Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
sequencelengths 1
91
| keywords
sequencelengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
The Huntington's disease ( HD ) CAG repeat , encoding a polymorphic glutamine tract in huntingtin , is inversely correlated with cellular energy level , with alleles over ∼37 repeats leading to the loss of striatal neurons . This early HD neuronal specificity can be modeled by respiratory chain inhibitor 3-nitropropionic acid ( 3-NP ) and , like 3-NP , mutant huntingtin has been proposed to directly influence the mitochondrion , via interaction or decreased PGC-1α expression . We have tested this hypothesis by comparing the gene expression changes due to mutant huntingtin accurately expressed in STHdhQ111/Q111 cells with the changes produced by 3-NP treatment of wild-type striatal cells . In general , the HD mutation did not mimic 3-NP , although both produced a state of energy collapse that was mildly alleviated by the PGC-1α-coregulated nuclear respiratory factor 1 ( Nrf-1 ) . Moreover , unlike 3-NP , the HD CAG repeat did not significantly alter mitochondrial pathways in STHdhQ111/Q111 cells , despite decreased Ppargc1a expression . Instead , the HD mutation enriched for processes linked to huntingtin normal function and Nf-κB signaling . Thus , rather than a direct impact on the mitochondrion , the polyglutamine tract may modulate some aspect of huntingtin's activity in extra-mitochondrial energy metabolism . Elucidation of this HD CAG-dependent pathway would spur efforts to achieve energy-based therapeutics in HD .
The CAG trinucleotide repeat in the Huntington's disease gene ( HD ) is highly polymorphic in humans , with alleles ranging from ∼6 to >100 units encoding a variable polyglutamine tract in huntingtin , a large ( >350 kDa ) HEAT domain protein [1] . The gene was discovered because alleles over ∼35–37 units , whether inherited in one copy or , in rare cases , two copies , are associated with the onset of Huntington's disease ( HD ) symptoms , including dance-like movements , cognitive decline , and psychiatric disturbance [1] . This intriguing disease-initiating mechanism , while relatively insensitive to dosage , is exquisitely progressive with allele size , such that the age at onset of HD symptoms is progressively decreased as CAG length is increased [1] . The early HD pathology comprises the selective loss of medium-size spiny neurons in the striatum that forms the HD pathological grading system [2] . The ability of mitochondrial respiratory chain poisons such as succinate dehydrogenase inhibitor 3-nitropropionic acid ( 3-NP ) to produce HD-like striatal-specific cell loss has implied a role for mitochondrial dysfunction in HD pathogenesis [3] . Indeed , numerous studies have demonstrated that deficits in measures of energy metabolism become manifest in presymptomatic and symptomatic HD brain and peripheral tissues [4–11] . Investigations of the early consequences of the HD CAG repeat in human lymphoblastoid cell lines have recently implicated the polyglutamine tract size in huntingtin in modulating cellular ATP/ADP ratio across the entire non-HD and HD range [12] . The longest alleles were associated with the lowest ATP/ADP ratios , while alleles in the non-HD range were associated with progressively higher energy levels [12] . Moreover , consistent with a role for the polyglutamine tract in influencing an intrinsic huntingtin function in energy metabolism , the targeted deletion of the short seven-glutamine tract from murine huntingtin yielded elevated cellular ATP , with early senescence , and improved motor performance in HdhΔQ/ΔQ mice [13] . The shared downstream consequences of the HD CAG tract and 3-NP have implied that the polyglutamine tract in huntingtin , like 3-NP , may directly affect the mitochondrion [14] . Huntingtin is detected throughout the cell , in the nucleus and in the cytoplasm , where it can associate with mitochondria [15] , implicating a direct “toxic” interaction [15–17] . However , recent findings , including studies in STHdhQ111/Q111 striatal cells , with a knock-in juvenile onset CAG repeat accurately expressed as endogenous huntingtin with 111 glutamines [18] , suggested that mutant huntingtin may influence mitochondrial biogenesis/function by decreasing Ppargc1a transcription [19 , 20] . This gene encodes peroxisome proliferative activated receptor gamma , coactivator 1 alpha ( PGC-1α ) , a key cofactor for Nrf-1 and other mitochondrial transcription regulators . To probe the earliest consequences of the HD CAG mechanism , we tested the mitochondrial hypothesis by using unbiased gene expression analysis to monitor the extent to which the accurate expression of the HD CAG repeat , in STHdhQ111/Q111 striatal cells , may reproduce the consequences of 3-NP challenge . The results confirmed a shared downstream energy collapse but did not indict direct 3-NP-like effects of the HD CAG repeat on the mitochondrion . Instead , the data have elevated the candidacy of extra-mitochondrial pathways in huntingtin modulation of energy metabolism .
We and others have demonstrated previously that STHdhQ111/Q111 cells , compared to wild-type STHdhQ7/Q7 cells ( expressing endogenous seven-glutamine huntingtin ) , exhibited mitochondrial energy phenotypes similar to the effects of 3-NP treatment , including decreased mitochondrial respiration [21] and ATP synthesis [12 , 21 , 22] . However , STHdhQ111/Q111 cells also have been shown to display phenotypes opposite to the effects reported for 3-NP challenge , such as increased , instead of decreased , levels of the reactive oxygen species ( ROS ) scavenger glutathione [23] , suggesting that 3-NP might not precisely mimic the effects of the HD mutation . To explore this notion , we examined additional energy phenotypes and observed , as reported for 3-NP challenge , that the HD mutation decreased mitochondrial membrane potential ( Figure 1A ) and elevated the lactate/pyruvate ratio ( Figure 1B ) indicative of altered energy homeostasis . Indeed , in human lymphoblastoid cells , this ratio was increased in severity with HD CAG repeat size ( Figure 1C ) , demonstrating that mitochondrial dysfunction is likely to be a consequence of the HD CAG size-dependent mechanism . However , while 3-NP , as reported [24] , was associated with increased ROS , as measured by hydrogen peroxide levels , the STHdhQ111/Q111 cells , compared to wild-type cells , exhibited decreased ROS ( Figure 1D ) . Moreover , while 3-NP treated wild-type cells , as expected , displayed decreased succinate dehydrogenase activity , the longer HD CAG repeat expressed in the STHdhQ111/Q111 cells did not significantly change the activity of this respiratory chain component ( Figure 1E ) . Thus , while superficially similar , the energy phenotypes displayed by mutant striatal cells did not in detail recapitulate the effects of 3-NP challenge , implying distinct underlying pathways . To delineate the effects of the HD CAG repeat and 3-NP treatment , without making a priori assumptions about the underlying biology , we performed global analysis to monitor the expression of the mitochondrial and the nuclear genomes . As sequences for genes located on the mitochondrial genome were not represented on the murine Affymetrix MG 430 2 . 0 arrays microarrays , which we used to analyze the nuclear genome ( below ) , the expression of nine of the 13 mitochondrial genes encoding respiratory components was assessed using specific RT-PCR assays . The mRNA levels for each of these genes was dramatically reduced in 3-NP treated cells , compared to untreated wild-type striatal cells , whereas mRNA levels did not differ significantly between STHdhQ111/Q111 and wild-type cells ( Figure S1 ) . Thus , consistent with unchanged succinate dehydrogenase activity , mutant huntingtin did not reproduce the effects of 3-NP on the mitochondrial genome , implying that the energy deficits in STHdhQ111/Q111 cells might instead stem from altered expression of the nuclear genes that regulate the mitochondrion . We then performed unbiased analysis of the nuclear genome expression datasets to determine the extent to which the consequences of the HD CAG repeat may mirror the effects of 3-NP challenge . The results of principle component ( Figure 2A ) and cluster analysis ( Figure 2B ) of all probes demonstrated that , whereas the replicate datasets were highly related , the wild-type , 3-NP treated , and mutant cell datasets were quite distinct . Thus , rather than giving rise to similar effects , which may differ in magnitude , the consequences of the HD CAG repeat and 3-NP appeared to be fundamentally different . Indeed , at a stringent false discovery rate ( FDR ) ( q < 0 . 005 ) , approximately the same proportion of all probes ( 3% ) was significantly altered either by the HD CAG repeat mutation ( comparing mutant versus wild-type cells ) ( Table S1 ) or by 3-NP challenge ( comparing 3-NP treated wild-type versus wild-type cells ) ( Table S2 ) , with the HD CAG yielding some changes of larger magnitude ( fold-change ) than 3-NP . However , little overlap was detected between the two probe lists ( summarized in Figure 2C and 2E ) , yielding low correlation coefficients between HD CAG and 3-NP changed probes ( Figure 2D ) . Indeed , for the most part , these represented different genes , which are plotted by chromosome in Figure S2 . Thus , while the impact , in terms of number and magnitude of changes , was similar , HD mutation did not reproduce the molecular effects of 3-NP . The mitochondrial hypothesis predicted that both the HD mutation and 3-NP treatment would alter mitochondrial energy genes , prompting an examination of the small fraction ( 0 . 18% ) of all probes significantly altered by both insults . These represented 83 known genes ( plotted by chromosome in Figure S3 and listed in Table S3 ) , which did not highlight the mitochondrion but instead spotlighted decreased cytosolic energy production ( Figure 3 ) . More sensitive tests of groups of genes in functional pathways , using the Gene Ontology ( GO ) biological process ( Figure 4A ) and sigPathway gene set analysis [25] ( Figure 4B ) , confirmed that the HD mutation and 3-NP were both associated with highly significant decreases in carbohydrate metabolism and , as reported previously [26] , lipid ( sterol/cholesterol ) biosynthesis ( Figure 4C ) . Thus , striatal cells , unlike glia or other cell types , may possess a limited capacity to adjust glycolytic flux in response to changes in mitochondrial ATP synthesis [27] , thereby providing a potential explanation for the ability of 3-NP to mimic the early loss of striatal neurons in HD striatum . Although the notion that the HD mutation and 3-NP might commonly alter the expression of nuclear encoded mitochondrial genes did not appear to be borne out , inspection of the data did reveal significant changes in a few mitochondria-related energy genes in mutant striatal cells , which , notably , were unchanged by 3-NP ( Figure 3 ) . Ppargc1a ( encoding PGC-1α ) was decreased , as reported previously [19] . Mitochondrial components in the transfer electrons from NADH to the respiratory chain ( Ndufa12 and Ndufa3 ) or in the transport of hydrogen ions ( Atp6v0e2 ) were decreased . By contrast , mRNAs encoding the iron sulfur-binding factor of complex III ( Uqcr ) and a nonenzymatic component of the ATP synthase complex ( Atp5j2 ) were elevated . Moreover , consistent with elevated glutathione [23] , the expression of genes that detoxify free radical derivatives ( Ldh2 , Gss , and Glo1 ) was increased , suggesting that altered redox state may contribute distinctly to low energy ( and lipid ) metabolism in mutant striatal cells . The decrease in Ppargc1a mRNA implied that mitochondrial regulatory transcription factors such as Nrf-1 , which are coregulated by PGC-1α , might be expected to improve energy metabolism in mutant cells . Indeed , over-expression of Nrf-1 in STHdhQ111/Q111 cells altered the mRNA levels of known Nrf-1 target genes ( Table S4 ) but , as reported for PGC-1α [19] , only mildly improved both cellular lactate/pyruvate ratio ( Figure 5A ) and cell survival following respiratory chain inhibition ( Figure 5B ) . Thus , coupled with the apparent dearth of changes in mitochondrial factors , the modest effects of boosting Nrf-1 strongly suggested that mutant huntingtin may not directly perturb the mitochondrion . Consequently , to more rigorously examine this possibility , we used gene set enrichment analysis ( GSEA ) to specifically determine whether mitochondrial pathways or processes may be perturbed by 3-NP or the HD mutation . GSEA examines , as a group , genes that form a functional pathway , thereby capturing small effect sizes that when considered in single gene analyses may not have reached our stringent statistical threshold . We tested two sets of ∼1 , 500 nuclear genes that were either GO annotated or predicted by the integrative genomics program MAESTRO [28] to encode mitochondrial products , as well as a third set comprising the group of 902 genes that were common to both lists ( Table S5; Figure 6A and 6B ) . The GSEA results , plotted as enrichment scores in Figure 6C , demonstrated that , compared to wild-type cells , each of the gene sets was highly enriched by 3-NP treatment . In striking contrast , no significant enrichment of any of the mitochondrial gene sets was detected in STHdhQ111/Q111 cells . Thus , these findings , which were consistent with the results of the mitochondrial genome analysis ( Figure S1 ) , clearly revealed that the HD mutation did not reproduce the direct effects of 3-NP on the mitochondrion . Therefore , contrary to predictions from decreased Ppargc1a mRNA or interactions of mutant huntingtin with the mitochondrion [14–17 , 19 , 20] , our results consistently implied that the HD CAG mechanism may primarily influence energy metabolism via extra-mitochondrial cellular pathways . Since our data supported the view that the processes by which the HD mutation and 3-NP may lead to energy starvation were largely distinct , we reasoned that the pathways that did mediate the effects of the huntingtin polyglutamine tract would be those that were not affected by 3-NP challenge . To support this approach , we tested whether the early presymptomatic consequences of the HD CAG repeat in medium-size spiny striatal neurons in vivo might be recapitulated by accurate expression of the expanded HD CAG repeat in cultured STHdhQ111/Q111 cells or by 3-NP challenge of wild-type cells . Microarray analysis of mRNA from medium-size spiny striatal neurons , obtained by laser capture microscopy ( LCM ) from post-mortem brain , has been reported [29] . The major class of LCM genes judged to be the most significantly altered by the HD CAG in that study , all with decreased expression , yielded a set of 38 mouse genes ( Table S14 ) that were tested by GSEA in our striatal cell HD CAG and 3-NP datasets . The results demonstrated that this human LCM gene set was significantly decreased in the STHdhQ111/Q111 cells ( enrichment score [ES] 0 . 51 , p-value 0 . 000 , FDR q-value 0 . 180 ) but was not altered in the 3-NP treated cells ( ES 0 . 38 , p-value 0 . 271 , FDR q-value 0 . 713 ) . Thus , the early molecular consequences of the HD CAG repeat in striatal neurons in human brain also become manifest as a consequence of accurate expression of the expanded repeat in the cultured STHdhQ111/Q111 cells but these were not reproduced by 3-NP challenge . This supported the approach of attempting to identify the processes by which huntingtin may modulate energy metabolism by examination of HD CAG repeat–specific changes . Therefore , stringent statistical filtering criteria were used to identify those gene sets that were enriched in mutant striatal cells but not in 3-NP treated wild-type cells ( Tables S6–S12 ) . Remarkably , a major class of pathways uniquely enriched in STHdhQ111/Q111 cells , summarized in Figure 4A and 4B , pointed to processes implicated in huntingtin's essential normal activities [30–33]: development/morphogenesis ( central nervous system , mesoderm , and embryo ) ; growth , cell motility , migration , and locomotion; cell adhesion; neuronal processes; and TGF-β and BMP signaling . A minor class , which denoted increased gap junction channel ( connexin ) , phosphate , and anion transport , may also prove to be related to huntingtin function . These results , therefore , strongly implicated huntingtin normal function in the capacity of the polyglutamine tract to influence mitochondrial function and cellular energy metabolism in STHdhQ111/Q111 cells . Notably , this finding is consistent with a genetic gain-of-function hypothesis for the mechanism that initiates the HD pathogenic process . The polyglutamine tract size may progressively influence energy homeostasis by increasing some intrinsic huntingtin activity . Alternatively , it may capitalize on an opportunity afforded by huntingtin function to modulate an unrelated cellular component . What might this normal huntingtin activity be ? One possibility , inferred from previous work with huntingtin-deficient cells [34 , 35] , as well as with STHdhQ111/Q111 cells [18] , may reflect a role for huntingtin in intracellular iron trafficking . However , huntingtin's normal activities impact a broad range of cellular processes , from vesicle trafficking to the proper regulation of gene transcription [33 , 36 , 37] . As a starting point , therefore , we reasoned that worthy candidates for huntingtin modulation of energy metabolism might be found among the transcriptional regulators that most frequently orchestrate the gene changes in STHdhQ111/Q111 cells , but not 3-NP treated cells . Unbiased transcription factor analysis coupled with GSEA yielded five HD CAG–specific candidates ( Figure 7 ) . Interestingly , NRSF/REST , associated with altered expression of genes in STHdhQ111/Q111 cells and in HD brain [37] , did not emerge as a candidate , and NRSF/REST target genes or genes with upstream RE1 binding sites were not enriched as a consequence of the HD CAG repeat ( unpublished data ) . The top candidates for HD CAG specific regulators were CAAT-Box ( e . g . , NF-Y , cEBP , and CTF ) and nuclear factor kappa B ( Nf-κB ) , with weaker support for Nrf-1 and Sp-1 . Most of these candidates , including Nrf-1 , have been shown to exert extra-mitochondrial effects on energy metabolism , affecting , e . g . , fatty oxidation , glycolysis , glutathione synthesis , and cell stress-sensing [38–40] . Moreover , by associating exclusively with mutant huntingtin ( or fragment ) , PGC-1α , a coregulator with Nrf-1 , Sp-1 , and its coregulator TAF4 , as well as core components of Pol II transcriptional machinery ( TFIID , TFIIF ) have been implicated in HD pathogenesis [19 , 41–43] . However , the mechanism by which the HD CAG modulates cellular energy levels was manifest even at non-HD-causing lengths , in the cadre of normal huntingtin function [12] . Nf-κB/Dorsal/RelB has been linked previously to huntingtin's normal function [44] . Nf-κB was a modifier of huntingtin carboxyl terminal fragment–induced phenotypes in Drosophila [44] . Furthermore , endogenous normal human huntingtin can associate with the p50 subunit of Nf-κB [44] , which binds to target gene promoters in a redox-dependent manner [45] . This raises a new hypothesis that merits investigation in a variety of suitable polymorphic HD CAG systems , namely , that huntingtin polyglutamine tract length may influence Nf-κB-mediated signaling , perhaps via redox sensing , either as a cause or a consequence of modulating energy metabolism . In summary , the results of this study demonstrated that the early ( presymptomatic ) consequences of the juvenile onset HD CAG allele , accurately expressed in STHdhQ111/Q111 striatal cells , did not in most details mimic mitochondrial respiratory chain inhibitor 3-NP , although both metabolic challenges produced a common response—the collapse of mitochondrial and cytosolic energy processes . This reveals the limited utility of 3-NP lesion models for uncovering the pathways by which the polyglutamine tract in huntingtin may influence energy metabolism or for prioritizing agents that may modify these processes . Indeed , the data uniformly refute the widely held view of a direct mutant huntingtin-specific effect on the mitochondrion . Instead , our data implicate an effect of the polyglutamine tract on some normal activity of the huntingtin protein in extra-mitochondrial energy metabolism , perhaps redox sensing . It is interesting to note that redox sensing cell signaling , via ROS-dependent pathways , is emerging as a regulator of glucose and lipid metabolism in health , aging , and disease [46] . If huntingtin proves to be involved , the molecules and gene products that modify oxidation-sensitive signaling in metabolic disorders may become candidates for modifying huntingtin-regulated mitochondrial metabolism in HD . By the same token , agents that modify redox sensing signaling in HD may also be of interest in cancer and a variety of complex metabolic disorders .
STHdhQ7/Q7 and STHdhQ111/Q111 cells , generated from striatal primordia of wild-type HdhQ7/Q7 and homozygous mutant HdhQ111/Q111 knock-in mouse embryos , respectively [18] , were cultured in DMEM ( 33 °C , 5% CO2 , 10% FBS , and 400 μg/ml G418 ) . Challenge of STHdhQ7/Q7 cells with 3-NP ( 1 mM ) was for 48 h . Previously genotyped human lymphoblastoid cell lines ( HD CAG 17/18 , 15/42 , 21/43 , 21/66 , 19/70 , 41/48 , and 45/50 ) were cultured in RPMI-1640 ( 37 °C , 5% CO2 , and 5% FBS ) . Mitochondrial membrane potential was measured by incubation of cells with JC-1 ( 2 . 5 μg/ml ) for 20 min at 33 °C , followed by flow cytometry ( FACScalibur , http://www . bdbiosciences . com/ ) , counting 104 cells in the FL1 ( monomer ) and FL2 ( aggregate ) channels and calculating FL2/FL1 ratio . Succinate dehydrogenase activity , measured using succinate as a substrate and 3- ( 4 , 5-dimethylthiazol-2-yl ) -5- ( 3-carboxymethoxyphenyl ) -2- ( 4-sulfophenyl ) -2H-tetrazolium as an electron acceptor , was normalized by protein concentration [47] . ROS levels were monitored by determinations of hydrogen peroxide concentration , using a chemiluminescent hydrogen peroxide detection kit ( Assay Designs , http://www . assaydesigns . com/ ) . Data were quantified compared to a standard curve , and normalized to cell number . Lactate and pyruvate concentrations in cleared lysates ( 3% perchloric acid , sonication ) were determined by HPLC analysis ( Aminex column , Bio-Rad Laboratories , http://www . bio-rad . com/; 0 . 6 ml/min 0 . 05 mM H2SO4; UV detection at 210 nm ) using appropriate standards . All measurements derive from triplicate experiments , using duplicate samples . Data are given as the mean , +/− one standard deviation . Reverse transcription PCR ( RT-PCR ) analysis for the genes listed in Table S5 was performed ( Bio-Rad iCycler ) with gene-specific primer sets listed in Table S13 . The ΔΔCT method was used to calculate gene expression levels , compared to β-actin control [48] . Total RNA ( 5 μg ) , isolated from triplicate cell cultures , was converted using SuperScript II reverse transcriptase ( Invitrogen . http://www . invitrogen . com ) to labeled cRNA , and 25 μg of labeled probe was hybridized to Affymetrix MG 430 2 . 0 arrays ( http://www . affymetrix . com ) . Expression data was generated and normalized using RMA [49] and significant probes were identified by FDR ( q < 0 . 005 ) [50] . A selection of genes judged to be significantly altered according to these criteria was tested by semi-quantitative RT-PCR analysis ( primer sets in Table S13 ) , with ∼80% ( 13 of 16 randomly chosen genes ) concordant with the microarray results . For pathway analysis , significant ( p < 0 . 05 ) GO terms in significantly altered probe sets were identified by GO analysis ( DAVID 2006 ) [51] and , using all probes , significantly enriched ( q < 0 . 01 ) pathways were identified by permutation-based GSEA ( sigPathway ) [25] . For the GSEA comparison of the mouse and human presymptomatic striatal cell changes , the LCM gene set comprised the 38 most robust LCM genes , all decreased by the HD CAG repeat , drawn from the table in Hodges et al . [29] , that could be unambiguously mapped to a corresponding mouse locus . Cluster [52] , dChip [53] , and Bioconductor [54] were used for clustering and data visualization . STHdhQ111/Q111 cells were transiently transfected with control vector ( pSG5 ) or Nrf-1 mammalian cell expression plasmid ( 0 . 2 μg of plasmid/well ) using lipofectamine as described by the manufacturer ( Invitrogen ) . At 48 h post-transfection , lactate/pyruvate ratios were determined , as described above , and viability following 3-NP challenge was measured by MTS cytotoxicity assay ( Promega , http://www . promega . com/ ) . Nrf-1 activity was judged by measuring the expression of a selection of Nrf-1 target genes ( Table S4 ) , using semi-quantitative RT-PCR analysis with gene-specific primer sets ( Table S13 ) . For significant genes , putative transcription factor binding sites were identified from the MAPPER database [55] , by searching 500 bp upstream from the transcription initiation site and then selecting the top ten transcription factors , by score , for each gene . A total of 4 , 400 hits ( 358 transcription factors ) and 6 , 684 hits ( 371 transcription factors ) were recorded for HD CAG and 3-NP treatment , respectively , and the frequency for each was then calculated relative to the total number of hits . As a statistical test , the 100 top target genes for each transcription factor that exhibited 2% or higher frequency in each comparison were identified and used in the transcription factor analysis , to build a custom set for gene set enrichment analysis [56] .
MIAME compliant microarray data were deposited at the Gene Expression Omnibus ( GEO , http://www . ncbi . nlm . nih . gov/geo/ ) with Accession number GSE3583 . | Huntington's disease ( HD ) is a tragic neurodegenerative disorder caused by a CAG repeat that specifies the size of a glutamine tract in the huntingtin protein , such that the longer the tract , the earlier the loss of striatal brain cells . A correlation of polyglutamine tract size has also implicated huntingtin in the proper functioning of mitochondria , the cell's energy factories . Here we have tested the prevailing hypothesis , that huntingtin may directly affect the mitochondrion , by using comprehensive gene expression analysis to judge whether the HD mutation may replicate the effects of 3-nitropropionic acid ( 3-NP ) , a compound known to inhibit mitochondria , with loss of striatal neurons . We found that , while mutant huntingtin and 3-NP both elicited energy starvation , the gene responses to the HD mutation , unlike the responses to 3-NP , did not highlight damage to mitochondria , but instead revealed effects on huntingtin-dependent processes . Thus , rather than direct inhibition , the polyglutamine tract size appears to modulate some normal activity of huntingtin that indirectly influences the management of the mitochondrion . Understanding the precise nature of this extra-mitochondrial process would critically guide efforts to achieve effective energy-based therapeutics in HD . | [
"Abstract",
"Introduction",
"Result/Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"mus",
"(mouse)",
"genetics",
"and",
"genomics"
] | 2007 | Unbiased Gene Expression Analysis Implicates the huntingtin Polyglutamine Tract in Extra-mitochondrial Energy Metabolism |
The intestinal microbiota enhances dietary energy harvest leading to increased fat storage in adipose tissues . This effect is caused in part by the microbial suppression of intestinal epithelial expression of a circulating inhibitor of lipoprotein lipase called Angiopoietin-like 4 ( Angptl4/Fiaf ) . To define the cis-regulatory mechanisms underlying intestine-specific and microbial control of Angptl4 transcription , we utilized the zebrafish system in which host regulatory DNA can be rapidly analyzed in a live , transparent , and gnotobiotic vertebrate . We found that zebrafish angptl4 is transcribed in multiple tissues including the liver , pancreatic islet , and intestinal epithelium , which is similar to its mammalian homologs . Zebrafish angptl4 is also specifically suppressed in the intestinal epithelium upon colonization with a microbiota . In vivo transgenic reporter assays identified discrete tissue-specific regulatory modules within angptl4 intron 3 sufficient to drive expression in the liver , pancreatic islet β-cells , or intestinal enterocytes . Comparative sequence analyses and heterologous functional assays of angptl4 intron 3 sequences from 12 teleost fish species revealed differential evolution of the islet and intestinal regulatory modules . High-resolution functional mapping and site-directed mutagenesis defined the minimal set of regulatory sequences required for intestinal activity . Strikingly , the microbiota suppressed the transcriptional activity of the intestine-specific regulatory module similar to the endogenous angptl4 gene . These results suggest that the microbiota might regulate host intestinal Angptl4 protein expression and peripheral fat storage by suppressing the activity of an intestine-specific transcriptional enhancer . This study provides a useful paradigm for understanding how microbial signals interact with tissue-specific regulatory networks to control the activity and evolution of host gene transcription .
The vertebrate intestine harbors a dense community of microorganisms ( gut microbiota ) that exerts a profound influence on distinct aspects of host physiology [1] , [2] . The gut microbiota has been identified as a potent environmental factor in a growing number of human diseases , including inflammatory bowel disease [3] , antibiotic-associated diarrheas [4] , cardiovascular disease [4] , and obesity [5] . As a consequence , there is considerable interest in understanding the mechanisms by which this resident microbial community influences health and disease in humans and other animals . The ability of the microbiota to modify host nutrient metabolism and energy balance is a prominent theme in host-microbe commensalism in the intestine . Recent mechanistic insights into this process have been provided by comparisons between mice reared in the absence of microbes ( germ-free or GF ) to those colonized with members of the normal microbiota , as well as high-throughput DNA sequencing analysis of the metabolic potential of gut microbial genomes . These approaches have shown that the gut microbiota contributes biochemical activities not encoded in the host genome that enhance digestion of dietary nutrients [6] , [7] . The resulting increase in digestive efficiency results in elevated plasma levels of triglyceride ( TG ) -rich lipoproteins [8] , [9] . TG within circulating lipoprotein particles is hydrolyzed through the rate-limiting activity of lipoprotein lipase ( LPL ) located at the luminal surface of capillaries . TG hydrolysis releases free fatty acids ( FFA ) for uptake by adjacent tissues for oxidation ( e . g . , in cardiac and skeletal muscle ) or fat storage ( e . g . , in adipose tissues ) [10] . The presence of a gut microbiota also results in a concomitant reduction in intestinal expression of Angiopoietin-like 4 ( Angptl4 , also called Fiaf , Pgar , and Hfarp ) [8] , [11] , encoding a circulating peptide hormone that acts as a direct inhibitor of LPL activity [12]–[15] . Studies in gnotobiotic mice have indicated that microbial suppression of Angptl4 expression is restricted to the intestinal epithelium and is not observed in other tissues that express Angptl4 , such as liver and adipose tissue . This restricted suppression leads to a significant increase in LPL activity and fat storage in adipose tissue of animals colonized with a microbiota , which is an effect abolished in mice lacking Angptl4 [8] . These results have established Angptl4 as a key host factor mediating the microbial regulation of host energy balance and have raised considerable interest in defining the mechanisms underlying the tissue-specific and microbial regulation of Angptl4 expression . The importance of understanding mechanisms regulating Angptl4 production is further underscored by reports suggesting that human ANGPTL4 functions as an important determinant of plasma TG levels [16] , [17] and by Angptl4's additional functions in angiogenesis [18] , tumor cell survival [19] and metastasis [20] , [21] , and wound healing [22] . Previous studies have revealed that mammalian Angptl4 expression is subject to complex cell type-specific regulation but the underlying mechanisms remain unclear . Angptl4 mRNA in humans and rodents is expressed in multiple tissues , including adipose tissue , liver , intestinal epithelium , pancreatic islets , and cardiac and skeletal muscle [8] , [19] , [23]–[26] . Preliminary insights into the trans- and cis-regulatory mechanisms controlling Angptl4 transcription have been provided by analyses in non-intestinal tissues . Members of the peroxisome proliferator-activated receptor ( PPAR ) family of nuclear receptors ( i . e . , PPARγ , PPARα , and PPARβ/δ ) have been identified as activators of Angptl4 expression in adipose tissue , liver [23] , [27] , skeletal [28] and cardiac muscle [29] , myofibroblasts [30] , and colon carcinoma cells [31] . A PPAR-responsive element ( element defined as a transcription factor binding site or TFBS ) located in the proximal portion of Angptl4 intron 3 has been shown to directly bind different PPAR family members in adipose tissue , liver [27] , and myofibroblasts [30] . Additional studies in non-intestinal cell types have identified functional TFBSs for SMAD3 and glucocorticoid receptor in the 5′ distal region and 3′ untranslated region ( UTR ) , respectively [30] , [32] . Angptl4 transcription is induced under hypoxic conditions in several non-intestinal cell types by hypoxia-inducible factor 1α ( HIF1α ) [33] , [34]; however , the TFBSs mediating this response have not been identified . These studies support a role for these trans- and cis-regulatory factors in controlling Angptl4 transcription in these cell types , yet the mechanisms underlying the transcription of Angptl4 in other tissues , such as the intestine and pancreatic islet , remain unknown . Moreover , the cis/trans-regulatory mechanisms underlying microbial suppression of Angptl4 transcription in the intestinal epithelium remain undefined . The zebrafish ( Danio rerio ) provides unique opportunities to study the transcriptional regulatory programs mediating tissue-specific and the microbial control of vertebrate gene expression . Robust transgenesis methods using the Tol2 transposon system [35] , large numbers of offspring , and optical transparency facilitate efficient spatiotemporal analysis of reporters driven by potential DNA regulatory regions in mosaic and stable transgenic animals [36] . The anatomy and physiology of the zebrafish digestive tract are highly similar to mammals , including an intestine , liver , gall bladder , and exocrine and endocrine pancreas [37]–[39] . The intestinal epithelium of the zebrafish displays proximal-distal functional specification and is composed of absorptive enterocytes as well as secretory goblet and enteroendocrine lineages [40] , [41] . The zebrafish intestine is colonized by a microbiota shortly after the animals hatch from their protective chorions at 3 days post-fertilization ( dpf ) [42] , [43] and reaches a stage sufficient to support nutrient digestion by 5 dpf [44] . To study the roles of commensal microbes on zebrafish development and physiology , we have developed methods for rearing GF zebrafish and colonizing them with members of the normal zebrafish microbiota [45] , [46] . By combining these methods with functional genomic approaches , we identified zebrafish transcripts that display altered expression levels in animals raised GF compared to those colonized with a normal microbiota , including microbial suppression of a zebrafish homolog of mammalian Angptl4 [47]–[49] . The expression pattern of this zebrafish Angptl4 homolog , and the mechanisms underlying the tissue-specific and microbial regulation of its expression , have not been previously described . These features position the zebrafish as a powerful model for assaying the regulatory potential of DNA involved in mediating cell-specific and microbe-responsive transcriptional events . Previous studies of DNA regulatory potential in the zebrafish system have focused primarily on developmental genes [50]–[54] , and it remains unclear if the lessons learned from these analyses [55] will apply to physiologic genes like Angptl4 that are regulated by endogenous as well as exogenous cues . Moreover , a paucity of available genome sequences for teleost species closely related to zebrafish has severely limited prior evolutionary analysis of cis-regulatory sequence and function . Here , we utilize the zebrafish to investigate the cis-regulatory mechanisms governing tissue-specific and microbial control of Angptl4 transcription . We focus our analysis on intestinal and islet expression , where the mechanisms regulating Angptl4 transcription have not been adequately examined . We first uncover distinct intronic cis-regulatory modules ( CRM , defined here as a discrete DNA region containing sufficient information to confer a regulatory function ) that mediate intestinal and islet expression . Using this information , we reveal that the intestine-specific CRM also responds to microbial stimuli to suppress angptl4 expression . These results provide novel insights into how vertebrates might control the tissue-specific transcription of Angptl4 and constitute an important advance towards understanding how commensal gut microbes regulate gene expression and energy balance in their vertebrate hosts .
A comparative sequence analysis revealed that the zebrafish genome encodes a single ortholog of mammalian Angptl4 that displays marked amino acid sequence conservation with other vertebrate homologs ( See Text S1 , Figure 1A , Figures S1 and S2 ) . We used RNA whole-mount in situ hybridization ( WISH ) to identify the tissues in which angptl4 is transcribed during zebrafish development . We found that zebrafish angptl4 mRNA is expressed ubiquitously in 1 dpf embryos ( Figure 1B ) but becomes enriched in specific tissues during post-embryonic stages . Transcripts for angptl4 are enriched in the intestinal epithelium by 4 dpf , shortly after the intestinal tract becomes completely patent ( Figure 1C ) , and become localized to the anterior intestine ( segment 1 ) by 6 dpf ( Figure 1D , 1E ) . Transcripts for angptl4 were also enriched in the pancreatic islet by 8 dpf ( Figure 1F ) and in the liver by 17 dpf ( Figure 1G , 1H ) . Notably , the intestinal epithelium [8] , [11] , liver [24] , [27] , and pancreatic islet [25] in mammals also express Angptl4 mRNA . These data establish that the zebrafish angptl4 ortholog is expressed in a tissue-specific pattern that is conserved across vertebrate lineages and suggest that the underlying transcriptional regulatory mechanisms may also be conserved . Previous studies have indicated that conservation in non-coding genomic DNA sequence across vertebrate lineages can be a reliable predictor of cis-regulatory DNA regions [56] , [57] . We therefore used this approach to discover regulatory regions controlling transcription of angptl4 in the liver , islet , and intestinal epithelium . Mammals and teleost fishes diverged approximately 438–476 million years ago [58] , whereas zebrafish ( clade Otocephala ) diverged from other teleost fishes with currently-available genome sequence [clade Euteleostei; i . e . , medaka ( Oryzias latipes ) , stickleback ( Gasterosteus aculeatus ) , fugu ( Takifugu rubripes ) , and tetraodon ( Tetraodon nigroviridis ) ] approximately 230–307 million years ago [59] . We generated multiple-species LAGAN alignments with Vista software using 10 kb of genomic sequence surrounding and including the angptl4 loci from four teleost fishes ( zebrafish , medaka , tetraodon , fugu ) and three mammals [human ( Homo sapiens ) , dog ( Canis familiaris ) , and mouse ( Mus musculus ) ] . Alignment of teleost and mammalian genomic sequences did not detect regions of primary sequence conservation within angptl4 non-coding regions ( >50% over 100 bp; data not shown ) , suggesting that these alignment methods are not sufficiently sensitive to detect existing non-coding conservation [56] or that the composition and/or location of non-coding regulatory regions are not stringently conserved between these lineages . We therefore separately aligned teleost angptl4 ( Figure 2A ) and mammalian Angptl4 loci ( Figure 2B ) and searched for non-coding sequence conservation in each lineage . These alignments revealed that human and zebrafish angptl4 loci both contain 7 conserved exons as well as a concentration of conserved non-coding sequences directly upstream of exon 1 and in intron 3 ( Figure 2 ) . Similarities in gene structure and locations of conserved non-coding regions , in addition to conservation in gene expression patterns , support the hypothesis that the regulatory mechanisms of angptl4 transcription may be evolutionarily conserved . We assayed the regulatory potential of DNA upstream and proximal to the zebrafish angptl4 transcription start site ( TSS ) for the ability to transcribe a reporter in the intestine , liver , and islet . We first employed 5′ rapid amplification of cDNA ends ( 5′RACE ) to determine the location of the TSS ( Figure S3B ) . We identified a single TSS located 89 base pairs ( bp ) upstream of the translation start site and a canonical TATA box at position −31 bp of the TSS ( Figure S3B ) . Based on this analysis and expressed sequence tag ( EST ) coverage of the zebrafish angptl4 locus ( data not shown ) , we found no evidence of alternative promoters farther upstream of the defined TSS . Using Tol2 transposon transgenesis , we assayed the regulatory potential of genomic DNA upstream of the zebrafish angptl4 TSS , including the 5′ untranslated region ( UTR ) ( Figure S3A ) , to drive expression of an enhanced green fluorescent protein ( GFP ) reporter in 0–7 dpf zebrafish larvae . We found that regulatory DNA within −1 kb , −3 . 5 kb , or −5 . 2 kb upstream of the TSS harbors the potential to drive GFP expression in mosaic animals in several tissues including liver at 6 dpf ( Figure S3C , S3E ) . Robust expression in the liver was confirmed in animals harboring stable germ-line incorporation of these transgenes ( Figure S3D , S3F ) . However , these angptl4 upstream regulatory sequences were not sufficient to drive detectable reporter expression in the intestine ( Figure S3G ) or islet ( data not shown ) . We therefore reasoned that information governing transcription in the intestine and islet must be located distal to the TSS and proximal promoter . Relatively high levels of DNA sequence conservation in both teleost and mammalian lineages ( Figure 2 ) prompted us to test the 3rd intron of zebrafish angptl4 for transcriptional regulatory potential . We cloned full-length zebrafish angptl4 intron 3 ( 2 , 136 bp; designated in3 ) into a Tol2 transposon reporter vector upstream of a minimal mouse Fos promoter ( Mmu . Fos ) driving transcription of a GFP or tdTomato reporter . Importantly , the minimal Fos promoter alone is relatively inactive in most tissues and is not sufficient to drive transcription of detectable levels of GFP in the intestine , islet , or liver [50] . Analysis of 6 dpf zebrafish larvae with mosaic expression of the Tg ( in3-Mmu . Fos:GFP ) transgene disclosed that full-length in3 is sufficient to confer reporter expression in multiple tissues including the liver , muscle , intestine ( Figure 3C ) , and islet ( not shown ) . This expression pattern was confirmed in fish with stable germ-line incorporation of the transgene ( Figure 3D ) . Guided by sequence conservation between zebrafish and medaka ( Figure 2A ) , we assayed serial truncations of in3 for spatial regulatory potential to determine whether reporter transcriptional activity in these distinct tissues is governed by the same CRM or through multiple discrete CRMs , ( Figure 3B ) . The first truncation separated liver expression ( 1 , 219 bp , designated in3 . 1 , Figure 3E , 3F ) from islet and intestinal expression ( 701 bp; designated in3 . 2 , Figure 3G , 3H ) . Further truncation of in3 . 2 uncoupled islet ( 387 bp; designated in3 . 3; Figure 3I , 3J ) and intestinal ( 316 bp; designated in3 . 4; Figure 3K , 3L ) expression . This analysis therefore revealed non-overlapping modules sufficient to confer mosaic and stable reporter expression in the liver , islet , and intestinal epithelium that is consistent with endogenous angptl4 mRNA expression ( Figure 1 ) . We next sought to identify the specific cell types in the intestinal epithelium and pancreatic islet in which modules in3 . 3 and in3 . 4 respectively enhance transcription . To define the cell type within the islet in which module in3 . 3 is active , we utilized a zebrafish transgenic line that drives expression of cyan fluorescent reporter ( CFP ) specifically in insulin-producing β-cells within the islet ( Tg ( ins:CFP-NTR ) s892 ) [60] . In vivo imaging of 6 dpf progeny from intercrosses of Tg ( ins:CFP-NTR ) s892 and Tg ( in3 . 2-Mmu . Fos:tdTomato ) adults revealed strong co-localization of CFP and tdTomato ( Figure 3O ) , indicating that the in3 . 3 module specifically enhances transcription in pancreatic β-cells . Immunofluorescence assays of sectioned 6 dpf zebrafish stably expressing the Tg ( in3 . 4-Mmu . Fos:GFP ) transgene revealed that GFP driven by the in3 . 4 module co-localizes with 4E8-positive absorptive enterocytes ( Figure 3M ) but not with 2F11-positive secretory cells in the intestinal epithelium ( Figure 3N ) . These data suggest that in3 . 4 functions as an enterocyte-specific transcriptional regulatory module . We next tested whether the intestine-specific reporter expression generated by module in3 . 4 is independent of the Fos minimal promoter , orientation , and proximal position to the TSS . This module is located downstream of the TSS in intron 3 of the endogenous angptl4 gene; however , our synthetic reporter construct positions it upstream of the TSS and the Fos minimal promoter . We therefore cloned in3 . 4 into a position downstream of GFP in either the forward or inverse orientation under control of either a Fos minimal promoter or the −1 kb angptl4 promoter . Each of these constructs was sufficient to promote robust reporter expression in the anterior intestine of 6 dpf mosaic and stable zebrafish ( Figure S4A and data not shown ) , similar to our observations with in3 . 4 located in the proximal position ( Figure 3K , 3L ) . These results establish that in3 . 4 is a bona fide transcriptional enhancer module active in enterocytes in the anterior intestine . We next used DNase I hypersensitivity to determine if the in3 . 4 module functions as an intestinal regulatory module in vivo at the endogenous angptl4 locus . To obtain a sufficient number of intestinal epithelial cells for this assay , we analyzed intestines from adult zebrafish . Stable transgenic zebrafish harboring the in3 . 2 or in3 . 4 reporter maintain reporter activity in the intestine into adulthood ( Figure S4B and data not shown ) indicating this module and associated trans-regulators are active in the adult zebrafish intestine . We find that the endogenous angptl4 promoter and in3 . 4 module , but not the adjacent in3 . 3 module , are hypersensitive to DNase I cleavage in intestinal epithelial cells isolated from adult zebrafish ( Figure 3P ) . The endogenous in3 . 4 module is therefore an active regulatory module in the intestinal epithelium , under regulatory control distinct from the adjacent in3 . 3 module , consistent with our transgenic reporter analysis of this same region . Together , these data reveal extensive transcriptional regulatory potential within intron 3 of zebrafish angptl4 and suggest that distinct intronic modules may mediate spatially restricted transcription of angptl4 in the intestinal epithelium , pancreatic β-cells , and liver . We used comparative genome sequence analysis from 12 teleost fishes and heterologous in vivo reporter assays to explore the evolution of the islet and intestinal regulatory modules . We originally postulated that evolutionary conservation of non-coding sequences could be used to predict the location of cis-regulatory regions controlling spatial and environmental regulation of angptl4 transcription ( Figure 2 ) . However , the significant amount of time ( approximately 230–307 million years ago ) [59] since the divergence between zebrafish ( clade Otocephala; order Cypriniformes ) and the other teleost fish with available genome sequence ( all from clade Clupeocephala , such as medaka ) did not permit high-resolution analysis of recent evolution of zebrafish angptl4 regulatory sequences ( Figure 2A ) . We therefore sequenced the intronic region orthologous to in3 . 2 from 10 additional Ostariophysi species , including 1 from order Siluriformes ( channel catfish , Ictalurus punctatus ) and 9 other members of order Cypriniformes ( Figure 4A ) . Because genome sequences are not currently available for these species , we took advantage of the intronic location of these regulatory modules by utilizing PCR primers targeting highly conserved sequences in flanking exons 3/4 or intron 3 to clone and sequence these putative regulatory regions . As expected , pairwise alignments of new sequences orthologous to zebrafish in3 . 2 revealed an inverse relationship between the phylogenetic distance between the two species and module sequence conservation , with the intestinal module diverging more rapidly than the islet module ( Figure 4B , Figures S5 and S6 ) . To test the functional consequences of the observed module divergence in these teleost species , we analyzed each module using our zebrafish mosaic transgenic assay for regulatory potential in the intestine and islet . Despite accounts of functional conservation in the absence of primary sequence conservation [50] , [61] , the non-coding sequence within medaka angptl4 intron 3 orthologous to zebrafish in3 . 2 ( Ol in3 . 2 ) failed to drive reporter expression in either the intestine or islet ( Figure 4C ) . Notably , all tested Ostariophysi modules elicited robust reporter expression in the islet ( Figure 4C ) . However , only in3 . 2 from Cypriniformes species within the Danio monophyletic group ( Danio nigrofasciatus , D . choprae , D . feegradei ) [62] , [63] were sufficient to confer reporter expression in the intestine ( Figure 4C ) despite marked regions of sequence conservation within the intestinal module in other Cypriniformes species ( D . aequipinnatus , C . auratus , C . carpio , P . conchonius ) . These results reveal differential evolutionary dynamics of the angptl4 intestinal and islet modules and support the hypothesis that high sequence conservation is required for tissue-specific transcription . Guided by our conservation analyses , we next sought to map the boundaries of critical regulatory regions in the zebrafish in3 . 3 islet and in3 . 4 intestinal CRMs by creating and testing truncations of these modules . Each truncation construct was injected into embryos and analyzed at 6–7 dpf for mosaic expression in the islet or intestine . These analyses defined a 164 bp region sufficient to confer islet expression ( in3 . 17; Figure 5A , 5B ) including a 129 bp region present in all islet-sufficient truncations ( Figure 5A ) . This 129 bp region overlaps with conserved regions identified in our comparative evolutionary analysis ( Figure 7A ) . In silico prediction of transcription factor binding sites in this critical region identified putative binding sites for multiple transcription factors known to be active in pancreatic islets such as Myc [64] , [65] and Arnt/HIF1b [66] , [67] , as well ubiquitously expressed transcription factors with important regulatory roles in β-cells such as USF [68] and CREB/ATF [69] ( Figure 7A ) . A distinct 116 bp region ( in3 . 12 ) was found to be sufficient to confer intestinal expression ( Figure 5A , 5C ) . Notably , the intensity driven by in3 . 12 in the intestine was lower than other larger truncations of this module that confer strong intestine-specific expression , such as in3 . 9 and in3 . 11 ( Figure 5C , 5D ) . The in3 . 12 truncation therefore represents a minimal intestinal regulatory module that requires additional flanking sequence information to facilitate maximal activity . Intriguingly , the in3 . 11 truncation , which displays strong intestinal activity , overlaps with two regions of high conservation identified in our comparative evolutionary analysis ( Figure 7B ) , suggesting that specific sequences within these conserved regions may be responsible for mediating intestine-specific enhancer activity . Together , these results define the approximate boundaries of functional regulatory DNA within angptl4 intron 3 required for intestinal and islet transcription . To complement our comparative genomic and truncation strategies , we used site-directed mutagenesis ( SDM ) to generate a higher-resolution understanding of the functional DNA motifs required for enterocyte-specific transcription of angptl4 . Ten base-pair substitutions were tiled across the region corresponding to in3 . 11 within the context of the entire in3 . 4 module , and assayed for competency to drive intestinal transcription ( Figure 6A ) . This analysis revealed two regions of 40 bp and 20 bp that disrupt intestinal reporter expression when mutated ( Figure 6B , 6C ) . DNA adjacent to these regions was not required for intestinal expression , validating the efficacy of the experimental approach . These data support our truncation mapping experiments ( Figure 5 ) by localizing a required region within the in3 . 12 truncation , as well as a second region within the larger , more active in3 . 11 truncation . We observed strong overlap between conserved sequences in intestine-positive in3 . 4 modules identified in our comparative genomic analysis and regions identified by SDM as required for intestinal expression ( Figure 7B ) . Specifically , SDM revealed that regions deleted in Daeq and Dn lineages do not harbor functional motifs required for intestinal expression . Most notably , mutation block 4–7 overlap with the single nucleotide polymorphisms between Devario and Danio species ( Figure S6 ) . This region harbors predicted binding sites for transcription factors involved in intestinal epithelial cell biology ( Figure 7B; see Discussion ) that represent attractive candidates for controlling enterocyte-specific angptl4 transcription . The presence of commensal gut microbiota in mice results in decreased levels of Angptl4 transcript specifically in the intestinal epithelium , which is thought to lead to increased peripheral fat storage [8] . However , it remained unknown whether this microbe-induced change in transcript levels was due to alterations in transcriptional activity or transcript stability . We speculated that the intestine-specific cis-regulatory module within intron 3 could impart this environmental response in the zebrafish . Our previous comparisons of 6 dpf GF zebrafish to age-matched ex-GF zebrafish colonized since 3 dpf with a normal microbiota ( conventionalized or CONVD ) indicated that the presence of a microbiota results in reduced angptl4 transcript levels [47]–[49] . To define the cellular origins of this response in zebrafish hosts , we used semi-quantitative WISH assays to reveal marked reduction of angptl4 mRNA in the intestinal epithelium in 6 dpf CONVD zebrafish compared to age-matched GF controls ( Figure 8A ) . These results indicate that intestinal epithelial suppression of angptl4 expression is a conserved response to the microbiota in zebrafish and mammalian hosts . We next tested the ability of the zebrafish intestinal CRM in3 . 4 to mediate the observed microbial suppression of the endogenous angptl4 gene . We reared stable Tg ( in3 . 4-Mmu . Fos:GFP ) zebrafish to 6 dpf under GF or CONVD conditions and assayed transcript levels for both GFP reporter and endogenous angptl4 using qRT-PCR . Consistent with our WISH results , endogenous angptl4 transcript levels were significantly and reproducibly reduced in CONVD compared to GF animals ( Figure 8B ) . Strikingly , transcript levels of the GFP reporter gene were similarly reduced in CONVD compared to GF animals ( Figure 8B ) . These observations were confirmed using an independent stable transgenic line , Tg ( in3 . 2-Mmu . Fos:tdT ) , harboring the in3 . 2 reporter which includes the in3 . 4 module ( Figure S7 ) . These data identify the angptl4 in3 . 4 module as a nodal cis-regulatory module that integrates transcriptional regulatory input from intestinal epithelial-specific and microbial factors .
Transcriptional regulation is a key determinant of gene function in the context of animal development and physiology . Recent biochemical and genetic studies in mouse and humans have identified Angptl4 as a critical hormonal regulator of TG-rich lipoprotein metabolism , angiogenesis , and tumor cell survival and metastasis . An improved understanding of the mechanisms controlling Angptl4 activity levels could therefore lead to new approaches for controlling multiple pathophysiologic processes . Although we have a working understanding of Angptl4's post-translational functions , our current knowledge of the mechanisms underlying Angptl4 transcription in different tissues is relatively limited . Here , we exploited the advantages of the zebrafish model system to examine the regulatory potential of DNA at the angptl4 locus in all cell types simultaneously and within an intact and living vertebrate organism that can be raised under gnotobiotic conditions . We found that the zebrafish angptl4 ortholog is expressed in many of the same tissues and cell types as mammalian Angptl4 ( i . e . , liver , pancreatic β-cells , and intestinal enterocytes ) . This finding suggests that the tissue-specific pattern of Angptl4 expression may have been conserved in the last common ancestor of mammalian and teleost lineages and might have important functional consequences on vertebrate physiology . Our results reveal that transcription of angptl4 in distinct tissues might be governed by independent cis-regulatory mechanisms . This modular design could have important implications for Angptl4 evolution and function . First , tissue-specific CRMs could have allowed independent evolution of CRM sequence structure . Consistent with this notion , we observed evidence of differential evolution of the islet and intestinal modules within teleost fish lineages ( Figure 4 ) . Differential selective pressures influencing CRM sequence evolution likely arise from the vastly different cellular contexts and exogenous stimuli of each cell type . Pancreatic β-cells are surrounded by other endocrine and exocrine pancreatic cells as well as vascular endothelial cells , whereas intestinal epithelial cells are exposed to complex and variable contents of the intestinal lumen and to the cells of the underlying lamina propria . Combining the observations that ( i ) functional conservation of the intestinal module is restricted to Danio species , ( ii ) transcriptional activity of the intestinal module is sensitive to the microbial status of the intestinal lumen , and ( iii ) this microbial regulation of angptl4 transcript levels is conserved in mammals , suggests an intriguing possibility that genes expressed in intestinal epithelia exposed to the dynamic and potentially hazardous luminal environment undergo relatively rapid regulatory evolution . Previous studies have suggested that the expression and function of defensin genes within the epithelia of the intestine and other exposed tissues has driven rapid evolution of their coding sequences [70] , and our results raise the possibility that similar selective pressures may also affect evolutionary rate of regulatory sequences for angptl4 and potentially other genes . Second , discrete cis-regulatory modules could have led to the independent evolution of Angptl4 synthesis in each respective cell type . This evolution would allow each expressing cell type to independently communicate its physiologic status and environmental exposures systemically by secreting Angptl4 into circulation , and locally by secreting Angptl4 into the extracellular space . The modular organization of these independent tissue-specific CRMs suggests that therapeutic strategies could be developed to control Angptl4 synthesis in specific target tissues without unintended effects on Angptl4 synthesis in other tissues . Previous studies of CRM evolution in vertebrates and invertebrates have focused primarily on enhancers regulating expression of genes involved in development [50] , [61] , [71] . These studies revealed that maintenance of regulatory function can be sustained over long evolutionary distances despite marked sequence dissimilarity and turnover of regulatory information . Our work provides a novel example of utilizing genomic DNA sequences from both close and distant relatives to define the evolutionary dynamics of multiple CRMs and marks the first time to our knowledge that such an extensive exploration ( i . e . , 12 related fish species ) was carried out in a vertebrate . We find that transcriptional output generated by both the intestinal and islet modules is maintained through a striking conservation in DNA sequence throughout the entire functional module , with little or no turnover of predicted binding sites . This finding suggests that these modules can comply with the “enhanceosome model” of regulatory information organization , as opposed to the “billboard model , ” which accommodates variation in binding site order , orientation , and spacing [72] , [73] . However , we detected little non-coding sequence conservation between zebrafish angptl4 and mouse Angptl4 intron 3 , and we did not detect islet or intestinal reporter expression in a heterologous assay in which we tested full and truncated versions of mouse introns 3 and 4 in the zebrafish ( data not shown ) . This result suggests either that regulatory information governing islet and intestinal expression of murine Angptl4 is not located within intron 3 or that compensatory cis/trans mutations render murine intron 3 sequences non-functional in the zebrafish . We suspect that rules governing CRM function and evolution are dependent on the distinct nature of the organism , the specific module , and the signals that the module integrates . It therefore remains an intriguing question as to what extent lessons learned from developmental gene regulation are applicable to the evolution of CRMs controlling expression of genes like Angptl4 that function in homeostatic physiology or in response to environmental factors like the microbiota [72] . Analyses of Drosophila genomes have elegantly shown that CRM “discovery power scales with the divergence time and number of species compared” [74] , and our results suggest that the same will be true in vertebrate lineages . Moreover , our data underscore the need for more reference genome sequences from phylogenetically diverse fish species , in combination with experimentally tractable fish models such as the zebrafish , to facilitate new insights into vertebrate CRM function and evolution . The intestinal microbiota has been identified as an important environmental factor that contributes to host energy storage and obesity , and our results provide critical new insights into how this might be achieved . Previous studies in gnotobiotic mice have shown that the intestinal microbiota regulates fat storage in part by suppressing Angptl4 transcript levels in the epithelium of the small intestine but not in liver or WAT [8] , [11] . However , it remained unclear whether these microbe-induced reductions of Angptl4 transcript levels were due to alterations in Angptl4 transcription or mRNA turnover . Furthermore , the molecular basis of the intestinal specificity of this response remained unknown . Our results reveal that zebrafish angptl4 transcript levels are also reduced in the intestinal epithelium in the presence of a microbiota , suggesting that the microbial regulation of angptl4 transcript levels might be an evolutionarily ancient feature of host-microbe commensalism in the vertebrate intestine . Our observation that transcript levels from the in3 . 4 reporter and the endogenous angptl4 gene respond similarly to microbial colonization strongly suggests that the microbiota regulates angptl4 expression , at least in part , by reducing the transcriptional activity of this enterocyte-specific enhancer module . These results indicate that enterocyte-specific and microbial control of angptl4 transcription is conferred through a shared intronic enhancer . Future investigation will be required to determine whether microbial regulation of in3 . 4 activity is achieved by ( i ) reducing the accessibility of this chromatin region to activating trans-factors , ( ii ) subverting the expression or activity of activating trans-factors , and/or ( iii ) inducing expression or activity of repressive trans-factors that function through this module . To distinguish between these models , it will be useful to identify the microbial activity and host transcription factors that regulate angptl4 transcription in the intestinal epithelium . We previously reported that colonization of GF zebrafish with a microbiota harvested from conventionally raised zebrafish or mice resulted in similar suppression of angptl4 transcript levels in the digestive tract [48] . This finding suggests that the microbial factor ( s ) regulating zebrafish angptl4 transcription is expressed by the ‘native’ zebrafish microbiota and in the ‘non-native’ and compositionally distinct mouse gut microbiota . Previous studies have identified individual microbial species sufficient to regulate angptl4 expression in gnotobiotic zebrafish [47] , [48] and mouse hosts [9] , [75] as well as in cultured colon cancer cells [31] , [76] , suggesting that reductionist approaches in these microbial species could be used to define the specific factors they utilize to control expression of angptl4 homologs and other host genes . In this study , we define two minimal regions within the in3 . 4 CRM that harbor regulatory activity in the intestine and are also conserved within the Danio lineage ( Figure 7B ) . Predicted transcription factor binding sites within these regions intimates potential roles for these factors in regulation of angptl4 tissue-specific transcription and/or microbial suppression . Because sequence-specific transcription factors typically recognize 6–12 bp motifs [77] , it is reasonable to assume that multiple factors cooperate to combinatorially regulate intestinal expression through this CRM . The Hnf4 family of fatty acid-regulated nuclear receptors has evolutionarily conserved roles in lipid metabolism [78] , [79] , and Hnf4a is expressed in the intestinal epithelium of zebrafish [80] and mouse [81] . Similarly , GATA factors 4 , 5 , and 6 are all expressed in the zebrafish [82] , [83] and mouse [84] , [85] intestinal epithelium and have proposed roles in regulating epithelial cell differentiation . Notably , C . elegans GATA family member elt-2 has been implicated in mediating intestinal epithelial cell immune responses [86] , suggesting that GATA factors could mediate tissue-specific as well as microbial regulatory inputs at angptl4 . PPAR family members have been identified as key regulators of mammalian Angptl4 expression in adipocytes and hepatocytes through PPAR responsive elements located in the 5′ portion of human ANGPTL4 intron 3 [27] , [30] , and zebrafish PPARγ [87] and PPARδ [88] homologs are expressed in the larval intestine . The zebrafish angptl4 locus contains multiple predicted PPRE sites , including several in both the 5′ and 3′ portion of intron 3 [89] . Most notably , a predicted PPRE was detected within the substitution blocks 16/17 in the intestinal enhancer in3 . 4 ( Figure 7B ) . However , the PPREs within zebrafish angptl4 intron 3 that display the highest sequence homology to the defined human ANGPTL4 intron 3 PPRE mapped outside of minimal regions for either intestinal or islet expression within the 5′ liver module ( data not shown ) . The location of these PPREs in the 5′ region of zebrafish angptl4 intron 3 , combined with the fact that the PPREs discovered in human ANGPTL4 are also located in the 5′ portion of intron 3 , suggests that the predicted PPREs within the 3′ islet and intestine CRMs of zebrafish angptl4 could represent novel elements for which functional equivalents have not been identified in mammals . Although these predicted factors represent candidates for controlling intestine-specific regulation of angptl4 , databases of predicted TFBSs are incomplete and commonly produce both false-positive and false-negative predictions . Moreover , critical regions identified by SDM might reflect sequences that alter nucleosome positioning or histone modification patterns rather than binding sites for sequence-specific transcription factors . Therefore , we anticipate that unbiased methods for transcription factor discovery will provide the most rigorous approach to an improved understanding of this cis/trans system . The structure-function analysis of the zebrafish in3 . 4 intestinal enhancer module reported here was designed to identify sequences critical for intestinal activity . It will therefore be interesting to determine whether exogenous microbial inputs are interpreted through the same or distinct motifs within this CRM and how the endogenous trans-acting factors mediating microbial and intestinal regulatory inputs interact to determine transcriptional output .
All experiments using zebrafish were performed in wild-type TL or Tg ( ins:CFP-NTR ) s892 [60] strains according to established protocols approved by the Animal Studies Committee at the University of North Carolina at Chapel Hill . New stable transgenic lines genereated in this study are listed in Table S3 . Conventionally raised zebrafish were reared and maintained as described [87] . Production , colonization , maintenance , and sterility testing of germ-free zebrafish were performed as described [45] , [49] . Protein sequences from top BlastP hits to human ( Homo sapiens , Hs ) ANGPTL4 and zebrafish Angptl4 ( Danio rerio , Dr ) were acquired through NCBI or Ensembl and aligned using MUSCLE with default settings [90] . Amino acids highlighted in black represent identical residues in at least 50% of species , whereas amino acids highlighted in grey represent biochemically similar residues ( Boxshade ) . The cleavage recognition sequence and LPL inhibition domain were annotated using information from previous publications [15] , [91] . The boundaries of the fibrinogen domain were annotated using in silico predictions [92] , [93] . Gaps in the alignment resulting from poorly annotated sequences were manually curated using primary DNA sequence and in silico translated using ExPASy [94] . The workflow for inferring phylogenetic relationships was performed at http://mobyle . pasteur . fr/cgi-bin/portal . py . A distance matrix was computed using Phylip 3 . 67 ( Protdist , JTT matrix , default settings ) , and trees were built using the neighbor-joining method . Bootstrap analysis was performed from 1000 replicates . PHYLIP software and the maximum likelihood probability model [95] using default settings were used to confirm the phylogeny inferred using distance methods . See Table S1 for a complete list of protein sequences used in this study . Genomic DNA sequences encompassing 10 kb upstream , including , and 10 kb downstream of the Angptl4 locus from Homo sapiens ( GRCh37:19:8419011:8449257:1 ) , Mus musculus ( NCBIM37:17:33900702:33928520:−1 ) , Canis familiaris ( BROADD2:20:55933601:55958821:1 ) , Danio rerio ( Zv9:2:23312551:23337293 ) , Oryzias latipes ( MEDAKA1:17:6095931:6120384:1 ) , Takifugu rubripes ( FUGU4:scaffold_212:367815:391593:1 ) , and Tetraodon nigroviridis ( TETRAODON8:15:3989265:4012887:1 ) were acquired through Ensembl . 10 kb was chosen as a cutoff because of proximity to neighboring gene loci . Genomic DNA sequence encompassing the angptl4 locus from Danio albolineatus was generously provided by David Parichy ( Department of Biology , University of Washington ) . For species without available genomic sequence , angptl4 intron 3 regions were PCR amplified from the relevant genomic DNA using a high-fidelity Taq polymerase ( Platinum , Invitrogen ) and the primers listed in Table S2 . PCR products were cloned into TOPO vector pCR2 . 1 ( Invitrogen ) prior to sequencing with M13F primers . An EST corresponding to an angptl4 homolog in Ictalurus punctatus ( CK419825 ) was used to design primers targeting exon 3 and exon 4 for PCR amplification of the full-length intron 3 . For Cypriniformes species , ESTs EG548328 ( Rutilus rutilus ) , DT085020 ( Pimephales promelas ) , GH715226 ( Pimephales promelas ) , and AM929131 ( Carassius auratus ) were aligned and used to design primers targeting highly conserved regions in angptl4 exon 3 and exon 4 , which we predicted would function for multiple Cypriniformes species . These primers were used to amplify , clone , and sequence the full-length intron 3 from Cyprinus carpio and Chromobotia macracanthus . Alignment of Cc , Cm , and Dr revealed 100% conservation at the extreme 5′ end of the in3 . 2 module . We used a forward primer targeting in3 . 2 in combination with a reverse primer targeting exon 4 for cloning of the remaining Cyprinidae species . The bacterial artificial chromosome golwb118_K01 containing the angptl4 locus from Oryzias latipes was provided by Hiroyo Kaneko ( Laboratory of Bioresource , National Institute for Basic Biology , Okazaki , Japan ) . Carassius auratus , Puntius conchonius , Cyprinus carpio , Devario aequipinnatus , and Chromobotia macracanthus genomic DNA was extracted from the fins of two individuals acquired from commercial suppliers . Genomic DNA from Ictalurus punctatus and Danio species ( Danio nigrofasciatus , Danio choprae , Danio feegradei ) from one individual were generously provided by Zhanjiang Liu ( Department of Fisheries and Allied Aquacultures , Auburn University ) and David Parichy ( Department of Biology , University of Washington ) , respectively . Novel angptl4 intron 3 sequences generated in this study were deposited in GenBank with accession numbers JN606312–JN606321 . Intronic sequences were aligned in mVISTA using LAGAN [96] and visualized using VISTA conservation plots ( 100 bp windows Figure 2 and 25 bp windows Figure 4 ) [97] . DNA sequences were queried for predicted transcription factor binding sites deposited in TRANSFAC [98] and JASPAR [99] databases using MATCH [100] and TESS [101] programs using default settings . We used a discriminative motif MEME [102] search to discover motifs common to islet-positive or intestine-positive intronic regions , using sequences orthologous to in3 . 4 or sequences orthologous to in3 . 3 , respectively , as negative selectors . To determine if MEME motifs were unique to islet- or intestine-positive regions , we used MAST [103] to query islet-negative ( Ol in . 3 ) or intestine-negative ( Daeq , Ca , Cc , Pc , Cm , Ip , Ol in3 . 4 ) sequences for islet-positive or intestine-positive MEME motifs , respectively . TOMTOM [104] was used to query MEME hits against TRANSFAC and JASPAR databases . In situ hybridization was performed in whole zebrafish as described [87] , except that heads and tails were removed from euthanized 17 dpf animals prior to fixation . Sense and anti-sense riboprobes targeting zebrafish angptl4 were generated by digesting plasmid fj89c07 in pBK-CMV ( NCBI Accession XM_686956 ) with NotI ( sense ) or BamHI ( anti-sense ) , and transcribed in vitro using T3 ( sense; Epicentre ) or T7 RNA polymerase ( anti-sense; Epicentre ) . Sense riboprobes were used in each experiment as a negative control . Total RNA was extracted from groups of 6 dpf whole zebrafish larvae from 6 dpf zebrafish ( 10 larvae per group , 2 biological replicate groups per condition per experiment , 2 experimental replicates total ) using TRIzol Reagent ( Invitrogen ) or the Qiagen RNeasy ( Qiagen ) kit using manufacturer's protocol . qRT-PCR was performed as described [49] . Primers used in qRT-PCR assays are listed in Table S2 . ESTs at the zebrafish angptl4 locus were analyzed using UCSC and Ensembl genome browsers . Total RNA was extracted from adult zebrafish intestines and subjected to 5′RACE using the FirstChoice RLM-RACE kit ( Ambion ) , according to the manufacturer's specifications ( see Table S2 for primers ) . Three clones were sequenced and mapped to the zebrafish angptl4 locus . All PCR reactions used for cloning were performed with high-fidelity DNA polymerase ( PfuTurbo , Stratagene; Phusion , Invitrogen; Platinum Taq , Invitrogen ) and TOP10 chemically competent E . coli ( Invitrogen ) . The bacterial artificial chromosome C177A22 containing the zebrafish angptl4 locus was used as the template for all zebrafish angptl4 promoter and intronic PCR amplification and cloning . Mouse BAC ( RP24-294G12 , CHORI ) , Medaka BAC ( golwb118_K01 ) , and sequenced pCR2 . 1 clones ( Ip , Pc , Cc , Ca , Daeq , Df , Dc , Dn ) containing intronic regions orthologous to zebrafish in3 . 2 from each species were used as source material for cloning in heterologous reporter assays . The plasmid pT2cfosGW [50] was used as the vector backbone for all Tol2 transgenic reporter assays . The Fos minimal promoter and angptl4 5′ upstream regions were PCR amplified and directionally cloned into pT2cfosGW using XhoI and BamHI restriction sites . This step removed both the original Fos promoter and the upstream Gateway site . Of note , we observed significant levels of reporter expression in muscle tissue upon removal of the Gateway cloning site ( Figure 5C , D and data not shown ) . Intronic DNA was cloned upstream of the Fos minimal promoter in pT2cfosGW using Gateway reagents as described [36] . The intronic module in3 . 4 was non-directionally cloned into Tg ( -1kbangptl4:GFP ) using the single BglII site located downstream of SV40polyA . A vector ( Tg ( in3 . 4-Mmu . Fos:GFP ) ) containing the angptl4 intronic module in3 . 4 was used as the source vector for site-directed mutagenesis . To create site-directed substitutions , 50 bp complementary primers containing two 20 bp regions complementary to in3 . 4 , separated by a 10 bp substitution block , were used in circular PCR followed by DpnI treatment to digest methylated parent plasmid . A ClaI restriction site was incorporated into the 10 bp region in order to screen for mutant bacterial colonies . Selection of nucleotide exchange was generally A–C and G–T , except in cases that would create a site amenable to DamI methylation . All plasmids were verified by Sanger dideoxy terminator sequencing . All primers used are listed in Table S2 . Co-injections of Tol2 plasmid and transposase mRNA were performed as described [36] . Generally , 100–200 zebrafish embryos were injected at the 1–2 cell stage with approximately 69 pg of plasmid DNA at a DNA∶transposase ratio of 1∶2 . Injections of each construct were performed with at least two sequence-verified plasmids in two independent experiments . Mosaic expression patterns were quantified as follows: at least 200 fish were visually observed , and at least 10 were scored per construct for positive/negative expression in selected tissues . At least 7–20 fish/construct were imaged at the same magnification and exposure time and densitometric measures were quantified in 8-bit grey scale images using ImageJ software [105] . Three mosaic patches within a given tissue of an imaged fish were quantified for mean fluorescence intensity and averaged . Statistical significance was analyzed using Kruskal-Wallis one-way analysis of variance and Dunn's multiple comparison test using GraphPad Prism software . Injected larvae were raised to adulthood and screened for stable germ-line insertion . Where indicated , patterns identified in mosaic animals were verified in a least two independent stable germ-line insertions ( 3 ) . In each case , independent pedigrees of the same Tol2 vector displayed the same specific pattern of expression in the intestine , liver , and islet , respectively . Staining of fixed and sectioned 6 dpf zebrafish was performed exactly as described [49] . Primary antibodies used in this study were anti-GFP ( Rabbit , 1∶500 , Invitrogen ) , 2F11 ( mouse , 1∶200 ) , 4E8 ( mouse , 1∶200; gifts from Julian Lewis ) , and secondary antibodies were AF568 ( goat anti-mouse , 1∶500 , Invitrogen ) and AF488 ( goat anti-mouse , 1∶500 , Invitrogen ) . Three intestines were dissected from adult zebrafish at 1 year post-fertilization , splayed , and washed extensively with 1× PBS . Intestines were incubated for 15 minutes on ice in 5 ml of Dissociation Reagent 1 ( 1× PBS , 30 mM EDTA , 1 . 5 mM DTT , 1× Complete protease inhibitors; Roche ) , then transferred to Dissociation Reagent 2 ( 1× PBS , 30 mM EDTA , 1× Complete protease inhibitors ) and shaken at 25°C until epithelial layers were sufficiently sloughed . Epithelial cells were collected , washed in 1× PBS , and re-suspended in 500 microliters of RSB ( 10 mM Tris-HCl pH 7 . 4 , 10 mM NaCl , 3 mM MgCl2 ) . Cells were gently lysed in 10 ml cold RSB plus 0 . 075% NP-40 and nuclei pelleted at 500× G at 4°C for 10 minutes . Nuclei were incubated with various concentrations of Dnase I ( 0–1 . 5 units , NEB ) for 10 minutes at 37°C . Reactions were stopped by adding an equal volume of 2× Lysis Buffer ( 1% SDS , 200 mM NaCl , 10 mM EDTA , 20 mM Tris pH 7 . 5 , 0 . 4 mg/ml proteinase K ) and incubated overnight at 37°C . Digested DNA was extracted using phenol/cholorform/isoamyl alcohol ( Fisher ) , precipitated with ethanol and sodium acetate , and quantified using a fluorimeter ( Qubit , Invitrogen ) . Quantitative PCR was performed as described above using primers listed in Table S2 . | Recent studies have revealed that the community of microorganisms residing in the intestine regulates fat storage . Microbes evoke this response in part by suppressing expression of the Angptl4 gene , which encodes a secreted inhibitor of fat storage . Although Angptl4 is expressed in multiple tissues , microbial suppression occurs only in the intestine . To determine how microbes control fat storage , we must elucidate the mechanisms underlying intestine-specific and microbial regulation of Angptl4 expression . Here , we take advantage of the unique features of the zebrafish model to define the regulatory DNA sequences controlling angptl4 expression . Our results reveal that different DNA regulatory regions within the angptl4 gene mediate expression of angptl4 in the intestine and other tissues . By assessing the evolution of angptl4 regulatory regions and subjecting them to structure-function analyses , we identify discrete DNA sequences that are required for intestinal expression . Strikingly , microbes suppress the activity of the intestine-specific regulatory region similar to the endogenous angptl4 gene . Therefore , intestinal microbes might regulate angptl4 production by suppressing the signaling pathway interpreted by an intestine-specific transcriptional regulatory region . Our results provide new mechanistic insights into how intestinal microbes might influence fat storage and contribute to the development of obesity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"microbiology",
"anatomy",
"and",
"physiology",
"animal",
"models",
"physiological",
"processes",
"developmental",
"biology",
"model",
"organisms",
"nutrition",
"gastroenterology",
"and",
"hepatology",
"aquaculture",
"molecular",
"genetics",
"gene",
"expression",
"marine",
"and",
"aquatic",
"sciences",
"biology",
"agriculture",
"physiology",
"genetics",
"computational",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | Intronic Cis-Regulatory Modules Mediate Tissue-Specific and Microbial Control of angptl4/fiaf Transcription |
Tumor cells do not develop in isolation , but co-evolve with stromal cells and tumor-associated immune cells in a tumor microenvironment mediated by an array of soluble factors , forming a complex intercellular signaling network . Herein , we report an unbiased , generic model to integrate prior biochemical data and the constructed brain tumor microenvironment in silico as characterized by an intercellular signaling network comprising 5 types of cells , 15 cytokines , and 69 signaling pathways . The results show that glioma develops through three distinct phases: pre-tumor , rapid expansion , and saturation . We designed a microglia depletion therapy and observed significant benefit for virtual patients treated at the early stages but strikingly no therapeutic efficacy at all when therapy was given at a slightly later stage . Cytokine combination therapy exhibits more focused and enhanced therapeutic response even when microglia depletion therapy already fails . It was further revealed that the optimal combination depends on the molecular profile of individual patients , suggesting the need for patient stratification and personalized treatment . These results , obtained solely by observing the in silico dynamics of the glioma microenvironment with no fitting to experimental/clinical data , reflect many characteristics of human glioma development and imply new venues for treating tumors via selective targeting of microenvironmental components .
Tumor cells and stromal cells actively “talk” to each other via an array of soluble signaling molecules , leading to co-evolution of the tumor and its microenvironment [1] , [2] , [3] , [4] , [5] . This also implies that the tumor microenvironment itself is a critical aspect of disease mechanism and that the microenvironmental components , including cells and soluble mediators , may represent a new set of targets for anti-tumor therapy [3] , [6] , [7] , [8] , [9] . However , due to the inherent heterogeneity of the tumor microenvironment and the complexity of the cell-cell communication network , it remains poorly understood at the systems level how these cells and their communication network collectively shape a heterogeneous tumor microenvironment and modulate tumorigenesis and metastasis . Conventional approaches that examine one or two selected pathways are incapable of fully assessing complex signaling networks and recapitulate the dynamics of the tumor microenvironment , and often result in contradictory conclusions . Thus , a systems approach that examines various cell types and the associated intercellular signaling networks in the tumor microenvironment is highly desired . In this work we choose to study the dynamics of glioblastoma multiforme ( GBM ) development . GBM is one of the most malignant brain tumors , with conventional therapies against “common” oncogenic targets usually ineffective due in part to the high degree of tumor heterogeneity . Astrocytes , microglia , and infiltrating immune cells actively interact with glioma and glioma stem cells via complex intercellular signaling networks mediated by an array of soluble signaling molecules , e . g . , cytokines , growth factors , and neuropoientins [10] . All these collectively shape a tumor microenvironment that could be distinct from one patient to another . Despite substantial research efforts and significant advances in cancer therapeutics , human GBM remains the most aggressive and lethal brain tumor in humans . In addition to inter-tumoral and inter-patient heterogeneity , GBM also exhibits significant intra-tumoral heterogeneity down to the single-cell level [11] , [12] . First , glioma cells originate from a variety of dynamically evolving progenitor cells [13] . It has been demonstrated that GBM cells demarcated by the neural stem cell marker CD133 exhibit much enhanced competencies for self-renewal and tumor initiation [14] , [15] . Recent studies have also shown instances in which CD133-negative cells were able to generate the same outcomes [16] , [17] , [18] , [19] . Second , glioma cells constantly interact with a variety of stromal cells . There is evidence that glioma cells acquire the ability to recruit and subvert their untransformed neighbor microglia into active collaborators to facilitate tumorigenesis . Direct correlation has been reported between the grade of glioma and the level of resident tumor microglia [20] , suggesting the mutual paracrine stimulation between microglial cells and glioma cells [21] , [22] , [23] , [24] . Microglial cells recruited by glioma can promote tumor growth [25] , [26] , [27] , dictated by paracrine loops responsible for glioma initiation and progression ( e . g . , IL-6 , IL-10 , TGF-β , prostaglandins , G-CSF , and GM-CSF , and growth factors such as EGF , VEGF , HGF , and SCF ) . The crosstalk between activated astroglial and glioma cells has also been documented , although the mechanism of their interactions has not been full revealed . For example , astroglial cells produce IL-1β [28] , [29] that promotes cell proliferation [30] , [31] , [32] and tumor angiogenesis [33] , [34] , [35] . Upon stimulation by the autocrine IL-1β these cells further secrete TNF-α and IL-6 [36] , [37] , [38] . The former was found to increase VEGF [39] , EGF receptor [40] , and MMP-9 [41] expression in glioma cells , suggesting that astroglia-produced cytokines may influence all the three most critical aspects of glioma cell survival: angiogenesis ( VEGF ) , proliferation ( EGFR ) , and migration ( MMP-9 ) . In silico models of tumor microenvironment integrate information about the biological context in which cancers develop , and thus represent a multi-scale consideration of oncogenesis as it occurs within somatic tissues [42] , [43] . Multiple factors involved in the development of an intrinsically complex tumor microenvironment have been studied including extracellular biomolecules , a spatially intricate and dynamic vasculature , and the immune system . Thus far , these models can be broadly divided into ‘continuum’ models , and discrete or ‘agent-based’ models as summarized in a review by Price and coauthors [43] . The latter describe the dynamics of individual interacting units , such as cancer cells , in small confined space; the former can be applied to a large tissue scale where agent-based modeling is computationally prohibitive . However , none of these methods have been integrated with a large cell-cell communication network in a complex tumor microenvironment . Herein we integrate all the intercellular signaling pathways known to date for human glioblastoma and generate a dynamic cell-cell communication network associated with the glioma microenvironment . Then we apply evolutionary population dynamics and the Hill functions to interrogate this intercellular signaling network and execute an in silico tumor microenvironment development . The observed results reveal a profound influence of the microenvironmental cues on tumor initiation and growth , and suggest new venues for glioblastoma treatment by targeting cells or soluble mediators in the tumor microenvironment .
Although much is known about the identities and biochemical activities of signaling molecules in the glioma microenvironment [1] , [2] , [3] , [4] , [5] , [44] , [45] , how these mediators coordinate and function collectively at the systems level to regulate tumor development is insufficiently understood . Here we first constructed an intercellular signaling network by incorporating all the autocrine/paracrine pathways known for human glioblastoma , as shown in the diagram of Fig . 1a . Five types of cells – quiescent and activated glioma initiating/progenitor cells , glioma cells , and astroglial and microglial cells – and a panel of 15 growth factors/cytokines/chemokines were included in the signaling network . Then we derived a quantitative model using stochastic population dynamics and the Hill functions . First , a basic population dynamic equation was employed to compute the growth rate of five cell types as a function of their proliferation rate , decay ( apoptosis ) rate , the rate of formation via direct mutation , the rate of formation via differentiation of their stem/progenitor cells , and the rate of de-differentiation . Second , the temporal growth rate of each cell type is also modulated by soluble signaling mediators present in the tumor microenvironment; this process is quantitatively described by the Hill functions . All differential equations are described in Supporting Text S1 and the initial settings of all parameters are detailed in Supporting Table S1 . As an example , we present here the procedure on how to construct the model for glioma cell population . It has been suggested that glioma can originate from cells at multiple differentiation stages during glial cell development , whereas the progenitor cells appear to be more susceptible to neoplastic transformation compared with mature glial cells [46] , [47] . Cytokine signalings , including IL-1 , IL-6 , IL-10 , TGF-β , EGF , VEGF , HGF , G-CSF , SCF , and MIF , participate in the mechanism of promoting GBM growth . PGE2 can transiently prevent glioma cell proliferation in vitro . EGF , FGF , and MIF are predominantly survival factors for GBM cells . To re-illustrate the underlying physics of this model , we show the population dynamics for glioma cells as in equation 1 , which integrate all the above signalings: ( 1 ) where ccell/cytoine is the concentration of cell/cytokine , f is the basal rate function , H1 , H2 , H3 , H4 , and H5 are Hill functions , L is a logistic function , and Aangiogenesis is defined as the angiogenesis factor . Similarly , the same algorithm was applied to derive population dynamics equations for other cells . More details may be found in Method and the sections 1&2 in Supporting Text S1 . The change of cytokines associated with tumor microenvironment development is described as the production and consumption by all the cells and modulation by other cytokines as revealed by prior experiments . For example , glioma stem cells , glioma cells , and microglial cells secrete substantial amounts of VEGF . MIF and TNF-α have been observed to induce a significant dose-dependent increase of VEGF . The dynamics of VEGF is thus governed by a differential equation ( Eq . 10 ) related to these cells and soluble mediators: ( 2 ) where f is the basal secretion/decay rate function and H is the Hill function . In the end , the temporal rate of growth and death of each cell population or the rate of production and decay of each cytokine is expressed as an ODE; a set of 20 inter-coupled ODEs were constructed to interrogate the dynamics of intercellular signaling network in a glioma microenvironment . To capture the stochastic nature of cell dynamics and cytokine signaling , we applied truncated Gaussian white noise , Poisson white noise , and bounded noise to describe the stochastic perturbation to production/regulation rate constants , recruitment rate , and proliferation/mutation/differentiation rates , respectively . Supporting Tables S1 and S2 and Methods give a complete description of all the signaling processes and summarize the input values for all differential equations . We performed an in silico stochastic study of glioma microenvironment development in a 1-ml control volume over a period of 12 months and observed a non-linear , synergistic co-evolution of all five cell types ( Fig . 1b–f ) . The dynamics of glioma cells ( GC ) exhibit three distinct phases ( Fig . 1b ) : the pre-tumor phase ( 1–5 months ) , the rapid expansion phase ( 6–10 months ) , and the malignant phase that corresponds to semi-steady high-grade glioblastoma ( 11–12 months ) . The starting cell populations are astrocyte ( 2 . 8×107/ml ) , microglia ( 2×106/ml ) , and quiescent stem cells ( QSC ) ( 1×104/ml ) . The initial conditions only change the quantitative timeline of the dynamics but would not affect the general trends observed in our model that properly reflect the dynamics of human glioma ( see Supporting Fig . S2 ) . The number of glioma cells at t = 0 is zero , and glioma cells develop via either neoplastic transformation of normal astrocyte or differentiation of glioma stem cells . The initial rate constants ( time = 0 ) are derived from literature reports [48] , [49] , and become gradually subjected to the modulation by soluble factors ( cytokines and growth factors ) . We observed that glioma stem cells are the major cell sources for glioma formation . At the early stage , QSCs upon stimulation are rapidly activated into activated stem cells ( ASC ) via a reversible process conferring self-renewal capability ( Fig . 1c ) . This step proceeds to completion within the first month . Then both QSCs and ASCs stay at a relatively steady state over the next four months before ASCs further differentiate into glioma cells in a stochastic manner . Despite the rapid lineage conversion of stem cells occurring as early as in the first month , glioma cells remain at a silent state with cell density way below the clinically detectable threshold . ( For all the experiments shown here , the threshold for detecting glioma in the clinic is assumed to be 1×106/ml , which is in agreement with the data from clinical studies [50] . ) During the growth of glioma cells within the space that astrocytes occupy , astrocytes strive to maintain their abundance as well as their functions until they are displaced by the glioma cells in the late stage . The number of microglial cells follows a steady increase all the way from the pre-tumor to the malignant stage , with a small kink occurring at the onset of rapid tumor expansion ( Fig . 1d ) . Although no “clinical” signs are observed at the first phase , the imperceptible changes occurring in the tumor microenvironment silently accumulate tumorigenic signals and eventually result in a switch of fitness dominance between astrocyte and glioma . The glioma cells acquire competitive advantages and are primed to rapid growth within a month to reach the diagnostic threshold ( ∼1×106/ml ) . This unique behavior is consistent with glioblastoma development observed in animal models [17] . It was assumed that after rapid expansion glioma cells follow a typical exponential growth mode in the next month until reaching a tumor cell concentration ( ∼1 . 5×107/ml ) , and then gradually turn into a slow growth phase dictated by the logistic growth model [51] , [52] . The astrocyte population shrinks due to competitive selection pressure exerted by a microenvironment unfavorable to astrocyte proliferation or favorable to astrocyte apoptosis that decreases the fitness advantages over time and eventually causes the loss of dominance . We examined the contribution of direct mutation of astrocyte and the differentiation of glioma stem cells to glioma growth . We observed that neoplastic transformation of astrocytes directly to glioma cells does result in the formation of small numbers of glioma cells in the pre-cancer phase , but contributes little to tumor development in rapid growth and expansion phases ( see Supporting Fig . S1 ) . The total cell concentration experienced a significant expansion during the seventh month , suggesting a density-gradient-driven potential for the glioma cells to invade neighboring tissues ( Fig . 1e ) . The total cell density we observed in the tumor microenvironment is higher than that in normal tissue , which is quantitatively consistent with the results obtained using tissue histology examinations [53] , [54] , [55] . A three-dimensional ( 3D ) stochastic simulation ( Fig . 1f ) shows that the evolution of all the cell types and the time course are consistent with clinical glioblastoma development . Cytokine dynamics also exhibit multi-stage non-linear characteristics ( Fig . 2a ) . Activated microglial cells were found to be an important source of cytokines in the early stage , yielding a steady increase of cytokine concentrations prior to the emergence of tumor . These cytokines participate in the modulation of rapid glioma cell expansion in the later stage , suggesting that microglial cells may play an important role in tumor initiation by priming glioma cells at very low concentrations . Glioma cells also secrete paracrine signaling factors that promote the proliferation and migration of microglia , and thus in turn benefit from the increase of microglia cells that reside in the vicinity of the glioma growth front . The normalized dynamics curves ( Fig . 2b ) show that 15 cytokines fall into three categories according to their time traces . TNF-α peaks at the end of the first phase , then gradually decreases presumably due to the consumption by glioma cells ( e . g . , rebind to TNF receptors and trigger the secondary signaling cascades ) . IL10 and PGE2 show a monotonic increase across all the three phases . All the other cytokines exhibit a rapid concentration increase in the second phase and reach a quasi-steady state correlated with the glioma population dynamics . We first designed a novel therapy by targeting the cellular components of the tumor microenvironment . According to cell population dynamics ( Fig . 1 ) , microglial cells produce an array of cytokines that often prime glioma cells to predispose them to rapid population expansion in the sixth month , and thus function as a tumor-promoting factor in the tumor microenvironment . Therefore , we designed a cell-targeting therapy that eliminates microglial cells in the tumor microenvironment . This therapy is realized by arbitrarily increasing the apoptotic rate of microglia by 10 times at the early , middle , and middle to late stages with the corresponding glioma cell density at 5×104/ml , 2×105/ml , or 1×106/ml , respectively . To examine the applicability of this therapy to patients with different biomolecular background and assess the effect of inter-patient heterogeneity on therapeutic response , three virtual patients with different profiles of initial parameters ( cytokine production rate , receptor expression level , etc . ) within the ranges reported in the literature [56] ( Supporting Table S5 ) were treated using the same microglia depletion therapy at three different stages . The results are compared as shown in Fig . 3a–c . Two interesting features were observed in the microglia depletion therapy experiments . First , all patients responded in a similar manner although the length of therapeutic benefit and the recurrence time varied from one patient to the other . Second , the efficacy strongly depends on how early the treatment was given to the patients ( Fig . 3d ) . All the patients treated at the early stage when glioma cell density ( ∼5×104/ml ) is far below the threshold for clinical tumor detection ( 1×106/ml ) showed no recurrence within the time of simulation . Treatment given at the early to middle stage ( glioma cell density ∼2×105/ml ) postpones the rapid tumor growth phase by two to four months and does give the patient therapeutic benefit . Patients treated right as the clinical sign emerges ( glioma cell density∼1×106/ml ) did not respond at all in terms of glioma growth rate , suggesting that tumor cells have been fully primed and become self-sustained with no need of paracrine signaling to drive glioma cell proliferation . These results , obtained by unbiased integration of basic biochemical parameters and cell signaling processes , were found to appropriately reflect clinical and experimental observations . There is a consensus that activated microglia promote glioma growth and promotion , which is consistent with our in silico glioma development experiments [20] , [39] , [40] , [41] . Recently , an animal model study indicated that clonal cooperation between different mutant cells can lead to tumor formation , whereas any single-cell type alone cannot develop into tumor [57] . What is more interesting is that the second clone , once activated by the first clone presumably through cytokine signaling , becomes fully self-sustained and develops into tumor without the presence of the first clone , which is strikingly similar to the glioma-microglia interaction observed in our model , and thus may share commonalities in molecular and cellular mechanisms . Our study suggests that cells in the tumor microenvironment can be good targets for therapeutic intervention or control of tumor progression , pointing to new venues for anti-tumor drug design and development . The results of microglia-depletion therapy indicate that patients do not show significant responses unless they are diagnosed at the very early stage – the time when no clinically detectable tumors have been formed . Thus , we turn to assess the possibility of combination therapy that directly targets a number of key cytokine signaling pathways , which is anticipated to give more focused and potent therapeutic effects . Due to inter-tumoral heterogeneity , the best therapeutic regimen must be an individually tailored combination of inhibitors that act on selected cytokines or their receptors optimized for the patient . We performed a sensitivity analysis to assess the tumorigenic potential of each cytokine and find the primary targets that , once subjected to blockade or promotion , exhibit the most effective responses in therapeutic intervention . The Methods and section 3 in Supporting Text S1 describe the details of this analysis . Basically , it measures the length of time taken by glioma cells to grow from the threshold concentration ( e . g . , 1×106/ml ) to an objective concentration ( e . g . , 1 . 5×107/ml ) reflecting the survival time of a patient after the therapy is given . Twenty-nine tests , each perturbing a cytokine production rate or a cytokine receptor expression level , were performed to give the sensitivity factor of each cytokine or its receptor with respect to patient survival probability . According to the results , forced activation of a signaling pathway with a positive sensitivity factor is expected to promote patient survival , and vice versa . Individualized combination therapy is designed by enhancing the signaling processes of cytokines with the largest “positive” sensitivity factors and inhibiting those with the largest “negative” sensitivity factors . To test this therapy , the same virtual patients ( patients 1 , 2 , and 3 ) that were randomly designed for microglia depletion experiments are examined here to generate sensitivity factor profiles for every patient ( Fig . 4a ) . Next we designed a four-cytokine combination therapy ( VEGF , MIF , IL6 , and HGF ) optimized for patient 1 , and all the patients were given the same treatment for comparison . First , we compared single-target therapy and combination therapy that are administered at the time of glioma cell density∼1×106/ml ( Fig . 4b and Supporting Fig . S3 ) . Although each of the four cytokines has a large negative sensitivity factor for promoting tumorigenesis , therapies that inhibit only one of these cytokines can hardly alter the time course of tumor progress , due to the homeostatic robustness of the cytokine network and the resulting intrinsic resistance to perturbation . To overcome this issue , we further applied to virtual patient 1 a combination treatment that simultaneously inhibits all four cytokines , and we observed substantial therapeutic responses that cannot be simply explained by the additive effect ( Supporting Fig . S4 ) . Second , the same therapy was given to patients 2 and 3 , but did not yield positive therapeutic responses ( Fig . 4c and Supporting Fig . S5 ) ; patient 2 exhibited a modest benefit by one month and patient 3 almost did not respond at all . Figure 4d shows the results of a 3D stochastic simulation of cell population dynamics in response to combination therapy administered at different times . Considering that these treatments were administered at a middle to late stage when clinically detectable tumors had already developed , we conclude that the combination therapy tailored to match individual patients is more focused and can give better therapeutic benefit even when microglia depletion therapy fails in the middle to late stages , highlighting the critical need for molecular diagnosis and patient stratification prior to the design of a combination therapy that targets the tumor microenvironment .
To the best of our knowledge , this is the first study that attempts to integrate a variety of cells and their intercellular signaling pathways into a cell-cell communication network and assess how this network controls tumor initiation and progression at the systems level . Through in silico experimentation of tumor microenvironment development , the dynamics of cells and cytokines correctly reflects general trends of tumorigenesis observed experimentally or clinically [17] , [51] , [58] . We also discovered interesting phenomena that can be seen only at the systems level and are often masked in conventional tumor biology studies . First , the cell population dynamics obtained using a set of coupled differential equations based upon population dynamics and the Monte Carlo method yield the full time courses of all five cell types . Although significant inter-patient heterogeneity has been observed , the time courses of glioma microenvironment development for all virtual patients we encountered do share common characteristics and all exhibit three-phase non-linear evolution dynamics . For example , all patients experience the pre-tumor phase; the mutual paracrine stimulation between microglial cell and glioma cell results in the continued growth of microglia . These results , obtained via in silico experimentation without fitting or optimization to any specific clinical or experimental data , were found to well reflect the general mechanisms of glioma development [17] , [20] . Second , soluble signaling proteins , e . g . cytokines , are the key components mediating the cell-cell communication network in a tumor microenvironment . We successfully integrated 15 cytokines in 69 paracrine/autocrine pathways in the cell population dynamics model . We further examined relative weight factors for all the paracrine/autocrine loops associated with tumor development . This study provides new insights into tumor microenvironment development and suggests that therapies targeting the cytokine-mediated intercellular signaling network in a tumor microenvironment need to be personalized . Third , we designed a microglia depletion therapy by adding a virtual drug in the tumor to increase the microglia apoptosis rate . The observation from in silico experimentation indicates that this therapy shows some efficacy only when patients are treated at very early stages , which is consistent with the general outcomes of anti-cancer treatment , but provides a new mechanism to explain the therapeutic resistance observed in the clinic . The ineffectiveness of microglia-targeted therapy in the middle to late phases indicates the emergence of an autocrine-dominant , self-propelled glioma proliferation . Then , we moved to look for another therapy that directly targets multiple key cytokines to assess the possibility of treating glioma in the middle to late stages . It turns out a more focused combination therapy can suppress tumor growth at the middle stage when the tumor becomes clinically detectable and microglia-depletion therapy is ineffective . Further study on virtual patients reveals inter-patient heterogeneity in response to the same combination therapy , and highlights the importance of designing therapy individually tailored to the patient's tumor microenvironment . While current anti-cancer drugs mostly target tumor cells , this study indicates the possibility and quantitatively assessed the effectiveness of new therapies that target cellular or molecular components of the tumor microenvironment , pointing to completely new venues for tumor control and treatment . In the end , a model as reported herein may serve as a tool to integrate clinical data obtained from informative molecular diagnosis of patients , predict the dynamics of tumor progression , and aid the design of personalized therapy . The technologies for such informative diagnosis are anticipated ( 1 ) to measure both tumor cells and a variety of cells in a tumor microenvironment , and ( 2 ) to analyze cytokine secretion profiles at the single-cell level such that a cytokine-mediated cell-cell communication network can be re-constructed for any individual patient . Currently , such technologies are not yet available in the clinic , but there have been significant research efforts in the past years that aim to develop single-cell proteomics technologies and clinical microchips for informative diagnosis of complex diseases including cancer [59] , [60] , [61] , [62] , [63] . In the future , integration of such technologies and the model described here can turn into a powerful clinical tool to diagnose the tumor microenvironment and the associated intercellular signaling network in individual patients and truly enable personalized therapy by selective targeting of the tumor microenvironment .
While the microenvironment exerts a significant selective pressure on the tumor , the tumor cells persistently reshape their microenvironment to synergistically support the growth and spread of the tumor . The dynamically changing levels of signaling molecules that rewire tumor-stromal interactions along with tumor progression will provide insights into the mechanisms of disease development . Since this work is focused on predicting tumor time course evolution , the model is based , for simplicity , on a well-mixed species system . Five types of cells ( quiescent and activated glioma stem/progenitor cells , glioma cells , astrocytes , and microglial cells ) and 15 growth factors/cytokines/chemokines are integrated in this model . We assume that the species included in the model evolve independently of species excluded from the model ( oligodendrocyte , etc . ) . In the end , we present the intercellular signaling network as a set of coupled ordinary differential equations in terms of population dynamics . The rates of change of cells are expressed by the conversion rate , proliferation rate , and decay rate ( see Supporting Text S1 ) . A set of coupled stochastic ordinary differential equations describing the co-evolution of tumor microenvironment is constructed using population dynamics and stochastic dynamics . The basic mathematical model is based on continuous logistic proliferation and discrete event type fluctuation , and can be described as ( 3 ) ( 4 ) where the first term of the right-hand side of Eq . ( 3 ) is the stochastic representation of discrete event–type fluctuation , including immigration , emigration , and production from progenitor . is the magnitude of the kth discrete event , i . e . , the number of cells increasing ( decreasing ) at time point . denotes a non-homogeneous Poisson counting process with arrival rate function ( i . e . , the number of events per unit time ) and gives the number of events that arrive in the time interval . The second term indicates the logistic proliferation of cell with a basic rate function , which can be up-regulated by cytokine and inhibited by . Parameter is the saturating concentration factor , whereas is the angiogenesis factor . The initial exponential growth will slow down and the cell concentration level is approached slowly in the late time . The third term is the decay due to natural lifespan that can be regulated by cytokine . The last term describes the mutation/differentiation/dedifferentiation from cell under stimulation of cytokine . The first term on the right hand side of Eq . ( 4 ) quantifies the production of cytokine from cell with basic secretion rate function , and the secretion is stimulated by cytokine and inhibited by . The last term is the decay term with half-life , and can be regulated in the presence of cytokine . To further assess how fluctuations in biological processes reflect the random nature and affect the performance of the system , we introduce the following stochastic process interpretation of the rate parameters: ( 5 ) ( 6 ) ( 7 ) ( 8 ) ( 9 ) ( 10 ) where W ( t ) is a standard Wiener process , and ξ ( t ) is a zero-mean Gaussian white noise with unit intensity . The sections 1&2 in Supporting Text S1 give a full set of deterministic ordinary differential equations ( ODE ) and detailed explanations of stochastic description . The stochastic dynamics are studied using Monte Carlo simulations . The corresponding time series of the species concentration are obtained by integrating these differential equations numerically using the fourth-order Runge-Kutta scheme or the fifth-order Dormand-Prince method . The parameters are assigned in the range over which the model output most closely matches experimental observation ( Supporting Table S1 ) . Although we calibrate the model with data from the literature , the model parameters can easily be changed to patient-specific clinical parameters as needed . To systematically evaluate the influence of each cytokine on tumorigenesis rate , we conduct a sensitivity test , in which the sensitivity factor of cytokine xi can be calculated as ( 11 ) where F ( x ) is the objective function ( e . g . , tumorigenesis time , cell density , cytokine concentration ) , and x0 is the local parameter profile . The results show marked effects of these cytokines on the development of glioma and suggest the possibility of designing therapeutic intervention by targeting cytokine signaling loops ( both cytokine production and receptor expression level ) ( Supporting Fig . S6 and Table S6 ) . The quantitative results are also found to be context specific; the exact time for observing tumor formation ( 1×106 cells/ml ) depends on the profile of all initial parameters for each virtual patient ( Supporting Tables S3 and S4 ) . The greater the difference between cytokine sensitivity factor landscapes , the greater is the inter-patient heterogeneity . In addition to the quantitative manifestation of inter-patient heterogeneity , sensitivity analysis also points to a venue to identify a cytokine profile that potentially can serve as a molecular signature for tumor sub-classification , and thus provides a means to stratify patients via their cytokine profiles and to design individualized treatment . | Tumor cells do not develop in isolation , but co-evolve with stromal cells via an array of soluble mediators . Here we report a model to integrate prior biochemical data and construct a glioma microenvironment in silico , which comprises 5 types of cells , 15 cytokines , and 69 signaling pathways . We observed a transition of the cytokine network from the microenvironmentally controlled , paracrine-based regulatory mechanism to the self-sustained , autocrine-dominant malignant state . A microglia-depletion therapy and a cytokine combination therapy were then designed and show significant efficacy on virtual patients . However , the optimal response depends on both the time the therapy is given and the molecular profiles of individual patients , suggesting the need for informative diagnosis and personalized treatment . These results , obtained solely by observing in silico tumor dynamics with no fitting to experimental/clinical data , reflect many characteristics of human glioma development and suggest new venues for anti-tumor treatment by selectively targeting microenvironmental components . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biology",
"computational",
"biology"
] | 2012 | In silico Experimentation of Glioma Microenvironment Development and Anti-tumor Therapy |
Bardet-Biedl syndrome , BBS , is a rare autosomal recessive disorder with clinical presentations including polydactyly , retinopathy , hyperphagia , obesity , short stature , cognitive impairment , and developmental delays . Disruptions of BBS proteins in a variety of organisms impair cilia formation and function and the multi-organ defects of BBS have been attributed to deficiencies in various cilia-associated signaling pathways . In C . elegans , bbs genes are expressed exclusively in the sixty ciliated sensory neurons of these animals and bbs mutants exhibit sensory defects as well as body size , feeding , and metabolic abnormalities . Here we show that in contrast to many other cilia-defective mutants , C . elegans bbs mutants exhibit increased release of dense-core vesicles and organism-wide phenotypes associated with enhanced activities of insulin , neuropeptide , and biogenic amine signaling pathways . We show that the altered body size , feeding , and metabolic abnormalities of bbs mutants can be corrected to wild-type levels by abrogating the enhanced secretion of dense-core vesicles without concomitant correction of ciliary defects . These findings expand the role of BBS proteins to the regulation of dense-core-vesicle exocytosis and suggest that some features of Bardet-Biedl Syndrome may be caused by excessive neuroendocrine secretion .
Bardet-Biedl Syndrome is a rare , multigenic , pleiotropic disorder characterized by defects in many tissues including the eyes , kidneys , central nervous system , and reproductive organs [1] . The clinical presentations of BBS include broadly prevalent disease conditions such as obesity and retinopathy . Mutations in at least 14 genes cause BBS [1]–[3] . Of these , seven interact to form a complex termed the BBSome [4] . The mammalian BBSome associates with ciliary membranes and interacts with a guanine nucleotide exchange factor ( GEF ) for Rab8 , a Rab GTPase implicated in cilia formation and function [4] . The BBSome is thought to act as a coat protein for cilia-bound vesicles , promoting the sorting and delivery of signaling molecules to the cilium [5]–[7] . Given that cilia are sensory organelles enriched in signaling molecules , the molecular functions of the BBSome have led to the current paradigm for understanding Bardet-Biedl Syndrome in which defects in cilia formation and signaling have been advanced as a unifying explanation for the wide range of phenotypes seen in BBS patients [3] , [8]–[11] . In support of this notion , conditional knockouts in components of intraflagellar transport , IFT , a multi-protein machinery required for building the cilium , have been shown to result in some of the phenotypes present in bbs-deficient mice such as obesity and kidney disease [12]–[14] . In C . elegans , the seven BBSome components are conserved and are exclusively expressed in the 60 sensory neurons of this animal [8] . Various subsets of these neurons have been implicated in sensation of environmental cues that affect animal growth , metabolism , and feeding behavior [15]–[17] . Disruptions of the C . elegans BBSome cause altered rates of IFT , defects in the structural integrity of sensory cilia , and diminished behavioral responses to various sensory cues [8] , [9] , [18] , [19] . Thus , in C . elegans as in mammals , loss of BBSome components leads to phenotypes that are also seen in IFT mutants . Consequently , the physiological abnormalities of C . elegans bbs mutants have also been attributed to defects in cilia formation and signaling [9] , [18] . Here we report that BBSome mutants display a dramatic increase in the secretion of dense-core vesicle cargoes from ciliated sensory neurons , while mutations in IFT components generally result in reduced secretion , indicating that the consequences of bbs deficiency cannot be fully explained by impaired intraflagellar transport . We show that the enhanced secretion of bbs-deficient animals depends on the evolutionarily conserved Rab27/rabphillin/CAPS exocytosis machinery but not on Rab8 , which participates in vesicular transport to the cilium . Importantly , we show that the altered size , feeding , and metabolic phenotypes of bbs mutants can be normalized to wild-type levels by abrogating the enhanced secretion of these mutants without simultaneous correction of ciliary defects . We also demonstrate that while certain phenotypes of IFT mutants mimic those caused by bbs mutations , distinct neural and molecular mechanisms underlie these phenotypes . These findings expand the role of the BBSome to the regulation of dense-core vesicle secretion and raise the possibility that enhanced neuroendocrine secretions rather than ciliary defects per se may be primarily responsible for some of the clinical features of BBS patients .
To survive and thrive in a dynamic environment , C . elegans , as in other animals , must sense changes in environmental conditions and respond through behavioral , physiological , and metabolic adaptations . C . elegans accomplishes this in part by secreting a diverse array of neuroendocrine hormones including insulin-like peptides from their various ciliated sensory neurons [20] . We had previously shown that insulin secretion from the ADL pair of ciliated sensory neurons can be assessed by expressing a fluorescently tagged insulin , DAF-28-mCherry , exclusively in this pair of neurons , and measuring the accumulation of the secreted insulins in coelomocytes , scavenger cells that non-specifically endocytose pseudocoelomic fluid [21] . It is thought that peptidergic secretions from neurons such as ADL , which are not directly in contact with the pseudocoelom , gain access to this space through the extracellular matrix and the ancillary pseudocoelom , a fluid fill cavity that bathes many of the neurons in the head of the worm . Eventual passage of secreted molecules from the ancillary pseudocoelom to the pseudocoelom is thought to be mediated by tight junctions . Coelomocytes within the pseudocoelom non-specifically endocytose pseudocoelomic fluids , concentrating them in endocytic vesicles; thus , the steady state level of circulating signaling molecules secreted from ADL can be measured by quantifying the amount of fluorescent insulins within the coelomocytes [21]–[23] . To better understand the molecular mechanisms that link environmental cues to secretion of insulin-like peptides , we examined the effects of various defects in cilia formation and function on insulin secretion from the ADL pair of sensory neurons . We found that mutations in intraflagellar transport , IFT , components such as osm-3 , osm-5 , che-2 , or che-11 [24] , which result in defective cilia , cause a ∼50% reduction in insulin secretion ( Figure 1A , B , E , F ) consistent with the idea that appropriate cilia function is required for detection of food-related cues and subsequent coordination of metabolic and growth pathways through insulin signaling . We were therefore surprised to find that cilia-defective osm-12 mutants [18] , the C . elegans homolog of bbs-7 , exhibit a dramatic 2–3-fold increased insulin accumulation in coelomocytes ( Figure 1A–D ) . Mutations in bbs-1 , bbs-5 , bbs-8 , and bbs-9 , encoding other components of the BBSome , also caused enhanced accumulation of the insulin reporter in coelomocytes ( Figure 1C , D ) , suggesting that the entire BBSome complex regulates insulin release from ADL sensory neurons . The increased secretion was not limited to ADL neurons , as bbs mutants also exhibited increased secretion when the fluorescent insulin reporter was expressed exclusively in the ASI pair of sensory neurons ( Figure 1E , F ) . Additionally , BBSome mutants exhibited a ∼2–3-fold elevated secretion of DAF-7 , a neurally expressed TGF-β ligand and FLP-21 , an FRMF-amide neuropeptide expressed under the ADL specific , srh-220 , promoter ( Figure 1G , H ) , indicating that BBSome mutants have elevated release of various dense-core vesicle cargoes from multiple ciliated sensory neurons . To assess whether the enhanced accumulations of the fluorescent reporters in coelomocytes indicate functional increases in the release of insulins and neuropeptides , we assayed phenotypes associated with hyperactive insulin and neuropeptide signaling . Insulin and TGF-β signaling pathways regulate whether C . elegans grow reproductively or enter a hibernating dauer stage [25] . Reductions in either of these parallel pathways initiate dauer entry , while favorable conditions promote dauer exit and reproductive growth . Accordingly , overexpression of either ins-4 or daf-28 , encoding distinct insulins , partially suppresses the constitutive dauer formation phenotype of TGF-β mutants [22] . We found that mutations in either bbs-7 or bbs-9 partially suppressed the dauer phenotypes of daf-7 or daf-1 mutants , encoding the TGF-β ligand and its receptor , respectively , but not that of mutants in daf-2 , the insulin receptor ( Figure 2A and unpublished data ) . Consistent with the notion that the enhanced secretion of bbs mutants mediates dauer suppression , loss of tom-1 , an inhibitor of the dense-core vesicle release [26] , also partially abrogated dauer formation of TGF-β mutants ( Figure 2A ) . By contrast , mutations in IFT components such as osm-5 and che-11 failed to alter dauer formation of TGF-β mutants [27] . Therefore , BBSome mutants display enhanced insulin signaling and have the same impact on dauer formation and maintenance as animals with enhanced dense-core vesicle secretion and as such are distinct from other ciliary defective mutants . To functionally assess neuropeptide signaling in bbs mutants , we examined the NPR-1 signaling pathway . FLP-21 is one of two FRMF-amide neuropeptides that activates NPR-1 , a neuropeptide-Y-like receptor that regulates C . elegans oxygen sensation and social feeding [28] . Overexpression of FLP-21 can partially suppress the hypomorphic npr-1 ( 215F ) but not the null mutation of npr-1 as the hypomorph can still bind to FLP-21 , albeit with a lower affinity [28] . We found that mutations in either bbs-7 or bbs-9 also partially suppress the social feeding phenotype of npr-1 ( 215F ) but not that of the null mutant , suggesting that BBSome mutants , as our reporter assay indicated ( Figure 1G , H ) , have increased FLP-21 release ( Figure 2B ) . The hypersecreting tom-1 mutants similarly strongly suppressed the hypomorphic allele of npr-1 but only weakly suppressed the null allele ( Figure 2B ) . These findings indicated that the enhanced secretion of neuropeptides as assessed by fluorescent reporters indeed correlate to functional increases in neuropeptide signaling in BBSome-deficient animals . To determine if the effects of the BBSome on dense-core vesicle secretion are cell-autonomous , we expressed wild-type bbs-1 cDNA under the control of various sensory neuron promoters in bbs-1 ( ok1111 ) mutants . This bbs-1 transgene was fully functional as judged by its capacity to fully rescue the small body size , feeding , and dye-filling defect of bbs-1 mutants ( Table S1; unpublished data ) . Expression of bbs-1 under its own promoter or that of ocr-2 , which drives expression in ADL in addition to a few other sensory neurons ( see Table S1 for expression pattern ) , abrogated the enhanced secretion of bbs-1 mutant animals ( Figure 3A , B ) . By contrast , use of the tax-4 promoter to drive expression in numerous sensory neurons excluding ADL had no effect on the secretion phenotype of bbs-1 mutants ( Figure 3A , B ) . Expression exclusively in ADL neurons showed partial but highly significant reduction in the enhanced insulin secretion of bbs-1 mutants , suggesting that the BBSome regulates secretion at least partially cell autonomously ( Figure 3A , B ) . The partial suppression could be due to differences in promoter strength or regulation of secretion from other sensory neurons . Similar results were obtained when a functional bbs-7 cDNA was expressed in a bbs-7 mutant: two independently generated transgenic lines showed partial but highly significant reductions of the enhanced insulin secretion of bbs-7 mutants when expressed exclusively in ADL ( Figure 3C , D ) . Finally , use of a heat shock inducible promoter to induce expression of wild-type bbs-7 in late fourth larval stage bbs-7 mutants when the developmental programs for ciliated neurons have been completed led to a partial but significant abrogation of enhanced secretion ( Figure 3E , F ) . Thus , the requirement of the BBSome for wild-type secretion is , in part , cell autonomous and not a consequence of altered neural development . As the cilia and BBS proteins have been shown to regulate transcription through the TCF/LEF1 transcription factor [29] , we investigated whether the observed enhanced secretions could be secondary consequences of increased transcription in bbs mutants . While expression of FLP-21-mCherry translational reporter fusion using the ADL-specific srh-220 promoter resulted in increased accumulation of the secreted FLP-21-mCherry in coelomocytes ( Figure 1H ) , expression levels of the Psrh-220::gfp transcriptional reporter were similar in wild type and bbs-9 mutants ( Figure S3A , B ) . Furthermore , transcript levels in several bbs mutants , as measured by semi-quantitative RT-PCR , did not show any significant increase in transcription of srh-220 , daf-28 , flp-21 , daf-7 , or several other neuropeptide genes endogenously expressed in ADL and ASI neurons ( Figure S3C ) . Thus , enhanced secretion levels of insulin , TGF-β ligand , and neuropeptides and the corresponding increase in their associated signaling pathways are unlikely to be secondary consequences of enhanced transcription of insulin and neuropeptide genes . To better understand the enhanced secretions of bbs mutants , we next investigated the requirements of various components of the dense-core secretion machinery . Regulated release of dense-core vesicles is triggered by Ca+2 entry leading to activation of UNC-31/CAPS , which promotes vesicle fusion with the plasma membrane [30] . We previously showed that UNC-31 is required for insulin release from ADL [21] . Therefore , we investigated whether enhanced secretion of dense-core vesicle contents seen in bbs mutants was dependent on UNC-31 . We found that secretion levels of bbs-7; unc-31 double mutants were indistinguishable from unc-31 mutant alone ( Figure 4A , B ) , indicating that the BBSome is a negative regulator of Ca+2 stimulated release of dense-core vesicles and does not affect the constitutive basal release seen in unc-31 mutants . UNC-31 acts in parallel to TOM-1 , the C . elegans tomosyn [26] . Tomosyns negatively regulate both dense-core and synaptic vesicle fusions by binding to and inhibiting syntaxins , an essential component of the vesicle fusion machinery [26] . As in bbs mutants , tom-1 mutants exhibited increased secretion from ADL ( Figure 3A , B ) and the tom-1 and bbs-7 secretion phenotypes were additive ( Figure 4A , B ) . The finding that loss of unc-31 fully abrogated the hypersecretion of bbs mutants cannot distinguish whether the BBSome regulates secretion through mechanisms that directly modulate the unc-31 pathway from indirect mechanisms that ultimately depend on UNC-31 mediated secretion . However , since the hypersecretion of bbs mutants was fully abrogated by unc-31 but additive with that of tom-1 mutants , these findings suggest that UNC-31 and the BBSome act in a common release pathway in parallel to TOM-1 . Dense-core vesicles are trafficked from their site of formation in the Golgi to the plasma membrane by Rab GTPases and their regulators . Rab GTPases are a highly conserved family of proteins involved in various aspects of vesicle fusion and transport . Previous studies have indicated that the BBSome regulates Rab8 localization and activity in mammalian cells [4] . bbs and rab-8 mutants in C . elegans show similar defects in cilia membrane morphology , suggesting that the BBSome is also likely to regulate Rab8 activity in C . elegans [31] . However , rab-8 mutants exhibited wild-type levels of secretion from ADL and loss of rab-8 did not change the hypersecretion phenotype of bbs-7 mutants ( Figure 4C , D ) . This suggests that the BBSome might regulate dense-core vesicle secretion through different Rabs . Previous studies have suggested that Rab3 and Rab27 regulate vesicular exocytosis by promoting the movement and tethering of dense-core vesicles to the plasma membrane [32] , [33] . We found that AEX-6 , the C . elegans homolog of Rab27 , was essential for dense-core vesicle secretion from ADL and that its loss abrogated the enhanced secretion of bbs mutants ( Figure 4E , F ) . By contrast , loss of rab-3 , another GTPase implicated in exocytosis , had no effect ( Figure 4E , F ) . The enhanced secretion of bbs mutants was also dependent on rabphilin/RBF-1 , an effector of RAB-27/AEX-6 ( Figure 4G , H ) , and AEX-3 , a RabGEF previously shown to regulate RAB-27/AEX-6 and RAB-3 ( Figure 3C , D ) [32] . Taken together , these data are consistent with a model whereby the BBSome acts as a negative regulator of AEX-6/RBF-1/UNC-31 , the worm counterpart of mammalian Rab27/rabphilin/CAPS exocytosis machinery . Given that the mammalian BBSome binds to Rabin8 , a RabGEF for Rab8 , and regulates Rab8 activity [4] , the BBSome might regulate dense-core vesicle secretion by regulating the activity of AEX-3 , the RabGEF for AEX-6/Rab27 . To further investigate the effects of bbs deficiency on dense-core vesicles , we examined subcellular localization of IDA-1-GFP , a tyrosine phosphatase-like receptor involved in protein secretion that has been used as a maker of dense-core vesicles [34] . The overall expression levels of IDA-1-GFP was similar in wild type and bbs mutants , and in both cases GFP puncta were clearly visible along axons as well as dendrites terminating in cilia ( Figure S5 and unpublished data ) . While the IDA-1-GFP reporter is largely excluded from the cilia of wild type animals , it was prominently visible in those of bbs mutants ( Figure S5 ) . Presence of the marker in cilia of bbs mutants could suggest either mis-sorting of proteins normally localized to dense-core vesicles or the appearance of intact dense-core vesicles , which are normally excluded from cilia , within these structures . One mechanism that has been proposed to ensure proper segregation of cilia-localized membrane proteins from other plasma membrane proteins is the proteinaceous structure at the base of the cilia known as the transition zone [35] . The transition zone is thought to provide a barrier that helps exclude non-ciliary proteins from the cilia [35] . Mutations in components of the transition zone underlie ciliopathies such as Merkel-Gruber syndrome and nephronophthisis , which share some clinical features such as renal abnormalities with Bardet-Biedl syndrome [2] . Unlike bbs mutations , however , mutations in genes encoding components of the transition zone , such as mksr-1 and mksr-2 did not cause enhanced neuroendocrine secretion ( Figure S2 ) . The finding that loss of BBSome components enhances secretion of dense-core vesicles prompted us to examine the relationship between hyperactive neuroendocrine signaling and ciliary defects of bbs mutants . Since losses of various IFT components result in reduced secretion , we first investigated the requirement of functional cilia for the enhanced secretion of bbs mutants . We found that mutations in each of che-2 and che-11 , which cause reduced secretion levels ( Figure 1A , B ) , completely abrogate the enhanced secretion phenotype of bbs mutants ( Figure S2 ) . These findings further validate the notion that the excess secretion phenotype of bbs-deficient animals is not simply a consequence of defective cilia but rather that some level of normal ciliary function is required in order for the secretion phenotype of bbs mutants to be manifested . We next investigated whether ciliary defects of bbs mutants were dependent on hypersecretion of dense-core vesicles . To assess structural integrity of cilia in various mutants , we used a dye-filling assay . Subsets of C . elegans sensory neurons have their ciliated endings located near the surface of the animals and are directly exposed to the environment for chemosensation . Upon immersion of whole animals in a fluorescent dye solution , these neurons become visible as the dyes can enter these neurons through their cilia ( Figure S1 ) [36] . Since animals with defective cilia formation such as the IFT mutants fail to dye fill ( Figure S1B ) [36] , this assay has been used to assess cilia integrity and proper opening of the cilia to the external environment . As previously reported , we found that BBSome mutants also failed to dye fill ( Figure S1 ) [18] , [37] , [38] . Interestingly , while mutations in the Rab27/rabphilin/CAPS exocytosis machinery abrogated the enhanced secretion of bbs mutants , they did not restore structural integrity to cilia in these animals as judged by the inability of the double mutants to dye fill ( Figure S1C , D ) . These findings suggested that the structural and functional defects of cilia in bbs mutants are unlikely to be a consequence of hypersecretion of dense-core vesicles . To determine if ciliary structural deficiencies that manifest as dye-filling defects are a pre-requisite for hyperactive secretion , we examined the hypersecretion tom-1 mutants and found that they exhibited normal dye filling ( Figure S1B ) . We also found that while bbs-5 mutants display similarly enhanced levels of secretion as other BBSome mutants ( Figures 1C , D and 5A ) , they dye fill like wild-type animals ( Figure S1A ) . Additionally , bbs-5 mutants displayed normal expression of the serotonin synthesis enzyme , tph-1 ( Figure S4C , D ) , whereas expression of this enzyme was shown to be elevated in many cilia-defective mutants [39] ( Figure S4C , D ) , further suggesting the presence of functional cilia , in bbs-5 mutants . Together these findings suggest that the enhanced secretion phenotype of bbs mutants depends on retention of some level of ciliary function and that ciliary structural defects are not a pre-requisite for enhanced release of dense-core vesicles . In turn , enhanced release of dense-core vesicles can be abrogated without concomitant correction of ciliary structural abnormalities of bbs mutants . As various organism-wide phenotypes are controlled by neuroendocrine signals , we asked whether some of the phenotypes seen in C . elegans bbs mutant might be due to enhanced neuroendocrine signaling in these mutants . Human BBS patients are short statured , bbs mutant mice are born small [13] , [14] , [40] , and bbs mutant C . elegans have a reduced body size ( Figure 5 ) , suggesting that size regulation might be a conserved function of the BBSome . We found that tom-1 mutants , similar to BBSome-deficient animals , exhibited reduced adult body size ( Figure 5B ) , suggesting that hypersecretion without ciliary defects can cause small body size . Considering that some IFT mutants also exhibit small body sizes ( Figure 5D ) [41] , we sought to distinguish whether the small body size of bbs mutants could be attributed to either hypersecretion or ciliary defects . To do so , we suppressed the enhanced secretion of BBSome mutants with mutations in the dense-core vesicle exocytosis pathway ( Figure 5B ) . Loss of rbf-1 , which suppressed the enhanced dense-core vesicle release of bbs mutants , fully restored normal body size to bbs-7 mutants ( Figure 5B , D ) . By contrast , rbf-1 mutation failed to change the small body of IFT mutants , suggesting that IFT and bbs mutants regulate size through different pathways ( Figure 5D ) . Similarly , loss of unc-31 and aex-6 , but not that of rab-3 , conferred nearly wild-type body size to bbs-7 mutants ( Figure 5B ) . Importantly , in all of these cases normalizations of body sizes were achieved without functional correction of ciliary defects , as the double mutants remained dye-filling defective ( Figure S1C , D , unpublished data ) . Furthermore , tom-1 mutants that exhibited an additive secretion phenotype with BBSome mutants caused a further reduction in the size of bbs mutant ( Figure 5B ) . Together , these findings suggest that the small size of BBSome mutants is due to enhanced secretion of dense-core vesicle cargoes and distinct from IFT mutants . In further support of the notion that distinct mechanisms underlie the small sizes of BBSome and IFT mutants , we also found that neurons that function in size regulation in bbs mutants are distinct from the neurons that function in size regulation in IFT mutants . Normal body size can be restored to che-2 IFT mutants by expressing wild-type che-2 cDNA in a subset of ciliated neurons using the tax-4 promoter [42] . By contrast , expression of functional bbs cDNAs using a tax-4 promoter failed to alter the small body size of bbs mutants ( Figure 5C and Table S1 ) . The body-size phenotype of bbs mutants could , however , be reverted to wild-type levels when bbs cDNAs were expressed using an ocr-2 promoter , which directs expression to a distinct subset of neurons than those expressing tax-4 . This result indicates the spatial requirements of the IFT machinery and the BBSome can be attributed to distinct , non-overlapping neurons in the regulation of body size ( Figure 5C , see Table S1 for expression pattern ) . Thus , despite phenotypic similarities , the size phenotypes of BBSome and IFT mutants are based on distinct cellular and molecular mechanisms . As in the case of small body size , we found that a metabolic phenotype exhibited by bbs mutants is mimicked by the hyperactive secretion mutant tom-1 but not by IFT mutants . Specifically , bbs mutants were previously reported to exhibit increased accumulation of Nile Red and bodipy-labeled fatty acids , vital dyes used as proxies for assessment of metabolic state in intact animals [17] , [38] , [43]–[45] . As in body size , tom-1 mutants exhibited a Nile Red phenotype that was reminiscent of that seen in BBSome mutants , while IFT mutants such as osm-5 and che-11 exhibited wild-type patterns of Nile Red staining ( Figure 6A , C ) . As in body size and release of dense-core vesicles , tom-1 and bbs-7 mutants exhibited additive Nile Red phenotype ( Figure S4A , B ) . Furthermore , mutations in the Rab27 exocytosis pathway , which abrogated the enhanced secretion of bbs mutants , restored wild-type levels of Nile Red staining to bbs-7 mutants ( Figure 6B , D ) , suggesting that enhanced secretion underlies the Nile Red phenotype of bbs mutants . In mice , the obesity seen in the setting of bbs deficiency has been attributed to hyperphagia [13] . C . elegans BBSome mutants similarly showed an altered food intake behavior ( Figure 7 ) . Food intake behavior in C . elegans is assessed by counting the pumping rate of the pharynx , an organ for ingesting bacteria [16] , [46] . Pumping rate is modulated by the availability of food supplies , food quality , and prior feeding experience of the animals [47] . Although under plentiful food conditions , wild type and bbs-deficient C . elegans display similar pumping rate [38] , we found that bbs-deficient animals , unlike wild-type animals , continue to exhibit rapid pharyngeal pumps even when food supplies are exhausted ( Figure 7 ) . This rapid pumping rate is unlikely to be merely a consequence of defective cilia as many other cilia-defective mutants exhibit wild-type pumping rate in the presence or absence of food ( Figures 7E , S4C ) [41] . As in the case of body size and vital dye staining patterns , reducing secretion of dense-core vesicles without concomitant correction of ciliary defects was sufficient to restore wild-type rates of food intake in bbs mutants ( Figure 7A ) . One molecular mechanism known to modulate pumping rate is serotonin signaling [48] . In C . elegans as in mammals , serotonin production is dependent on the tryptophan hydroxylase enzyme , tph-1 . This enzyme is expressed in only a few neurons , only one pair of which , the ADF neurons , is ciliated [48] . To investigate whether abnormal serotonin signaling may underlie the enhanced pumping rate of food-deprived bbs mutants , we expressed functional bbs-1 cDNA using a tph-1 promoter in bbs-1 mutants . bbs-1 reconstitution was sufficient to fully restore wild-type pumping rate to bbs-1 mutants ( Figure 7B , Table S1 ) . Similarly , all other promoters used to reconstitute wild-type bbs-1 that express in ADF neurons fully rescued the feeding phenotype , whereas expression of bbs-1 using promoters that do not target ADF failed to alter the enhanced pumping rate of bbs-1 mutants ( Figure 7B , C , see Table S1 for expression pattern ) . Similar results were obtained when bbs-7 cDNA was expressed in the ADF neurons of bbs-7 mutants ( Table S1 ) . Consistent with the rescue data , loss of tph-1 fully abrogated the enhanced pumping rate of bbs mutants ( Figure 7D ) . These findings suggest that the elevated pumping rate of BBSome mutants is due to excessive serotonin signaling initiated from ADF neurons . Since elevated expression of tph-1 in ADF neurons has been reported in numerous cilia-defective mutants including bbs ( Figure S4C , D ) [39] , we sought to distinguish whether the excess serotonin signaling of bbs-deficient animals could be attributed to excessive production of serotonin or to increased release of this biogenic amine through dense-core vesicles . Despite elevated tph-1 expression , cilia-defective mutants such as che-2 , che-11 and osm-5 exhibited wild-type pumping rates in the presence or absence of food ( Figures 7E , S4C ) , suggesting that increased tph-1 expression does not necessarily correlate with excessive serotonin signaling . In contrast , the hypersecretion mutant tom-1 mimicked the bbs mutant feeding phenotype in that tom-1 mutants exhibited elevated pumping rate when removed from food ( Figure 7E ) . Furthermore , bbs-5 mutants , in which cilia are relatively normal as assessed by dye filling ( Figure S1A ) , exhibited wild-type level of tph-1 expression in ADF neurons yet displayed elevated pumping in the absence of food ( Figure S4C , D ) . Thus , the elevated pumping rate of bbs mutants is likely due to excessive release of serotonin rather than increased synthesis via elevated expression of tph-1 . Taken together , our data indicate that while some ciliary-defective mutants display size , metabolic , and feeding phenotypes that are reminiscent of those seen in the bbs mutants , the underlying molecular and cellular bases of these phenotypes are likely to be distinct . In turn , mutations that cause enhanced secretion of dense-core vesicles without gross structural defects of cilia mimic several physiological phenotypes seen in bbs mutants . Increased levels of circulating leptin and insulin in bbs-deficient mice and humans have been reported [1] , [11] , [14] , [49] . In the case of insulin , the increase is generally assumed to be a secondary consequence of obesity and insulin resistance . However , our data raise the possibility that elevated levels of these circulating hormones may be in part a consequence of enhanced dense-core secretion rather than secondary outcomes of obesity . To further support this possibility , we employed Min6 cells , a mouse pancreatic β-cell line that is ciliated and expresses BBSome components . Treatment of Min6 cells with siRNAs targeting each of three distinct BBSome subunits , Bbs5 , Bbs7 and Bbs9 , resulted in ∼1 . 5–2-fold increase in secreted insulin compared to control siRNA ( Figure 7F ) . Importantly , since these experiments were conducted in cell lines , the enhanced secretion could not be merely attributed to a secondary consequence of organismal obesity . Consistent with this interpretation , excess circulating level of leptin is evident in bbs-deficient mice before the onset of weight gain [11] .
Although disruptions in either the IFT machinery or the BBSome can lead to similar structural and functional defects , here we show that they elicit opposite effects on dense-core vesicle secretion . Specifically , defects in the BBSome cause elevated secretions of dense-core vesicles , while IFT defects generally cause reduced secretions . Our findings indicate that this excess secretion is largely cell autonomous , is not a consequence of altered development , and is dependent on the C . elegans counterpart of Rab27/rabphillin/CAPS exocytosis machinery but distinct from the Rab8/Rabin8 vesicular transport machinery that helps target membranes and cargoes to the cilium [4] . Therefore , we propose that the role of the BBSome complex in vesicular transport within ciliated cells should be expanded from ciliary functions [4] , [5] , [8] , [10] , [18] , [19] and melanosome movement [4] , [50] to also include dense-core vesicle exocytosis . The current paradigm for understanding the myriad manifestations of bbs deficiency is largely framed in the context of defective ciliary functions arising from a failure to properly sort receptors and other signaling molecules to the cilia [5] , [6] , [8] , [10] , [11] , [18] , [19] . As such , mutations in the IFT machinery and the BBSome are thought to cause similar ciliary defects albeit with different levels of severity . This view emerged , in part , by the observations that mutations in IFT components result in organism-wide phenotypes that resemble those of BBSome deficiency [12] . Our findings here indicate that while bbs deficient animals share some of the defects of IFT mutants , the consequences of these mutations on dense-core vesicle secretion and resultant phenotypes are dramatically different from one another . More importantly , our findings challenge the view that similar mechanisms underlie all of the phenotypic similarities caused by BBSome and IFT mutants . For instance , while mutants in both IFT and BBSome components share defects in body size , proper body size regulation requires these components in distinct subsets of sensory neurons . Moreover , the small body sizes of bbs mutants could be reverted to wild type upon abrogation of enhanced secretion of dense-core vesicles , while similar manipulations had no effects on the small body sizes of IFT mutant . Similarly , it has been reported that BBSome and IFT mutants displayed distinct phenotypes of a specific learning behavior in C . elegans [51]; our findings suggested that this difference is likely to reflect a role of dense-core vesicle release in this behavior . Finally , consistent with the notion that some of the phenotypic consequences of bbs deficiency are due to hypersecretion of dense-core vesicles , we found that tom-1 mutants , which cause hypersecretion without obvious ciliary defects , exhibit the size , metabolic , and feeding abnormalities similar to bbs mutants . We do not currently know the precise mechanisms through which bbs deficiencies cause hypersecretion of dense-core vesicles . Although the enhanced secretion phenotype of bbs mutants can be manipulated independently of their ciliary defects , we found that wild type IFT activity is required for manifestation of the hypersecretion phenotype . These findings are consistent with several models . In one model , bbs mutants may mislocalize or excessively accumulate receptors that modulate dense-core vesicle release within the cilia [9] . Given that cilia bound vesicles and dense-core vesicles both arise at the trans-Golgi membrane , it is possible that as a coat-protein , the BBSome not only sorts the appropriate cilia-localized proteins but also prevents the inappropriate accumulation of other molecules in this organelle . Consistent with this scenario , we found that accumulation of IDA-1-GFP , a marker of dense-core vesicles normally excluded from the cilia , accumulates in cilia of bbs mutants ( Figure S5 ) . An alternative but not mutually exclusive possibility is that the BBSome may regulate dense-core vesicle release as a separate function from its role in sorting membranes and cargoes to the cilium . Given that the BBSome regulates cilia-bound vesicles through binding Rabin8 [4] , it is possible that the BBSome regulates the Rab27/rabphillin/CAPS exocytosis pathway through binding specific regulatory components of this machinery such as AEX-3 , the RabGEF for RAB-27/AEX-6 . The idea that the transport machinery associated with cilia formation and function may not be restricted to these organelles and may have evolved from existing cellular machinery has been suggested by the recent demonstration that IFT proteins also play a role in polarized receptor trafficking to a cellular region reminiscent of the cilia in non-ciliated cells [52] . Additionally , the BBSome mediates the movement of melanosomes on microtubules tracks , an organelle not directly associated with the cilia , suggesting that the BBSome might have cilia-independent functions [50] . The genetic epistasis studies presented here do not distinguish between these possibilities . Whether the effects of bbs deficiency on dense-core vesicle secretion are cilia-independent functions of the BBSome or secondary consequences of ciliary abnormalities , our findings challenge the existing paradigm that phenotypic manifestations of the bbs deficiency are recapitulated by IFT mutations . The most salient insight to emerge from our findings is that several of the organismal phenotypes seen in bbs mutants are shared with hypersecretion tom-1 mutants and can be reverted to wild-type by abrogating the hypersecretion of these mutants without concomitant correction of ciliary defects . Although a number of defects seen in bbs patients are likely to be consequences of mis-localized ciliary proteins , we believe that the etiologies of some features of BBS merit re-evaluation in light of the role of this complex in dense-core vesicle secretions . While the specific hormonal pathways that underlie the behavioral and physiological abnormalities of C . elegans may or may not play similar roles in mammals , excessive hormonal secretions are likely to contribute to disease manifestations in BBS patients . For instance , obesity and hyperphagia of BBS patients may be a direct consequence of or exacerbated by increased secretion of appetite-promoting hormones , and hypertension in these patients could be caused by increased release of catecholamines . The finding that hypersecretion in bbs deficiency can be abrogated without correcting concomitant ciliary defects may open the door to new strategies for treating patients with Bardet-Biedl syndrome .
Strains were constructed by standard C . elegans methods [53] . Strains obtained from the C . elegans Genetics Center or the National Bioresourse Project were backcrossed at least 4× with wild type before phenotypic assessment . Transgenic animals were generated by injecting the plasmid of interest at 100 ng/µl with 20–50 ng/µl of Punc-122::GFP or Pmyo-3::GFP as co-injection marker . Plasmids were construct by Gateway cloning as described in [21] . Coelomocyte uptake assays were performed as described in [21] with the exception that 20–40 animals were imaged for each strain at sub-saturating exposure with the shutter half open to split fluorescence between the eyepiece and the camera . ∼30 sychronized adults for Ptph-1::GFP and L4 of Psrh-220::GFP expressing animals were imaged at 40× magnification on a Zeiss Axioplane microscope at a sub-saturating exposure . The neuron of interest was circled and GFP fluorescent within that neuron quantified with OpenLab software . The minimum fluorescence is subtracted from the mean for each cell . The mean and standard error of the mean was determined for each strain and and normalized to wild type . Synchronized mid L4 animals were heat shocked on plates at 37°C for an hour and allowed to recover at 20°C for 4 h before phenotypic assessment . DIC images of 20–40 animals were taken at 5× magnification and the outline of each animal was traced by OpenLab software . The mean and standard error of the perimeter was calculated for each strain and normalized to wild type . ∼200 synchronized L1 animals for each strain were plated onto five plates and incubated at 25°C for 2 d before the percentage of dauer animals was determined along with standard error of the mean . The assay was preformed as described in [28] on six plates per strain . The percentage of animals feeding in groups was determined along with standard error of the mean . ∼300 synchronized L4 worms were washed into 1 . 5 ml eppendorf with S-basal and incubated with 1 µl of 10 mg/µl of DiI in 1 ml of S-basal for ∼2 h rotating at room temp . Worms were than pelleted and washed once with S-basal and plated onto an OP50 seeded plate for ∼1 . 5 h before imaging at 16× on a Zeiss Axioplan microscope . Pumping rate was measured essentially as described in [16] with the exception that pumping rates were determined in 10-s intervals for 10 animals for each genotype . The average pumping rate and standard error of the mean were determined and normalized to wild type . ∼200 synchronized L1 animals were plated onto a 6 cm plate containing 0 . 05 µg/ml of Nile Red and incubated at 20°C for 3 d . Gravid animals were imaged using a TRITC filter at 16× magnification at a sub-satrating exposure with the shutter half open to split fluorescence between eyepiece and the camera on a Zeiss Axioplan microscope . The first pair of intestinal cells was traced using the DIC image in OpenLab and the mean Nile Red fluorescent minus the background was determined for 10–20 animals for each genotype . The average fluorescent along with the standard error of the mean was determined for each strain and normalized to wild type . Min6 cells were treated with siRNAs from Qiagen with Hiperfect ( Qiagen ) for 4 d in triplicates . Cells were rested in Krebs-Ringer bicarbonate buffer for 2 h and then allowed to release insulin into the media for an hour . Insulin was measured by an ELISA kit ( Mercodia ) in duplicates and normalized to the total insulin obtained after cell lysis . The mean and standard error was calculated and expressed as a percentage of the control siRNA . ∼10 , 000 synchronized L1 for each genotype were grown to L4 at 20°C . RNA extraction , cDNA prepartion , and RT-PCR were preformed as described in [54] with primers and syber green PCR mix from Qiagen . The data were standardized to the actin gene , act-1 . Data from two independent growths were aggragated for statistical analysis . All statistical analyses were performed using a two-tailed student t test . | Bardet-Biedl syndrome , BBS , is a rare human genetic disease caused by mutations in many genes . The BBS phenotype is very complex; it is principally characterized by early-onset obesity , progressive blindness , extra digits on the hands and feet , and renal problems . BBS patients may also suffer from developmental delay , learning disabilities , diabetes , and loss of the sense of smell . This complexity suggests that BBS proteins function in a variety of tissues , causing defects in many organs . A unifying theme for the diverse features of BBS emerged when BBS genes were identified and their protein products were found to function in the cilium , a sensory structure found in many cell types . Since then , the various manifestations of BBS have been attributed to the loss of ciliary function in the corresponding tissues . This notion was also supported by the finding that mutations in several genes required for proper cilia formation and function reproduce some of the features seen in BBS patients . Here , we have further investigated the defects found in Caenorhabditis elegans strains carrying mutations in BBS genes ( bbs mutants ) . We find that not only do they display sensory deficits associated with loss of ciliary function , but they also exhibit increased release of multiple peptide and biogenic amine hormones contained in dense-core vesicles of ciliated sensory neurons . Importantly , limiting this excessive hormonal release without correcting the ciliary defects of bbs mutants was sufficient to restore normal body size , feeding , and metabolism to these mutants . Moreover , we show that although non-bbs ciliary mutations can mimic some of the phenotypes of bbs mutants , these effects can be attributed to distinct spatial and molecular mechanisms . Our findings indicate that C . elegans bbs mutants exhibit features of both ciliary and endocrine defects and suggest that some of the clinical manifestations of human BBS may result from excessive endocrine activity , independently of the loss of ciliary function . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"behavioral",
"neuroscience",
"model",
"organisms",
"genetics",
"biology",
"genetics",
"of",
"disease",
"neuroscience",
"genetics",
"and",
"genomics"
] | 2011 | Hyperactive Neuroendocrine Secretion Causes Size, Feeding, and Metabolic Defects of C. elegans Bardet-Biedl Syndrome Mutants |
To faithfully encode mechanosensory information , auditory/vestibular hair cells utilize graded synaptic vesicle ( SV ) release at specialized ribbon synapses . The molecular basis of SV release and consequent recycling of membrane in hair cells has not been fully explored . Here , we report that comet , a gene identified in an ENU mutagenesis screen for zebrafish larvae with vestibular defects , encodes the lipid phosphatase Synaptojanin 1 ( Synj1 ) . Examination of mutant synj1 hair cells revealed basal blebbing near ribbons that was dependent on Cav1 . 3 calcium channel activity but not mechanotransduction . Synaptojanin has been previously implicated in SV recycling; therefore , we tested synaptic transmission at hair-cell synapses . Recordings of post-synaptic activity in synj1 mutants showed relatively normal spike rates when hair cells were mechanically stimulated for a short period of time at 20 Hz . In contrast , a sharp decline in the rate of firing occurred during prolonged stimulation at 20 Hz or stimulation at a higher frequency of 60 Hz . The decline in spike rate suggested that fewer vesicles were available for release . Consistent with this result , we observed that stimulated mutant hair cells had decreased numbers of tethered and reserve-pool vesicles in comparison to wild-type hair cells . Furthermore , stimulation at 60 Hz impaired phase locking of the postsynaptic activity to the mechanical stimulus . Following prolonged stimulation at 60 Hz , we also found that mutant synj1 hair cells displayed a striking delay in the recovery of spontaneous activity . Collectively , the data suggest that Synj1 is critical for retrieval of membrane in order to maintain the quantity , timing of fusion , and spontaneous release properties of SVs at hair-cell ribbon synapses .
Vertebrate hair cells generate graded receptor potentials in response to mechanical stimuli , resulting in neurotransmitter release with high temporal accuracy and fidelity . Precise timing of release supports the phase locking of post-synaptic activity to stimuli in the kilohertz range [1] . These high rates of transmitter release require the specialized ribbon synapse [2]–[4] . For example , in hair cells of the mouse inner ear , ribbon synapses support remarkable rates of exocytosis approaching 1000 SV/sec [5] , [6] . Such high rates are probably achieved by ribbon-mediated rapid exocytosis of a large pool of readily releasable vesicles [3] , [7]–[10] . The role of the ribbon in orchestrating timed SV release is not fully characterized . Perhaps specialized , ribbon-associated proteins are required for accurate timing of release . Additionally , the availability of vesicles at hair-cell ribbons may be necessary and sufficient for both release rate and timing . A recent modeling study of ribbon transmission depicts how increased numbers of docked , releasable vesicles reduces both the latency of release and the associated temporal jitter at ribbon synapses [11] . To maintain a steady supply of SVs at the ribbon , hair cells maintain a large reserve pool of vesicles . Thus , thousands of SVs can be released in response to strong stimulation with membrane turnover approaching a quantity of half the cell surface area within seconds [12] , [13] . It has been shown that disruption in vesicle recycling results in a reduced transmission at neuromuscular junctions [14] . An effect on temporal precision of release in hair-cell synapses has not been investigated in vivo . A key player in vesicle recycling is synaptojanin1 ( Synj1; reviewed by [15] ) . Synj1 is a multidomain lipid-phosphatase that contains a sac1-like phosphatase domain , a 5-phosphatase domain , and a proline-rich domain ( PRD ) . Synj1 regulates phosphoinositide turnover , especially that of phosphatidylinositol ( 4 , 5 ) bisphosphate ( PI ( 4 , 5 ) P2 ) , resulting in the shedding of the clathrin coat on internalized vesicles [16] , [17] . In zebrafish , synj1-deficient nrc ( no optokinetic response ‘c’ ) mutants have visual defects and floating synaptic ribbons in photoreceptors , but not bipolar cells [18] , [19] . Van Epps et al . noted an affect on balance , but did not examine the expression and role of synj1 in hair cells [19] . We report here that the comet gene , which was isolated during a large-scale mutagenesis screen , encodes synj1 . However , mutant zebrafish larvae display balance defects and a corresponding impaired vestibulo-ocular reflex . Given the previously described role of synj1 in SV recycling , we examined synaptic transmission and vesicle pools in mutant hair cells . We found a decrease in spike rate along with a decrease in number of ribbon-associated SVs . Our recordings also revealed an impairment of phase locking in synj1 mutants and a striking delay in the return of spontaneous release of SVs following periods of prolonged stimulation in mutants . Taken together , these results suggest that active maintenance of vesicle pools is required for robust and accurate timing of release of SVs in hair cells .
Three comet mutants were isolated in a large-scale chemical mutagenesis screen for larvae with balance defects . To characterize the mutant comet phenotype at the molecular level , we mapped the comet gene and noted that the recessive lesions mapped closely to a previously identified locus of the nrc gene [19] . The nrc mutant has a nonsense mutation in synj1 , leading to visual defects . We confirmed the identity of comet as synj1 by complementation analysis with nrc heterozygous fish . Sequencing revealed that two alleles , synj1Q296X and synj1W943X , contain nonsense mutations that lead to truncations of the protein product within the first and after the second phosphatase domain , respectively ( Figure 1A ) . The genomic lesion of the third allele , synj1c . 188+2T>A , inactivates the donor splice site of exon 2 that contains the translation start site ( ATG ) , thereby leading to the deletion of exon 2 and presumably no protein product ( Figure 1A ) . The synj1Q296X allele causes a truncation within the SAC1 domain similar to the original nrc allele ( R499X ) , which results in undetectable levels of Synj1 protein in larval brain extracts [19] . Because differences in behavior , cellular or physiological phenotypes could not be detected among larvae carrying the three alleles , we conducted our experiments with synj1Q296X mutants . Next we investigated synj1 expression in the larval auditory/vestibular system . In 96–120 hours post fertilization ( hpf ) zebrafish larvae , we detected synj1 mRNA expression in the central nervous system: foremost in the brain , retina , and in the ear ( Figure 1B , upper panel ) . Sections of the developing ear showed expression of synj1 mRNA in hair cells ( Figure 1B , lower left panel ) . In addition to visualizing synj1 expression by in situ hybridization , we injected a plasmid with a 6 . 5 kb fragment of the synj1 promoter driving EGFP and observed transient EGFP expression in the nervous system and in hair cells of both the ear and lateral line ( data not shown ) . Our results confirm a previous report of expression of synj1 in zebrafish hair cells using microarray analysis [20] . Using RT-PCR on tissue isolated from neuromasts , we detected the synj1-145 but not the synj1-170 isoform , suggesting that the short splice variant is specific to hair cells ( Figure 1C ) . As positive controls , PCR reactions using random decamer mRNA from whole larvae showed robust amplification with each primer set ( data not shown ) . In addition to in situ hybridization and RT-PCR of isolated neuromasts , we attempted to label Synj1 with antibodies , but were not successful with either published [19] or newly generated antibodies . The characteristic tilting posture displayed by synj1 mutant larvae is indicative of vestibular dysfunction . When mutant larvae were continuously swirled in a Petri dish of water they adopted more severely affected postures , including floating upside down . To determine the extent to which vestibular function was affected in synj1Q296X mutants , we tested the vestibulo-ocular reflex ( VOR ) in 120 hpf larvae . The VOR is an involuntary compensatory eye movement in response to stimulation of the vestibular system . The VOR in response to rotation of the head was significantly reduced in synj1Q296X mutants ( Figure 2A ) . We observed a 48% reduction of mean relative eye movement ( quantified by the total power at 0 . 2 Hz ) in mutants compared to wild-type siblings ( Figure 2B ) . Although the response is clearly present , these data suggest that transmission of vestibular information in synj1 mutants is impaired . Unlike mutant synj1 photoreceptor ribbons , the ribbons in synj1Q296X hair cells are localized to the plasma membrane ( Figure S1 and Figure 3H ) . However , we detected the presence of large basal membrane protrusions or blebs in approximately a third of mutant hair cells ( 34 . 5±0 . 9% , n = 388 cells; Figure 3 ) . Large blebs emanating from mutant hair cells could be visualized in live , intact fish using the vital dye , FM1-43 ( Figure 3B , arrow ) . Blebbing was also observed in inner ear or neuromast mutant hair cells transiently expressing soluble GFP ( plasmid myo6b∶GFP; data not shown ) , and in mutant hair cells stably expressing a membrane-targeted form of GFP ( Tg ( brn3c∶mGFP ) ; Figure 3F ) . When counterstained with anti-Ribeye b antibody , we observed that the extrusions of membrane occurred mainly near ribbons ( Figure 3H ) . In addition , we examined the synaptic vesicle marker Vglut3 using immunofluorescence and saw that Vglut3 was present but not noticeably enriched in blebbing membranes ( Figure S2 ) . We also transiently expressed a GFP-tagged pleckstrin homology domain of phospholipase Cδ1 protein that binds to PI ( 4 , 5 ) P2 [21] and did not note any obvious elevation in PI ( 4 , 5 ) P2 in mutant hair cells ( data not shown ) . To determine if the basal blebbing was dependent on synaptic transmission , we generated double mutant larvae that were homozygous for synj1Q296X and cav1 . 3R1250X . Cav1 . 3 L-type calcium channels mediate influx of calcium near hair-cell ribbons , causing the fusion of SVs [22] , [23] . In cav1 . 3aR1250X single mutants , the amount of blebbing was comparable to wild-type levels ( Figure 3I; cav1 . 3R1250X: 5 . 2±0 . 2% , n = 358 cells , 15 larvae; wild-type: 7 . 3±0 . 2% , n = 200 cells , 8 larvae ) . In contrast to synj1Q296X single mutants , we observed low levels of blebbing in cav1 . 3R1250X/synj1Q296X double mutants ( Figure 3I; 6 . 0±0 . 3% , n = 150 cells , 6 larvae ) . Thus , the presence of Cav1 . 3a calcium channels was required for the blebbing phenotype observed in mutant synj1 hair cells . In contrast to cav1 . 3R1250X/synj1Q296X double mutants , blebbing occurred independently of the presence of Cadherin 23 , which is required for mechanotransduction ( 29 . 6±2 . 4% , n = 81 cells , 4 larvae; [24] , [25]; Figure 3I ) . These results suggested that an imbalance of exo- and endocytosis led to blebbing in synj1 mutant hair cells . We investigated the impact of synj1 mutations on hair-cell transmission by stimulating neuromast hair cells and recording the evoked postsynaptic spikes from posterior lateral line ganglion ( PLLg ) cells [26] . Initially , we quantified the spontaneous activity of neuromast hair cells by recording action currents in the absence of water-jet stimulation . Wild-type afferent neurons displayed robust spontaneous activity ( 6 . 5±1 . 0 spikes/sec , n = 7 ) . There was no significant difference in spontaneous activity in synj1Q296X neurons ( 7 . 0±0 . 9 spikes/sec , n = 7 ) . To test whether mutant synapses displayed altered behavior following direct hair-cell activation , we mechanically stimulated neuromast hair cells at 20 Hz . We chose this rate or higher ( see below ) because lateral line hair cells are sensitive to frequencies of 1–150 Hz [27] . Stimulation at 20 Hz resulted in no significant difference in spike rate in the first 60 seconds of stimulation ( mutant: 11 . 4±2 . 4 spikes/sec , n = 6; wild-type: 11 . 9±3 . 4 spikes/sec , n = 4; Figure 4A , B ) . Based on the previously described role of synj1 in neurons , we hypothesized that sustained stimulation would fatigue the hair cell synapse . Indeed , stimulation of synj1Q296X hair cells at 20 Hz for 15 minutes resulted in a significant decrease in the number of spikes in synj1Q296X larvae compared to wild-type larvae ( mutant: 6 . 6±1 . 5 spikes/sec , n = 6; wild-type: 11 . 7±3 . 0 spikes/sec , n = 4; Figure 4B ) To determine whether the reduction in activity seen in synj1Q296X afferent neurons was exacerbated at higher frequency , we increased the stimulation frequency to 60 Hz . Within the first minute of 60 Hz stimulation , mutant spike rate was largely reduced compared to wild-type ( mutant: 10 . 3±2 . 3 spikes/sec , n = 8; wild-type: 19 . 4±3 . 2 spikes/sec , n = 7; Figure 4C , D ) . Next , we attempted to fatigue the synapse with 15 minutes of sustained 60 Hz stimulation . Following 15 minutes of continuous stimulation , the mutant spike rate ( 5 . 1±1 . 4 spikes/sec , n = 8 ) was significantly reduced compared to both the wild-type rate ( 15 . 7±1 . 7 spikes/sec , n = 7 ) and its own initial release rate ( Figure 4D ) . To rule out the possibility that the sustained stimulation affected mechanotransduction , we obtained microphonic potentials before and after 15 minutes of 60-Hz stimulation . We saw no differences between wild-type and mutants in the recordings pre-and post-stimulation ( Figure S3 ) . Given the previously described role for Synj1 in vesicle recycling and the decline of spiking in synj1 mutants , we examined the various membrane components in stimulated wild-type and mutant hair cells using transmission electron microscopy . For these experiments , we stimulated whole larvae for 15 minutes at 60 Hz before fixation ( see Methods ) . To quantify changes , we used single thin sections that featured the center of ribbon and counted structures within a 450 nm radius of the ribbon in neuromast hair cells ( Figure 5 ) . In mutant hair cells , we observed an increase in the number of large coated vesicles approximately 70 nm in diameter , which is larger than a typical SV with a diameter of 40 nm ( synj1Q296X :1 . 6±0 . 2 large vesicles , n = 20 ribbons; wild-type: 0 . 7±0 . 2 large vesicles , n = 14 ribbons; Figure 5C ) . Large coated vesicles were also reported by Lenzi et al . upon maximal stimulation of frog hair cells ( 2002 ) . Small clathrin-coated vesicles ( CCVs; ≥50 nm ) with a diameter similar to SVs were infrequently seen near the ribbon . Previous synj1 mutant studies in the worm and fly neuromuscular junction have suggested that the rundown of synaptic events in these mutants is the result of SV depletion [14] , [28] . We counted the SV pools surrounding hair-cell ribbons on single sections as above . We define the first shell of SVs that decorate the ribbons as ribbon-tethered . With respect to this releasable pool of SVs , we observed a reduction of approximately 30% of ribbon-tethered SVs in mutant neuromast hair cells ( mutant: 1 . 6±0 . 1/100 nm of ribbon perimeter , n = 30 ribbons; wild-type: 2 . 2±0 . 1/100 nm of ribbon perimeter , n = 18 ribbons; Figure 5D ) . We also observed far fewer reserve pool or non-tethered vesicles within 450 nm of ribbons in mutant neuromast hair cells ( mutant: 14 . 5±1 . 6 , n = 13 ribbons; wild-type: 29 . 3±2 . 5 , n = 11 ribbons; Figure 5E ) . With the exception of a two-fold increase of large coated vesicles , we noted only minor differences in membrane compartments of inner ear hair cells ( data not shown ) . The relatively weak effect on inner ear hair cells is likely due to the 60 Hz stimulus being suboptimal for the inner ear maculae . In zebrafish larvae , an acoustic startle reflex can be elicited with frequencies ranging from 100–1200 Hz or potentially higher , with a higher sensitivity at ≥300 Hz [29] . The difference in sensitivity thus predicts the differential effect of the 60 Hz stimulus on inner ear and neuromast hair cells that we observed in our micrographs . We also noted blebbing structures near ribbons in mutant hair cells in our micrographs . Three examples are shown in Figure S2 . Protrusions or the appearance of destabilized membrane can be seen directly adjacent to the ribbon . In order to address whether the disruption in SV recycling affects the timing of hair-cell transmission to PLLg neurons , we examined spike timing in response to stimulation . First , for both wild-type and synj1Q296X larvae , we determined the timing of spikes relative to the start of each 60 Hz cycle for all spikes in the first and fifteenth minute of stimulation . The spike times from all larvae are represented in a histogram in Figure 6A–B . We then calculated the mean spike time for each PLLg recording ( Fig 6C ) . In wild-type larvae , similar to the results for spike rate , there was no change in mean spike time from the first ( 4 . 5±0 . 3 ms , n = 7 ) to the fifteenth minute ( 4 . 7±0 . 2 ms , n = 7 ) of stimulation ( Fig 6C ) . In contrast , the mean spike time was significantly delayed following 15 minutes of sustained stimulation of synj1Q296X hair cells ( 5 . 9±0 . 6 ms , n = 8 ) compared to the first minute ( 4 . 9±0 . 4 ms , n = 8; Figure 6C ) . In addition to the overall delay in synj1 mutant hair-cell response to stimulation , inspection of both the traces in Figure 4C and the histogram of mutant activity from Figure 6B revealed an impairment in the timing of release in response to the stimulus . Proper phase locking of hair-cell output to mechanical stimuli is a reflection of precisely timed transmission . In order to determine the fidelity of timing , we quantified the quality of phase locking of afferent activity by calculating its vector strength ( see Methods ) . A vector strength ( r ) value of 0 implies essentially random activity , whereas a value of 1 describes perfect synchrony between stimulus and response . We determined the vector strength of 60 seconds worth of spikes in the first and fifteenth minute of stimulation in both wild-type and mutant afferent neurons . As expected , stimulus and response were tightly phase-locked in wild-type larvae at 60 Hz ( 1st min: r = 0 . 88±0 . 03; 15th min: r = 0 . 89±0 . 01; n = 7; Figure 7B ) . In contrast , synj1Q296X mutants , similar to the reduction in spike rate seen at 60 Hz , had significantly reduced vector strength in the first minute and a further reduction after fifteen minutes of sustained stimulation ( 1st min: 0 . 75±0 . 05; 15th min: 0 . 61±0 . 08; n = 8; Figure 7B ) . In addition , mutants had reduced vector strength at 20 Hz , but only after 15 minutes of stimulation ( data not shown ) , further supporting the relationship between the reduction in spike rate and the impairment of spike timing ( Figure 4B ) . We define spontaneous activity as random spikes seen in the absence of mechanical stimulation of hair cells . Typically , after a period of direct stimulation , hair cells resume spontaneous output within seconds . We noted in our recordings after ceasing prolonged stimulation that mutants displayed a delay in return of spontaneous activity . To illustrate the phenomenon , we stimulated wild-type and mutant larvae at 60 Hz for 15 minutes , removed the stimulus and continued recording until greater than 300 spontaneous spikes were collected ( Figure 8A ) . Wild-type larvae resumed spontaneous activity within 5±2 seconds ( average of first spike times; n = 4; Figure 8A ) . In contrast , the average time for recovery of first spikes in synj1Q296X mutants was six-fold slower ( 31±17 seconds , n = 4; Figure 8A ) . Comparison of the time elapsed for the first 300 spontaneous spikes to occur in both wild-type and mutant larvae gave an approximate rate of recovery ( Figure 8B ) . Mutants took roughly three-fold longer to generate 300 spikes , with the majority of output delayed greater than two minutes post-stimulation . To further describe the delay in recovery of spontaneous activity , we determined the spike rate in the first 60 seconds post-stimulation in both wild-type and synj1Q296X larvae ( Figure 8C ) . While the wild-type post-stimulation rate had nearly resumed normal levels , the firing rate in synj1Q296X mutants remained near zero ( mutant: 0 . 4±0 . 2 spikes/sec , n = 4; wild-type: 4 . 0±1 . 3 spikes/sec . n = 4 ) .
In this article , we report three new synaptojanin1 mutant alleles in zebrafish: synj1Q296X , synj1W943X , and synj1c . 188+2T>A . The first two alleles generate truncated products; the third allele is a splice-site mutation that eliminates the translation start site . We found that the zebrafish synj1 gene is highly expressed in the nervous system and retina as previously described by Van Epps [19] , but we also observed expression in hair cells of the larval inner ear and lateral line organ . Synj1 has been implicated in both early and later steps in classical clathrin-mediated endocytosis ( CME ) [30] , and one might expect that CME would occur near ribbons where the greater part of exocytosis takes place [31] . Indeed , CME was observed in goldfish retinal bipolar neurons [32] , [33] , but molecular analysis of this type of membrane retrieval has not been examined in hair cells . Alternatively , there is evidence that bulk endocytosis occurs within the vicinity of the hair-cell ribbon . Following sustained depolarizations , large endocytic profiles and cisternae , which are hallmarks of bulk endocytosis , have been observed near hair-cell ribbons [12] . This type of membrane retrieval is especially apparent during prolonged stimulation of hair cells . We find that Synj1 function was most critical during high demands on synaptic transmission as the synj1 mutant phenotype was more severe during sustained stimulation at higher frequencies ( see below ) . At mutant ribbon synapses there was an increase in large coated vesicles and decreased numbers of reserve pool vesicles . Apparently , the disrupted step in membrane recycling in synj1 mutant hair cells was the breakdown of endocytic structures into small synaptic vesicles . Therefore , it remains a possibility that Synj1 is involved in one or several steps of bulk endocytosis . Lesions in synj1 also cause similar defects in central or NMJ synapses of other species . Caenorhabditis elegans ( C . elegans ) , Drosophila melanogaster ( Drosophila ) , and mouse Synj1 mutants have been reported to show slowed endocytosis [14] , [34] , [35] , depletion of SVs , accumulation of clathrin-coated vesicles ( CCVs ) , and aggregation of cortical actin [16] , [28] . A unique feature of the mutant zebrafish phenotype was the presence of basal blebs in hair cells , suggestive of an imbalance of exo- and endocytosis . Indeed , the dependence of the hair-cell phenotype on the presence of Cav1 . 3 channels suggests that exocytosis leads to blebbing near ribbons . Another unique feature was the increase in large coated vesicles , as opposed to small coated vesicles . However , an increase of endosome-like structures is also noted in C . elegans unc-26 synapses [28] . These authors suggest that UNC-26/SYNJ may be involved in converting endosomal compartments into synaptic vesicles . Moreover , a subpopulation of large vesicles is observed among the remaining SVs in Drosophila synj1 mutant NMJ terminals , suggesting a defect in vesicle formation [14] . Collectively , these phenotypes support the notion that Synj1 participates in multiple steps of membrane recycling . The apparent defect in membrane recycling in mutant synj1 hair cells had the highest impact on the reserve pool—the number of vesicles was reduced by 50% . A smaller reserve pool is consistent with an impairment of bulk retrieval [36] . We speculate that the reduction of the reserve pool at neuromast ribbons in turn resulted in fewer tethered SVs . For our TEM experiments , we used a 60 Hz whole-larval stimulation to simulate the conditions of our physiological recordings . Whether this form of stimulation replicates the direct sinusoidal waterjet stimulation remains to be determined . Nevertheless , this rate of stimulation falls into the physiological range of lateral line hair cells [27] . Our experiments measuring the onset of spontaneous activity after prolonged stimulation suggest that spontaneous transmission in synj1Q296X mutants is fatigable and requires several minutes to fully recover . This lengthy delay in the return of spontaneous release may reflect a preference to restock the reserve pool before un-evoked SV release is allowed to resume . Our recordings of evoked action currents in post-synaptic neurons revealed that the mutant hair-cell synaptic properties were relatively normal during short periods of low frequency ( 20 Hz ) stimulation . The requirement for Synj1 became apparent only when high demands on transmission were made . A decrease in the rate of transmission emerged after tens of minutes of stimulation at 20 Hz and occurred without delay at a three-fold higher frequency ( 60 Hz ) . Under these conditions , we saw a 50% or greater reduction in the number of spikes . Furthermore , at the higher frequency of 60 Hz , we observed two additional phenotypes - a delay in the spike timing and an impairment of phase locking to the stimulus . Following prolonged periods of stimulation , wild-type hair cells displayed both tight phase locking and no change in the average spike response time to stimulation . In contrast , a clear delay of spiking occurred in mutant synapses with some spikes delayed by as much as 8 ms within a 60 Hz cycle . When quantified for the degree of synchrony between the response and stimulus , mutants also displayed a significant reduction in the quality of phase locking . The defect in phase locking , combined with the decrease in tethered vesicles and reserve pool vesicles , suggests that refilling of the ribbon or formation of SVs from endocytic compartments was rate limiting in mutant hair cells . Our results support the idea that vesicle numbers are critical for temporal fidelity of SV release at ribbon synapses [11] . Other possibilities may also explain the loss of phase locking , such as an indirect effect on the exocytic machinery . For example , prolonged stimulation could lead to diffusion or mislocalization of active zone proteins such as Cav1 . 3a channels , which mediate the calcium influx required for SV fusion . In our hands , labeling of hair cells stimulated for 15 minutes at 60 Hz with antibodies against Cav1 . 3a and Ribeye did not reveal any differences between wild type and mutant hair-cells ( Sheets , L . and Nicolson , T . , unpublished observations ) . Another possibility is a more subtle disruption of the attachment of the ribbon to the plasma membrane . While a slight displacement of the ribbon would likely delay vesicle fusion , it isn't clear how it would disrupt phase-locking fidelity . Furthermore , ribbon mislocalization was not apparent in our immunofluorescence images or TEM micrographs . Another possibility is that the third or so hair cells with basal blebs near ribbons may account for the loss of phase locking . Perhaps the destabilization of the plasma membrane in these cells leads to abnormal fusion of SVs . A further possibility is the presence of an endocytic defect on the post-synaptic side of mutant ribbon synapses . A recent study describes the failure to endocytose pHluorin-tagged α-amino-3-hydroxy-5-methylisoxazole-4-proprionic acid ( AMPA ) receptors in cultured synaptojanin-KO hippocampal neurons [37] . In this study , an increase of miniature excitatory postsynaptic currents ( mEPSC ) is attributed to an increase in surface exposure of AMPA receptors . An increase in exposed AMPA receptors would likely lead to an increase in spike rates . A recycling defect of AMPA receptors in synj1Q296X PLLg neurons is possible , however , rather than observing increased spontaneous or stimulus-evoked spiking , we saw a decrease in post-synaptic activity . In actively swimming larvae , synj1 mutants suffer from an easily observable balance defect . If swirled continuously in a stream of water , the severity of the vestibular defect increases , suggesting fatigue of synaptic transmission . Such an effect on behavior is consistent with our recordings of afferent neuron activity where a decrease in synaptic ribbon function was seen during sustained stimulation . Similarly , the reduction of a robust response in the VOR test presumably resulted from fewer synaptic transmission events during stimulation of vestibular hair cells . Light tapping on a Petri dish of synj1 mutants will elicit a response , which suggests that their acoustic startle reflex is intact ( data not shown ) . Perhaps continuous stimulation of the auditory system in synj1 mutants would also affect this acoustic startle reflex . Whether hearing is impaired in terms of sensitivity or frequency range in mutant synj1 zebrafish awaits further study . Our results suggest that maintenance of SV pools is critical for reliable and temporally precise transmission at the hair-cell ribbon synapse . Transduction in hair cells induces the release of thousands of vesicles during strong stimulation , which poses a particular challenge to endocytic retrieval mechanisms . Based on our data , we propose that Synj1 plays a critical role in facilitating vesicle recycling and if impaired , vesicle recycling not only affects the number of vesicles released at ribbon synapses , but also the timing of release .
Mutant alleles were maintained in Tübingen or Top Long Fin wild-type backgrounds . The transgenic lines Tg ( brn3c∶mGFP ) and Tg ( neurod∶GFP ) were previously described [26] , [38] . In addition , the gemini/cav1 . 3R1250X ( tc123d allele ) mutant was previously described [22] . The cdh23-MO ( targeted to the ATG start site ) was also previously described [25] . For all experiments , the genotype of larvae was confirmed . Larvae were tested as previously described with some modifications [26] . Zebrafish larvae placed in 2% low-melting agarose dorsal side up on a cover glass were mounted facedown on a vertical sample platform . The platform was driven by a motor ( BE2310J-NPSN , Parker Hann . Inc . ) , controlled by a servo controller ( GV6K-U3E , Parker Hann . Inc . ) , at a frequency of 30°/s and amplitude of ±60° . Movement was controlled by Motion Planner software ( Parker Hann . Inc ) . Eye movements were recorded by digital camera ( DCM130; Hangzhou Scopetek Opto-Electric , Zhejiang , China ) through a 10× objective lens ( Mitutoyo , Neuss , Germany ) , illuminated by infrared ( 800 nm ) , at 7–9 frames per second and at a resolution of 1024×768 dpi ( ScopePhoto ) . In order to read platform position directly from video , a servo controller generated 100 ms pulse , timed 100 ms after the motor reached an edge , briefly shut down the illumination . Video was processed in MATLAB ( Mathworks , Natick , MA ) , using functions in the Image Processing Toolbox . The ratio ( R ) of the long and short axes of the fish eye was extracted from each frame . For quantifying eye movement , the relative ratio change was calculated as ( R – mean ( R ) ) /mean ( R ) . Since the eye movement represented a response to sinusoidal stimulation , the magnitude of eye movement was the power at the stimulation frequency ( determined by discrete Fourier transformation ) . Genomic DNA and mRNA were extracted from day 5 zebrafish larvae by standard methods . First-strand cDNA was generated with the SuperScript III First-Strand Synthesis System ( Invitrogen ) and an oligo d ( T ) primer , and the alleles are designated as follows: synj1c188+2T>A ( formerly called IG459 ) , donor splice site , exon2; synj1Q296X ( formerly called JV039 ) , post sac1-domain truncation; synj1W943X ( formerly called synj1HT039 ) post IPPc domain truncation . Primers used for sequencing were previously described [19] . Neuromast hair-cell cores were obtained using the same glass pipet used for hair-cell stimulation filled with 5 µl of lysis buffer . The pipet was attached via tubing to a hand-held 60 mL syringe . Upon positioning the pipet against a rosette of hair cells , gentle suction was applied while at the same time raising the pipet away from the larva . The entire hair-cell group was usually collected within seconds . On average 5 neuromasts ( ∼50 hair cells ) were pooled , transferred to a PCR tube and then cDNA extraction and amplification was performed . Digoxygenin-labeled riboprobes were generated with the DIG Labeling Kit ( Roche ) from synj1 ( nt 1–692 ) cDNA that had been cloned into pCR-II ( Invitrogen ) . In situ hybridization was conducted as previously described with 200 ng of probe [25] . Larvae at 72 hours post fertilization were imaged on a Zeiss Axio bright-field microscope using a 10× or 20× dry lens objective . Images were acquired via an AxioCam MRc5 color digital camera using Axiovision software and exported as TIFF files to Adobe Photoshop for analysis . Hair cells were labeled with either anti-Ribeye b or anti-Vglut3 antibodies as previously described [26] . Mutant synj1 and wild-type larvae were stimulated on a mini-shaker ( Bruel and Kjaer ) at 0 . 5 V at 60 Hz for 15 minutes [29] . Larvae were fixed immediately in 3% glutaraldehyde and 1 . 5% paraformaldehyde in 0 . 1 M phosphate buffer for ≥12 h , stained with 1% osmium , dehydrated in ethanol and embedded in araldite . Transverse thin sections ( 60–80 nm ) through the ear and neuromasts were imaged on a Phillips CM100 Electron Microscope . Images of ear and neuromast hair-cell ribbons were collected at 25 , 000×–64 , 000× magnification . Images were acquired on an analog camera and negatives were scanned into Adobe Photoshop at 1200 dpi . Sections of stimulated larvae were collected from a total of 5 wild-type and 5 mutant specimens . We generated ≥5 sections per specimen , and each section contained ≥3 hair cells . Our analysis was confined to those ribbons that were directly adjacent to a plasma membrane connected to an afferent bouton . Tethered synaptic vesicles ( 30–50 nm diameter ) were counted as described previously [26] . Same-sized vesicles that were not tethered were counted as reserve pool vesicles . To examine the same cross-sectional area in each section ( cut through the center of ribbons; off-center sections were not counted ) , a circle with a radius of 450 nm was created using ImageJ ( NIH ) and centered on the ribbon body in images at the same magnification ( 64 , 000× ) . We chose 450 nm because longer distances often included large mitochondria or nuclei within the cross-sectional area . The recording setup was similar to that described previously [26] . Briefly , day 5 larvae were anesthetized , mounted and microinjected in the heart with 125 µM α-bungarotoxin to suppress muscle activity . Larvae were then rinsed and maintained in normal extracellular solution ( in mM: 120 NaCl , 2 KCl , 2 CaCl2 , 1 MgCl2 and 10 HEPES , pH 7 . 3 ) . The recording micropipette was also filled with normal extracellular solution . Pipettes were fabricated ( P-97 , Sutter Instruments ) from borosilicate glass with tip diameters of approximately 1 µm and resistances between 15 to 20 MΩ . Signals were collected with an Axopatch 200B , a Digidata 1440A and pClamp 10 software ( Molecular Devices ) . Extracellular currents were collected in voltage clamp mode , filtered at 1 kHz and sampled at 100 µs/pt . The recording electrode and waterjet pipet were positioned with stepper-motor micromanipulators ( Sutter Instruments ) . The waterjet pipet was positioned ∼100 µm from a given neuromast and displacement of the cupula was verified by eye . The recording electrode was positioned within the posterior lateral line ganglion against a cell body of choice . Following observation of spontaneous spiking , the corresponding neuromast was located by detection of phase-locked spiking upon waterjet stimulation . The microphonic recordings were obtained as in [26] . Neuromast hair cell stimulation was performed as described previously [24] . In short , we stimulated hair cells with a pressure clamp ( HSPC-1 , ALA Scientific , New York ) attached to a glass micropipette ( tip diameter ∼30 µM ) filled with normal extracellular solution . In order to elicit activation of hair-cell transduction channels , the waterjet was oriented along the long body axis in order to deflect the cupula laterally . The pressure clamp was driven by a sinusoidal voltage command , which resulted in bidirectional deflection of the cupula . For a given experiment , the voltage command was driven concurrently by pClamp via a second analog output from the Digidata . During each experiment the quality and timing of the waterjet pressure was monitored via a feedback sensor located on the HSPC-1 headstage . For each experiment , this feedback pressure signal was collected in Clampex alongside the extracellular current recording and used for alignment of the apparent stimulus phase to recorded postsynaptic currents . Stimulus trains of 0 , 20 and 60 Hz were delivered in 1-second episodes with minimum time between episodes . Data were analyzed and plotted with Axograph X ( Axograph Scientific ) and Prism 5 ( Graphpad ) software . The number of spikes per second was averaged from the total number of spikes in 60 consecutive episodes . Individual spike time was calculated to correspond with each stimulus sine-wave cycle ( i . e . t = 0 at the start of each sine-wave period ) . This resulted in every spike time occurring within 0 and either 50 ( 20 Hz ) or 16 . 66 ( 60 Hz ) milliseconds . To quantify the synchronization of the hair cell response to stimulus phase , we converted the individual spike times to unit vectors with specific phase angles ( see Figure 7A ) . The equation for vector strength ( r ) was then used as a measure of the degree of phase locking between stimulus and response [39] . Values in the text are expressed as mean±SEM . Statistical significance was determined by using either paired or unpaired Student's t tests , as appropriate , and are indicated in figures as * p<0 . 05 , ** p<0 . 01 , and *** p<0 . 001 . | Ribbon synapses are found in the ear and eye and facilitate the transmission of sensory information to the brain . In hair cells of the ear , the molecules required for ribbon function have not been fully explored . Zebrafish are ideal for investigating molecular components of these specialized synapses because of the ability to study ribbon function using genetic , cellular , and physiological methods . Here , we explore the role of the lipid phosphatase Synaptojanin at the hair cell synapse . Synaptojanin has been previously implicated in synaptic vesicle recycling in conventional synapses , and we also find that the number of synaptic vesicles are reduced in mutant synaptojanin hair cells . Mutant synaptojanin larvae have obvious equilibrium defects , and our electrophysiological recordings revealed that synaptic transmission from hair cells to neurons projecting to the brain is impaired in terms of both rate and accuracy . When stimulated at high frequency or for prolonged periods , mutant synaptojanin hair cells release vesicles out of phase with mechanical stimuli , thus compromising the transfer of sensory information to the brain . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/animal",
"genetics",
"physiology/sensory",
"systems",
"neuroscience/sensory",
"systems"
] | 2009 | synaptojanin1 Is Required for Temporal Fidelity of Synaptic Transmission in Hair Cells |
During ischemic stroke , occlusion of the cerebrovasculature causes neuronal cell death ( infarction ) , but naturally occurring genetic factors modulating infarction have been difficult to identify in human populations . In a surgically induced mouse model of ischemic stroke , we have previously mapped Civq1 to distal chromosome 7 as a quantitative trait locus determining infarct volume . In this study , genome-wide association mapping using 32 inbred mouse strains and an additional linkage scan for infarct volume confirmed that the size of the infarct is determined by ancestral alleles of the causative gene ( s ) . The genetically isolated Civq1 locus in reciprocal recombinant congenic mice refined the critical interval and demonstrated that infarct size is determined by both vascular ( collateral vessel anatomy ) and non-vascular ( neuroprotection ) effects . Through the use of interval-specific SNP haplotype analysis , we further refined the Civq1 locus and identified integrin alpha L ( Itgal ) as one of the causative genes for Civq1 . Itgal is the only gene that exhibits both strain-specific amino acid substitutions and expression differences . Coding SNPs , a 5-bp insertion in exon 30b , and increased mRNA and protein expression of a splice variant of the gene ( Itgal-003 , ENSMUST00000120857 ) , all segregate with infarct volume . Mice lacking Itgal show increased neuronal cell death in both ex vivo brain slice and in vivo focal cerebral ischemia . Our data demonstrate that sequence variation in Itgal modulates ischemic brain injury , and that infarct volume is determined by both vascular and non-vascular mechanisms .
Stroke is the second leading cause of death and the most common cause of acquired adult disability worldwide [1] , [2] . Ischemic stroke is caused by disrupted blood flow within the territory of an occluded blood vessel that results in death of brain cells ( infarct ) . The severity of cerebral infarction primarily depends on the re-perfusion and response of the blood vessels , but neuronal cell death is also determined by intrinsic molecular cascades including excitotoxicity , oxidative stress , apoptosis , and inflammation [3] . More recently , emerging data suggest that the dynamic interaction between vascular cells , glia , neurons and associated tissue matrix proteins – the neurovascular unit – plays a crucial role in the pathogenesis of ischemic brain injury [4] . Although genome-wide association studies have made some progress in the identification of stroke susceptibility genes [5] , the genetic determinants for stroke outcome have yet to be fully explained . Because variation in the anatomic location of the occluded artery , the extent and duration of occlusion , time until treatment , and other contributing factors cannot be controlled in patients , very few genetic factors have been identified that contribute to the severity of brain damage in human ischemic stroke . By contrast , infarct volume in murine models of focal cerebral ischemia ( stroke ) has been shown to vary widely among inbred strains , suggesting strong genetic control [6]–[8] . In a mouse model of ischemic stroke , we have previously demonstrated that infarct volume varies up to 30-fold in 16 common inbred mouse strains . Using a forward genetic mapping analysis in an F2 intercross between C57BL/6J ( B6 hereafter ) and BALB/cByJ ( BALB/c ) strains , we have identified three distinct quantitative trait loci ( QTLs ) that modulate the volume of cerebral infarction . In particular , a single locus mapping to distal chromosome 7 , Civq1 ( cerebral infarct volume QTL1 ) , accounts for the major portion of variation ( 56% ) in infarct volume [8] . In the present study , through the use of multiple QTL mapping analyses , generation of sub-congenic mouse lines , genome-wide association across inbred strains , and ancestral SNP haplotype analyses , we have identified that genetic variation in integrin alpha L ( Itgal ) modulates ischemic brain injury in mice .
Using 16 common inbred mouse strains , we have previously demonstrated that infarct volume after permanent distal middle cerebral artery occlusion ( MCAO ) is under strong genetic control , and had mapped Civq1 on chromosome 7 as a major genetic determinant of infarct volume [8] . To further explore the naturally occurring genetic variation in infarct volume , and provide additional statistical power and map resolution , we determined infarct volumes on additional 16 inbred strains , representing the priority strains for the Mouse Phenome Project ( http://phenome . jax . org ) . We excluded wild-derived strains to avoid spurious false positive association [9]–[11] . The ischemic damage was localized exclusively in the frontal and parietal cortex and infarct volumes were highly reproducible among individual animals of the same inbred strain . Similar to our previous report [8] , we observed large variability in infarct volumes among strains ( Figure 1A , B ) . Strains I/LnJ , LP/J , C3H/HeJ , AKR/J , A/J , BALB/c , SWR/J ( SWR ) , BUB/BnJ , MRL/MpJ , and C58/J exhibited marked sensitivity to ischemic injury and developed large infarct volumes ( >17 . 0 mm3 ) . By contrast , strains C57BLKS/J , FVB/NJ ( FVB ) , C57BR/cdJ , CBA/J , NON/LtJ , DBA/2J , NZL/LtJ , RIIIS/J , 129X1/SvJ , NOD/ShiLtJ , SJL/J , DBA/1J , C57L/J , B6 , and 129S1/SvImJ were relatively resistant to cerebral ischemia , showing infarct volume smaller than 5 mm3 ( Figure 1B ) . Mean infarct volume ranged from 0 . 9 to 43 . 0 mm3 between the strain pairs at the phenotypic extremes ( C57BLKS/J vs . C58/J ) , representing a 47-fold difference in the trait . From these combined data on 32 total strains ( 255 total mice ) , we preformed genome-wide associations for this trait . To correct for population structure and genetic relatedness among inbred strains of mice , we employed the Efficient Mixed Model Association ( EMMA ) [12] . The association scan was carried out using the 4 million high-density SNP panel using the EMMA server ( http://mouse . cs . ucla . edu/emmaserver ) . We identified a significant region of association ( Chr7: 132 . 35–134 . 81 Mb ) that co-localized within the 95% confidence interval of Civq1 that we previously mapped in linkage analyses ( Figure 2A ) [8] . A total of 69 SNPs across the ∼2 . 5 Mb region reached the statistically significant threshold ( P<10−5 ) for cerebral infarct volume and the most significant association without any missing alleles was at a SNP ( rs32965660 , P = 1 . 25×10−6 ) at 132 , 390 , 712 bp on the chromosome . Our EMMA associated region for Civq1 on chromosome 7 covers approximately 2 . 5 Mb ( 132 . 35–134 . 81 Mb ) . A previous report calculated a median distance of 3 Mb between the actual causal variant and the closest marker in EMMA analysis [11] , so we expanded our candidate interval for an additional 1 . 5 Mb region flanking either side of the associated SNPs . There are 124 known genes in the expanded associated region ( 130 . 85–136 . 31 Mb ) on chromosome 7 ( Figure 2A ) . Because of lack of precision in genome-wide association studies due to incomplete understanding of linkage disequilibrium in the mouse genome , statistical power is highly dependent on the number and genetic relatedness of the inbred mouse strains used [13] , [14] . A previous study suggests that for a trait with a genetic effect contributing in the range of 30% to the total variance ( Civq1 = 56% ) , 30 strains or more are required for acceptable power [15] . This suggests that our analysis is sufficiently powered , and that the causative gene for the Civq1 locus is located within the expanded 5 . 5 Mb region , and most likely , in the 2 . 5 Mb region , reduced by our EMMA analysis using 32 inbred strains . Since we previously identified Civq1 in two different genetic crosses ( B6×BALB/c and B6×SWR ) , as well as in the Chromosome Substitution Strain 7 ( CSS7 ) where A/J chromosome 7 was introgressed into the B6 background , and each cross includes B6 as one of the parental strains [8] , it is possible that the sequence variant underlying Civq1 is unique to B6 , occurring in this strain only after it was separated from its last common ancestor with the other strains [16]–[19] . To determine whether allelic variation at Civq1 is unique to the B6 strain or instead due to a sequence variant mapping within an ancestral murine haplotype block [20] , we performed an additional intercross between the large infarct strain , BALB/c , and a different small infarct strain FVB; two strains which exhibit a 10-fold difference in infarct volume ( Figure 1B ) . By substituting FVB for B6 as the “small infarct” strain in this new cross , we could effectively determine whether FVB and B6 share the “protective” allele at Civq1 . Because our goal was to determine whether we would remap Civq1 in this new cross , and to date , Civq1 had shown effect sizes in excess of 50% in other crosses , we surmised that if Civq1 was responsible for the difference between these two parental strains , the locus could be identified with a minimum number of F2 animals . Even with only 35 F2 ( FVB×BALB/c ) mice , we identified a statistically significant locus ( LOD = 5 . 2 ) that mapped to the identical position ( peak LOD at rs13479513 ) on chromosome 7 as that of Civq1 ( Figure 2B ) . Interestingly , the locus identified in this F2 ( FVB× BALB/c ) cross exhibits a large effect size ( ∼85% ) , even stronger than observed in our original two crosses ( 56–57% ) . In this cross , Civq1 accounts for nearly all of the phenotypic difference in infarct volume observed between FVB and BALB/c strains ( Figure S1 ) , and this may explain the highly significant LOD score obtained with only 35 F2 progeny . These combined data further validate the importance of Civq1 in the determination of infarct volume across common inbred mouse strains ( Figure 2C ) . More importantly , these data strongly suggest that the sequence variant underlying Civq1 is located within an ancestral haplotype block that has been inherited across multiple inbred mouse strains , rather than being unique to strain B6 . This further supports the use of ancestral haplotype association mapping approaches to fine map the Civq locus , towards the identification of the causative gene variant ( s ) . To validate the phenotypic effects of the isolated Civq1 locus , and to narrow down the critical region of the QTL , we created recombinant congenic mouse lines ( C . B6-Civq1 ) carrying different segments of the Civq1 region from B6 introgressed into the BALB/c background . Congenic Line 1 contains approximately 22 . 6 Mb of the B6 region of Civq1 , and Line 3 is a fully nested sub-congenic line containing a smaller region completely contained within the larger congenic region ( C . B6-Civq1-1: 119 . 3–141 . 9 Mb and C . B6-Civq1-3: 126 . 2–135 . 8 Mb ) . We also attempted to generate a reciprocal congenic line ( B6 . C-Civq1 ) containing the Civq1 region from BALB/c on the B6 background ( Figure 3A ) , but for unknown reasons , the reciprocal congenic line was embryonic lethal when crossed to homozygosity ( 0 homozygotes out of 52 progeny from a heterozygous congenic cross ) . Nonetheless , since both Civq1 exhibit heterozygous effects in the mapping crosses , we retained the heterozygous congenic line ( B6 . C-Civq1 ( Het ) ) for analysis ( retaining approximately 16 Mb of BALB/c genomic from D7Mit238 to rs32420445 ) . We first analyzed infarct volume of the congenic lines . Both C . B6-Civq1-1 and -3 lines showed significantly reduced ( ∼30% ) volume of cerebral infarction compared to control BALB/c mice ( Figure 3B , C ) . There was no significant difference in infarct volume between the two lines carrying the larger or fully nested , smaller , segment of B6 Civq1 region , providing evidence that the 9 . 6 Mb region ( 126 . 2–135 . 8 Mb ) located between the markers D7Mit238 and rs45999701 defines the critical interval for Civq1 ( Figure 3A ) . In a B6×BALB/c intercross , a locus modulating the number of pial collateral arteries ( Collateral artery number QTL , Canq1 ) [21] was mapped that appears to coincide with our infarct volume locus , Civq1 . This suggested that allelic variation in the same gene ( s ) might modulate the phenotypes of both infarct volume and collateral vessel formation in the brain . We next determined the collateral artery phenotype of these same congenic lines , measuring pial collateral arteries connecting between MCA and ACA ( Figure 3D ) . Consistent with previous reports [21] , BALB/c mice have on average less than one collateral artery per cerebral hemisphere , compared to an average of 10 in B6 mice . Both Lines 1 and 3 of the ( C . B6-Civq1 ) congenic lines showed an approximately 50% increase in the number of collateral arteries when compared to control background BALB/c mice , and similar to the infarct volume data , there was no difference between C . B6-Civq1-1 and -3 lines ( Figure 3E ) . Surprisingly , although the heterozygous reciprocal congenic mice ( B6 . C-Civq1 ( Het ) ) showed no difference in pial collateral number when compared with B6 controls ( Figure 3E ) , the infarct volume of the heterozygous congenic mice was significantly increased when compared to B6 mice ( Figure 3C ) . Infarct volume of the congenic mice ( B6 . C-Civq1 ( Het ) ) was ∼3-fold larger than that of B6 mice ( 14 . 7 mm3 vs . 4 . 5 mm3 ) . To examine whether the Civq1 locus confers a collateral-independent , tissue-intrinsic effect on cerebral infarction , we performed the widely used brain slice-based assay where transient oxygen-glucose deprivation ( OGD ) is used to induce neuronal cell death . Neuronal degeneration was measured via biolistic transfection of the vital fluorescent reporter , YFP , which creates a dispersed ‘sentinel’ population of cortical pyramidal neurons that can be used to quantify neuronal vitality and numbers [22] , [23] . We first examined neuronal cell death for parental B6 and BALB/c strains . Subjecting YFP-transfected coronal brain slices to transient OGD resulted in the degeneration and clearance of a large proportion of cortical pyramidal neurons by 24 hr post OGD treatment . Intriguingly , similar to the sensitivity to focal cerebral ischemia , cell viability in YFP-transfected brain slices in B6 mice was significantly higher than that in BALB/c mice ( Figure 3F , G ) . Based on this finding , we next determined the phenotype of the C . B6-Civq1-3 congenic mice . The congenic mice displayed a significantly increased number of YFP-positive live cortical neurons in OGD-treated brain slices when compared with control BALB/c mice ( 40%; Figure 3F , G ) . There was no difference in YFP transfection efficiency and viability in non OGD-treated brain slices between these mouse strains ( Figure S2 ) . In support of differential resistance to OGD in brain tissues between these strains , western blot analysis showed that the level of cleaved Caspase-3 was significantly reduced in lysates of brain slices from C . B6-Civq1-3 mice compared to control BALB/c mice after OGD treatment ( Figure 3S ) . Taken together , these results show that the B6 allele ( s ) of at least one of the causal gene ( s ) underlying Civq1 provides non-vascular , tissue-intrinsic resistance to ischemic brain injury . More importantly , although the sum of genetic evidence to date suggested that Civq1 , regulating infarct volume , and Canq1 , controlling collateral artery density , are mapped to the identical genomic region , the data from the congenic strain ( B6 . C-Civq1 ( Het ) ) and these ex vivo , OGD experiments using brain slice explants that lack functioning vasculature suggest that the Civq1 locus may be more complex than Canq1 , containing at least one gene variant that modulates ischemic brain injury independent of collateral artery density . The congenic line ( C . B6-Civq1-3 ) reduces the critical QTL interval to a 9 . 6 Mb interval between D7Mit238 at 126 . 2 Mb and rs45999701 at 135 . 8 Mb , but this region still harbors over 200 genes . Although genome-wide association analysis can be employed to significantly reduce a QTL interval for candidate gene identification [24] , the phenotype-associated EMMA interval in this region of chromosome 7 encompasses approximately 2 . 5 Mb , consisting of a genomic region of unusually high gene density , harboring more than 100 potential candidate genes . To further dissect the interval , we compared ancestral SNP haplotype patterns across the inbred mouse lineages , specifically focusing on those strains for which we had generated independent genetic mapping information [25]–[27] . Interval-specific SNP haplotype block analysis can reduce confidence intervals by identifying high-priority regions within a QTL interval that are likely to harbor the causal polymorphism [27] , [28] . Because Civq1 was identified in three different genetic crosses ( B6×BALB/c , B6×SWR , and FVB×BALB/c ) and mapped more broadly to chromosome 7 using the CSS ( B6×A/J ) series , allelic variation at Civq1 is most likely harbored by a gene that maps within an ancestral haplotype block that is shared between BALB/c , A/J , and SWR ( large infarct volumes ) , but that is different from B6 and FVB strains ( small infarct volumes ) . As illustrated in Figure 4 , defining a haplotype block to be three or more adjacent consecutive shared SNP alleles [25] , we identified all SNP haplotype blocks throughout the 3 . 3 Mb critical region of Civq1 ( 132 . 5–135 . 8 Mb ) , a region consistent with each of the 95% confidence intervals of the 4 independent linkage peaks for Civq1 . Only 4 genes ( 4933440M02Rik , Fam57b ( 1500016O10Rik ) , Qprt , and Itgal ) fall within haplotype blocks matching the phenotype pattern of the mapping strains ( Figure 4 ) . To identify the causal gene for Civq1 , we first sought the presence of non-synonymous coding SNPs in these genes . Re-sequencing of the 4 candidate genes identified non-synonymous amino acid substitutions in Qprt and Itgal . Qprt encoding quinolinate phosphoribosyltransferase harbors two coding SNPs ( E205K and D253N ) that segregate with the infarct volume phenotype ( Figure S4 ) . These changes occur at residues that are not well conserved between mammalian species and that are predicted to be functionally benign by in silico amino acid substitution analysis in the three different databases , PolyPhen ( http://genetics . bwh . harvard . edu/pph/ ) , PMut ( http://mmb2 . pcb . ub . es:8080/PMut/ ) , and Panther ( http://www . pantherdb . org/ ) [29] . Itgal ( CD11a ) encodes an α subunit of β2-integrin Lymphocyte Function associated Antigen-1 ( LFA-1 ) that mediates adhesion and migration of leukocytes . The gene harbors two coding SNPs that create W972R and P978L polymorphisms located in the calf-2 extracellular domain of the protein . Strains B6 and FVB that exhibit small infarcts , encode the W972 and P978 isoform , whereas BALB/c , A/J , and SWR strains that exhibit large infarcts , encode the R972 and L978 isoform . Interestingly , despite a lack of conservation at these residues across other species , the W972R change is predicted to be deleterious to the protein in the three in silico databases . No coding changes were identified in the two uncharacterized genes , 4933440M02Rik and Fam57b . Next , to identify genes that show different levels of mRNA between the mapping strains , a surrogate measure of the effects of “regulatory” sequence variants ( broadly defined ) , we performed quantitative real time PCR ( qRT-PCR ) for the 4 candidate genes . In adult brain cortex , only a single gene , Itgal , shows a strain-specific expression difference; an 8-fold higher transcript level in B6 cortex than seen in cortex from the large infarct strains ( BALB/c , CSS7 , and SWR ) ( Figure 5A ) . Since the causative variant for the infarct phenotype would need to reside within the mapped Civq1 interval , we performed allele-specific gene expression analysis to determine whether the observed expression difference is due to cis-acting variation [30] . Similar to the qRT-PCR results , allele-specific gene expression analysis confirmed that the level of B6 Itgal transcript is approximately 6-fold higher than BALB/c transcript in the adult cortex in F1 ( B6×BALB/c ) animals ( Figure 5B ) , providing further evidence that regulatory genetic variation in cis causes the difference in Itgal mRNA abundance in the brain . These differences at the mRNA level were also seen at the protein level , as detected by flow cytometry of CD45-positive cells isolated from adult brain . We found that the level of ITGAL protein is significantly higher in B6 than in BALB/c mice ( Figure 5C ) . Because the Civq1 locus contains at least one gene that modulates infarct volume via effects on collateral artery formation [21] , we also examined the mRNA level for each of the 4 candidate genes in postnatal day 1 ( P1 ) cortex ( Table S1 ) , consistent with the time of development of these vessels [31] . Itgal did not display allele-specific differential gene expression ( Figure S5 ) , consistent with a collateral-independent effect on infarct volume . We also noted that neither Fam57b nor Qprt showed allele-specific expression , and we were unable to detect 4933440M02Rik in either P1 or adult cortices . These other genes are therefore also unlikely to play a role in infarction via effects on collateral vessel anatomy . While performing the complete re-sequencing of all of the coding exons ( including at least 50 bp of flanking intron ) for the 4 candidate genes , we found that the large infarct strains , BALB/c , SWR , and A/J , harbor a complex rearrangement in the distal region of the Itgal gene; a ∼150 bp deletion in intron 29 and multiple insertions and deletions ( indels ) in the coding and 3′-UTR of exon 30b of an alternative splice form of Itgal ( Itgal-003 , ENSMUST00000120857 ) ( Figure 6A , B ) . Further sequencing of cDNA of the Itgal-003 transcript identified a 5-bp insertion in the coding region of exon 30b in BALB/c , SWR , and A/J strains causing a frameshift in the encoded protein , resulting in novel amino acid sequence and a shorter cytoplasmic tail of the protein , as compared to strains B6 and FVB ( Figure 6C ) . We also found that the mRNA level of Itgal-003 markedly differed between the 5 mapping strains in the P1 cortex . The mRNA level of this splice variant was substantially higher ( >11-fold ) in the large infarct strains ( BALB/c , SWR , and CSS7 ) than that of the small infarct strains , B6 and FVB mice ( Figure 6D ) . Using the more accurate allele-specific expression analysis , the level of BALB/c Itgal-003 transcript is approximately 20-fold higher than B6 transcript in P1 cerebral cortex in F1 ( B6×BALB/c ) animals ( Figure 6E ) , indicating that sequence variation in cis leads to increased levels of the Itgal-003 transcript in the BALB/c strain . Since the interplay between neurons , endothelial cells , and glial cells plays a crucial role in the early development of the neurovascular unit , and thus , the pathogenesis of cerebral ischemia [4] , we investigated the mRNA profile of Itgal-003 in both CD146 ( LSEC ) -positive endothelial cells and CD11b-positive brain macrophages isolated from F1 embryos ( E18 . 5 ) . As illustrated in Figure 6E , BALB/c-specific Itgal-003 transcript is expressed 43 times and 25 times higher in endothelial cells and macrophages , respectively , than the B6-specific transcript in the allele-specific expression analysis . Interestingly , the Itgal-003 transcript was not detected in the adult cerebral cortex by RT-PCR , suggesting that this splice variant is acting primarily or exclusively during brain development . To date , 5 Itgal splice isoforms have been identified in mice ( Figure S6 ) , so we determined the relative allele-specific expression of all the other Itgal splice isoforms . No difference was found in P1 and adult cerebral cortices for the other isoforms ( data not shown ) . To further determine whether the strain specific level of Itgal-003 transcript is also reflected at the level of the protein , we generated an Itgal-003-specific antibody against the unique cytoplasmic tail peptide to assay protein expression ( Figure 6C ) . Consistent with the mRNA level , western blot analysis of P1 cerebral cortex demonstrated markedly increased protein level in strain BALB/c compared to strains B6 and FVB ( Figure 6F ) . To determine whether Itgal is involved in ischemic brain injury in vivo , we next examined the phenotype of Itgal knockout mice [32] . A recent study reported that there was no difference in collateral vessel number or in infarct volume between Itgal knockout and control B6 mice [33] . However , despite also finding no difference in number of collateral arteries ( Figure 7A ) , we observed that infarct volume in Itgal knockout mice ( n = 30 ) of B6 background ( 17 generations backcrossed into B6 ) was ∼3-fold larger than that observed in B6 mice ( 15 . 5 mm3 vs . 4 . 7 mm3 ) ( Figure 7B , C ) . We believe that this difference in infarct volume results from the use of different surgical techniques . When performing permanent distal MCAO , care has to be taken to position the occlusion proximal to the lenticulo-striate branches [34] , [35] . If the artery is occluded more distally , smaller and more variable infarcts are produced . In support of this explanation for the difference between the studies , they report a difference in infarct volume of ∼3-fold between B6 and BALB/c strains [33] , whereas we routinely observe that infarct volume in BALB/c mice is ∼7-fold larger than that of B6 mice ( 4 . 7 mm3 vs . 34 . 2 mm3 ) ( Figure 1B ) . Since the Civq1 locus retaining infarct volume phenotype displays both vascular and non-vascular neuroprotective effects on ischemic tissue injury , we hypothesized that the increased infarct volume in Itgal knockout mice might be related to neuroprotection after focal cerebral ischemia . To examine this further , we determine the level of OGD-induced neuronal cell death in brain slices from Itgal knockout mice , again counting YFP-transfected cortical pyramidal neurons in brain slices . As a control , there was no difference in YFP-transfection efficiency and viability in non OGD-treated brain slices between B6 and Itgal knockout mice ( Figure S2 ) . Consistent with the infarct volume data , Itgal knockout mice showed an approximately 50% increase in neuronal cell death after transient OGD , compared to control B6 mice ( Figure 7D , E ) . After OGD treatment , the level of cleaved Caspase-3 was also significantly increased in lysates of brain slices from Itgal knockout mice compared to B6 mice ( Figure S3 ) . To further investigate whether reduced levels of Itgal modulate ischemic brain injury , and particularly whether these effects were collateral vessel-independent , we employed an ex vivo model of cerebral ischemia , using siRNA to knockdown Itgal expression in cortical brain slices ( where collateral circulation is no longer relevant ) . Consistent with the increased neuronal cell death observed in Itgal knockout mice , after transient oxygen deprivation , Itgal siRNA-treated brain slices from B6 mice show markedly increased levels of cleaved Caspase-3 , a marker of neuronal cell death ( Figure 8 ) . These results indicate that Itgal plays a protective role in ischemic brain damage , independent of tissue reperfusion through the collateral vessels . It should be noted that the Itgal knockout mouse line used in our study was generated by replacing the exons 1 and 2 with a Neo-cassette [32] but more recent data shows that there are 5 known alternative splicing transcripts of the gene , including transcripts that do not include these two exons ( Figure S6 ) . Thus , we examined whether this Itgal knockout mouse line generates null alleles for all of the splice variants of the gene . We found that a splicing variant that uses a different initiation site ( Itgal-004 , ENSMUST00000118405 ) was detected in P1 and adult cortices by RT-PCR ( Figure S6 ) , suggesting that the phenotypes we observed in the knockout mice represent only partial knockout of the entire complement of Itgal gene transcripts/functions . Thus , this well-established Itgal knockout line may retain residual or additional isoform protein functions , and in the sense of total Itgal gene function ( s ) , thus represents a hypomorphic allele . The Itgal knockout allele was generated using 129Sv ( Stevens ) genomic DNA and a 129S7 ( 129S7/SvEvBrd-Hprtb-m2 ) ES cell line . Thus , it is formally possible that the some of the differences in infarct volume that are seen in the Itgal knockout were contributed by 129 alleles flanking the deleted Itgal locus . Although the original strains used in the knockout construction are not widely available , we have determined infarct volume and collateral artery number for the related 129S1/SvImJ strain which is derived from the original 129/Sv strain [36] . Both infarct volume and collateral artery number of 129S1/SvImJ mice are not significantly different from those of B6 mice ( Figure S7 ) . Thus , we conclude that the effects seen with the Itgal KO mice are primarily due to the loss of the Itgal gene , and not linked ( 129Sv or 129S7 ) polymorphisms , although we cannot rule out modest bystander effects .
Although several approaches have been proposed to reduce ischemic brain damage , including reperfusion , neuroprotection , and neuronal regeneration , for the vast majority of stroke patients , current therapies are limited . We reasoned that given the multiplicity of mechanisms causing cell death in stroke , approaches that augment endogenous protective pathways might be more likely to lead to success . Thus , we have attempted to identify novel genetic factors modulating stroke outcomes by exploiting naturally occurring endogenous genetic variation determining ischemic brain injury . In crosses between inbred mouse strains that exhibit large differences in infarct volumes , we previously identified a QTL ( Civq1 ) mapping to distal chromosome 7 that determines more than 50% of the variation observed between the strains . In this study , we present evidence that Itgal is one of the genes underlying the complex Civq1 locus . Despite the almost routine detection of QTLs for important disease traits in both rodents and humans , identification of causal genes underlying QTLs remains a major obstacle , in large part due to the large confidence intervals for the typical QTL , often covering hundreds of candidate genes [26] . To narrow the Civq1 interval we employed three different methods that capitalize on the structure of the mouse genome: ( 1 ) generation of interval-specific congenic lines , ( 2 ) genome wide association ( EMMA ) analysis across more than 30 inbred mouse strains , and ( 3 ) interval-specific SNP haplotype analysis using the 5 inbred strains from our experimental crosses . The latter approach was most effective at reducing the list of candidate genes in the Civq1 interval to only 4 genes that clearly fall within shared SNP haplotype blocks . Of these 4 genes , Itgal was the only gene that harbored non-synonymous coding SNPs and exhibited altered mRNA abundance , with both molecular phenotypes co-segregating with the infarct volume phenotypes in the 5 strains used in QTL mapping . We also found that allelic variation in an alternative splicing variant of Itgal ( Itgal-003 ) resulted not only in differential transcript abundance , but also in a truncated cytoplasmic tail of the protein , consisting mostly of a novel amino acid sequence . Itgal encodes the α subunit of LFA-1 ( αLβ2 ) integrin , which is highly expressed in microglia , spleen , bone marrow , and most immune cell populations . The binding of intracellular proteins to the cytoplasmic tail of Itgal is essential to the activation of LFA-1 integrin . The canonical splice isoform of Itgal ( Itgal-002 , ENSMUST00000117762 ) contains this important functional domain , conserved among all integrin family members . Mutation of the cytoplasmic tail of Itgal ( Ital-002 ) has been shown to inhibit its interactions with intracellular proteins , destabilize integrin conformation , and disconnect to cytoskeleton [37] , [38] . By contrast , the function of the unique , truncated sequence of the cytoplasmic tail found in Itgal-003 remains unknown . The increased expression of the Itgal-003 in large infarct strains may interfere with the functions of the well-studied , reference isoform of Itgal , or with other α subunits ( αD , αM , and αX ) that can bind to the β2 subunit , resulting in inhibition of cell adhesion and migration during the development and/or tissue injury . In addition to these differences in the cytoplasmic tail of Itgal-003 , the inbred mouse strains show two amino acid substitutions ( W972R and P978L ) in the calf-2 domain that segregate with the infarct volume phenotypic difference . These two coding changes in ITGAL fall in relatively poorly conserved residues of the protein and the 972R BALB/c allele is shared with other species including cow , sheep , and opossum . However , the lack of conservation at these residues itself does not allow us conclude that these changes are inconsequential because non-synonymous coding SNPs causing risk for complex genetic traits tend not to fall within highly conserved residues/regions of proteins [39] . In point of fact , W972R is predicted to be deleterious by multiple in silico amino acid substitution databases . Although the exact role of the calf-2 domain is not fully understood , a point mutation in the calf-2 domain of αvβ3 integrin in Glazmann thrombasthenia patients disrupts the normal contacts between α and β subunits , resulting in impaired cellular transport from the endoplasmic reticulum [40] . Thus , the two amino acid substitutions in the calf-2 domain of ITGAL may have an effect on α-β formation and stabilization of LFA-1 integrin . Taken together , natural DNA sequence variation in the mapping strains in Itgal generates both qualitative ( a strain-specific splice variant and two amino acid coding changes ) and quantitative ( overall transcript level ) changes in the transcript and encoded protein . We do not know exactly which of these changes is most relevant to the infarct volume phenotype but we surmise that it is likely to be a combination of some or all of these . Importantly , we have shown that a congenic animal that retains all of these strain-specific differences in Itgal variation from B6 shows a similar phenotype to B6 mice , whereas the Itgal gene knockout appears similar to BALB/c strain . In addition , siRNA knockdown of Itgal in an ex vivo cerebral ischemia model using cortical brain slices from B6 mice also increases the extent of neuronal cell death . These combined data suggest that in comparison to the B6 Itgal allele , the net effect of the BALB/c allele is at least a partial loss of function . The identification of Civq1 has raised the question of the role of the causative gene ( s ) in pathophysiology of ischemic stroke . A previous study suggested that the differential ischemic outcomes between inbred mouse strains is related to intrinsic differences in ischemic tolerance or protection pathways in neural tissue [7] . More recently , the extent of pial collateral circulation in the brain has been shown to be inversely correlated with infarct volume data between 15 inbred strains [41] . The authors also identified a QTL ( Canq1 ) for collateral vessel number mapping to the identical genomic position as Civq1 , thereby proposing that the causative gene controlling collateral circulation might also determine the differential infarct volume [21] . Surprisingly , we have observed that the genetically isolated Civq1 locus retaining infarct volume phenotype displays both vascular ( collateral circulation ) and non-vascular ( neuroprotection ) effects in the modulation of ischemic brain damage after MCAO . These observations are in accord with our previous study that this locus displays a non-vascular protective effect on ischemic insult in skeletal muscle [42] . In a mouse model of hind-limb ischemia , we previously mapped a strong genetic locus determining limb necrosis and recovery of perfusion ( Lsq1 ) at the identical genomic position as Civq1 on chromosome 7 [43] . Using an in vitro model of hypoxia and nutrient deprivation , where collateral circulation and indeed all circulation is absent , we have found that isolated primary myocytes from BALB/c are more sensitive to hypoxia and nutrient deprivation than B6 myocytes , recapitulating the strain-specific response to hind-limb ischemia . More importantly , muscle cells from the congenic mice ( C . B6-Civq1-3 ) are protected from this same in vitro hypoxic insult , indicating that the B6 allele ( s ) of the causative gene ( s ) plays an important role in survival of muscle cells , independent of any vascular contribution to ischemia [42] . Therefore , given the physiological similarities between the two ischemic models and the identical map position , we propose that the same causative gene ( s ) underlying Civq1/Lsq1 determines ischemic tissue damage in multiple tissues , possibly through the same physiological mechanism . A number of studies have demonstrated that microglia play an important protective role in ischemic brain damage through microglial migration to the site of injury [44] , [45] , which is controlled by the Itgal protein [46] , [47] . Microglial cells with down-regulated Itgal expression fail to protect neurons after OGD in cultured hippocampal brain slices , suggesting that the migration and adhesion of microglial cells regulated by Itgal is important for the beneficial effect of microglia in stroke [47] . These published data further support Itgal as one of the genes underlying Civq1 . However , Itgal also shows deleterious effects in stroke , as Itgal is also involved in inflammatory injury after cerebral ischemia . After the ischemic insult , the brain is invaded by blood-circulating leukocytes , and LFA-1 ( containing Itgal ) regulates the interaction between circulating leukocytes and endothelial cells . Deficiency of Itgal shows a protective effect on ischemia-reperfusion injury using a transient MCAO model [48] . However , it should be stressed that permanent ( our study ) and transient MCAO [48] models exhibit different pathophysiologies [49] . Large numbers of circulating blood cells enter the brain at time points later than 24 hr after the ischemic insults [50] , [51] , but we measure infarct volume at 24 hr after MCAO . Thus , microglia , expressing Itgal , may play their critical , protective role in the early stages of stroke . Thus far , we have emphasized vascular-independent functions of Itgal in the modulation of infarct volume . But it is clear that the Civq1 locus also contains genetic determinants that modulate infarct volume via changes in collateral vessel anatomy . The identity and nature of these genes and gene variants remains to be determined . Recent studies have demonstrated that microglia regulate vascular anastomosis and increase vascular complexity by assisting endothelial tip cell fusion during brain development [52] . LFA-1 ( containing Itgal ) modulates adhesion of monocyte to collateral endothelium involved in arteriogenesis [53] , [54] . Given the important role of microglia and Itgal in these processes related to collateral vessel development , the question remains then why the Itgal knockout mouse line used in this study did not exhibit a collateral vessel phenotype . Because an engineered knockout allele is rarely equivalent to a naturally-occurring variant allele at a QTL [14] and an Itgal splice isoform was in fact detected in these Itgal “knockout” mice , we cannot exclude the possibility that collateral vessel development and neuroprotection are genetically regulated by different alleles or splice isoforms of Itgal . Alternatively , it is quite likely that more than one gene underlies the complex Civq1 locus . One or more genes may modulate infarct volume by their effect on neuroprotection or inflammatory physiology , and another gene or genes may modulate infarct volume by regulating collateral artery formation . Recent successes in QTL gene identification support this conjecture . Multiple , physically linked , smaller effect genes often contribute to the overall effect of robust , large effect QTLs [14] , [55]–[58] . In this light , we note that a number of gene products mapping within the Civq1 interval have well-defined roles in the immune response ( cytokine receptors , integrins ) , tissue remodeling ( MMP21 , ADAM12 ) , or metabolism ( cytochrome c oxidase ) , each of which could be involved in the overall response to ischemia . Generation of sub-congenic strains that further divide the Civq1 locus may help identify these genes . Although the Itgal haplotype across 32 inbred strains generally correlates with infarct volume , the correlation falls off for several outlier strains . For example , despite that fact that the NOD strain shares the BALB/c haplotype for the Itgal locus , this strain exhibits a small infarct volume and its mRNA expression level is significantly higher than those of the large infarct strains ( Figure S8 ) . Similarly , strain C3H shares the B6 haplotype including a high mRNA expression level , the W972 and P978 SNPs , and lacks the genomic deletion , but the C3H infarct volume is larger than most of the small infarct strains . These data suggests for certain outlier strains , loci other than the otherwise large-effect Civq1 locus can cause profound phenotypic effects . In line with this hypothesis , we have identified a novel genetic locus mapping to mouse chromosome 8 in an intercross between B6 and one of these outlier strains , C3H [59] . Overall , these results are consistent with what we have shown from our previous mapping data , namely , that genetic variation in Itgal is not solely causative for the differential phenotype . Even within the Civq1 locus , there are additional genes for determining infarct volume ( ie , a gene ( s ) regulating collateral vessel density ) . Nonetheless , our data using congenic and Itgal knockout mice show that at least one of the causative gene ( s ) underlying Civq1 functions in the survival of brain tissue independent of a vascular contribution to the ischemic response . Taken together , these data show that the extent of the collateral circulation will not be the sole mechanism underlying Civq1/Lsq1 , and possibly for other loci as well . In summary , by showing evidence of its role in regulation of ischemic brain injury in a mouse model of stroke , we have identified Itgal as one of possibly many quantitative trait genes for Civq1 . Natural DNA sequence variation in Itgal generates qualitative and quantitative change of the transcript and the encoded protein , and a knockout allele , even while retaining one of the newly described splice isoforms , shows a robust effect on infarct volume in vivo and in vitro . Future studies using cell type-specific knockout mice will help dissect cellular and molecular mechanisms of Itgal in both vascular and non-vascular contributions to ischemic brain injury . Ultimately , this work will provide insight into the endogenous protective pathways involved in the pathophysiology of ischemic tissue damage , and in the long-term , provide novel targets for potential therapeutic intervention of ischemic stroke .
All experiments were performed under protocols approved by the Animal Care and Use Committee of Duke University . All inbred strains and Itgal knockout mice ( B6 . 129S7-Itgaltm1Bll/J ) were obtained from the Jackson Laboratory ( Bar Harbor , Me ) either directly or bred locally from breeding pairs of each strain . Mice were age-matched ( 12±1 week ) for all experiments . Focal cerebral ischemia was induced by direct occlusion of the distal MCA as described previously [8] . Briefly , mice were anesthetized with ketamine ( 100 mg/kg ) and xylazine ( 5 mg/kg ) , and the right MCA was exposed by a 0 . 5-cm vertical skin incision midway between the right eye and ear . After the temporalis muscle was split , a 2-mm burr hole was drilled at the junction of the zygomatic arch and the squamous bone . While visualizing with a stereomicroscope , the right MCA was cauterized using an electrocauterizer ( Fine Science Tools ) . The cauterized MCA segment was then transected with micro scissors to verify that the occlusion was complete . The surgical site was closed with 6-0 sterile nylon sutures , and 0 . 25% bupivicaine was applied . Animals were maintained at 37°C during and after surgery until fully recovered from anesthesia , when they were returned to their cages and allowed free access to food and water . All mice were housed in an air-ventilated room with ambient temperature maintained at 25±0 . 5°C . Twenty-four hours after surgery , the animal was euthanized and the brain was removed , chilled at −80°C for 3 min to slightly harden the surface and sliced into 1-mm coronal sections using a brain matrix . Slices were stained with 2% TTC ( 2 , 3 , 5-Triphenyl-tetrazolium chloride ) as previously described [8] . Infarct volumes were calculated by measuring infarct areas on the separate slices , multiplying areas by slice thickness , and summing all slices; this “indirect” morphometric method corrects for edematous swelling . Under ketamine ( 100 mg/kg ) and xylazine ( 5 mg/kg ) , the left ventricle of the heart was cannulated . The right atrium of the heart was incised to allow for venous outflow and the circulation was cleared and maximally dilated with heparin ( 50 µg/ml ) , adenosine ( 1 mg/mL ) and papaverine ( 40 µg/mL ) in phosphate-buffered saline ( PBS ) . Immediately after the PBS infusion , the skull and dura were carefully removed and blue polyurethane ( PU4ii , Vasqtec ) with a viscosity sufficient to restrict capillary transit ( 1∶1 resin∶methylethyl ketone ) was injected . Formalin ( 10% in PBS ) was applied topically to the cortex , and the polyurethane was allowed to cure for 20 minutes . After post-fixation in 10% formalin overnight , the pial circulation was imaged ( Leica MZ16FA ) . Analysis was confined the measurement of collaterals between the MCA and ACA trees [41] . Brain slice isolation and OGD experiments were performed as previously described [22] , [23] . B6 , BALB/c , C . B6-Civq1-3 , and Itgal KO mice were euthanized at postnatal day 10 . Each brain was cut into 250 µm coronal slices on a Vibratome ( Vibratome ) in chilled culture medium containing 15% heat-inactivated horse serum , 10 mM KCl , 10 mM HEPES , 100 U/ml penicillin/streptomycin , 1 mM MEM sodium pyruvate , and 1 mM L-glutamine in Neurobasal A supplemented with NMDA inhibitor ( 1 µM MK-801 ) . Brains were divided into “hemi-coronal” slices . For OGD , slices were suspended at 34°C for 5 . 5 min in glucose-free , N2-bubbled artificial CSF containing 140 mM NaCl , 5 mM KCl , 1 mM CaCl2 , 1 mM MgCl2 , 24 mM D-glucose , and 10 mM HEPES . Control and OGD-treated brain slices were plated onto a 12-well plate with solid culture medium made by the addition of 0 . 5% agarose . After explanting the brain slices , plates were placed for recovery at 37°C for 30 min in a humidified incubator under 5% CO2 . Gold particle-coated plasmids containing Yellow Fluorescent Protein ( YFP ) were introduced into the brain slices by biolistic transfection using a Helios Gene Gun ( Bio-Rad , Herculues ) . Slice cultures were maintained at 37°C for 24 hr in humidified incubator under 5% CO2 . Single nucleotide polymorphism ( SNP ) genotyping was performed using the GoldenGate genome-wide mouse 377 SNP panel ( Illumina ) . Genomic positions of genetic markers ( NBCI Build37/mm9 ) were retrieved from the UCSC genome browser ( http://www . genome . ucsc . edu/ ) and converted to cM using the mouse map converter ( http://cgd . jax . org/mousemapconverter ) . Additional SNP ( rs13479513 , rs13468481 , rs45999701 , rs32420445 , rs37240399 and rs6216320 ) and microsatellite ( D7Mit281 , D7Mit238 , and D7Mit68 ) markers were used for fine-mapping of the locus . The distal boundary of the critical interval ( 9 . 6 Mb ) of C . B6-Civq1-3 was determined by locations of maximal breakpoints for the genotyped markers . The critical interval was homozygous for BALB/c alleles at 126 . 2 Mb ( D7Mit238 ) and at 135 . 82 Mb ( rs45999701 ) . Genome-wide scans were plotted using the J/QTL mapping program , version 1 . 2 . 1 ( http://research . jax . org/faculty/churchill/ ) . Suggestive ( P = 0 . 63 ) and significant ( P = 0 . 05 ) thresholds were established empirically for each phenotypic trait by 1 , 000 permutation tests using all informative markers . The percentage of total trait variance attributable to each locus was determined using the Fit-QTL function provided within the J/QTL software . Genome-wide association mapping for infarct volumes for the 32 strains was performed with EMMA using the UCLA web-based server ( http://mouse . cs . ucla . edu/emmaserver ) [12] . Analysis of individual phenotypic data was performed with SNP panels consisting of 4 million SNPs [11] . We determined confidence intervals by expanding the interval around the peak SNP to include all neighboring SNPs surpassing the significance threshold ( P = 10−5 ) . For single SNP associations , the QTL confidence interval was set at 3 Mb ( 1 . 5 Mb kb on either side of the peak SNP ) . SNP-associated P values were transformed with −log10 ( P value ) for graphing association scores . For the 9 . 6 Mb interval on chromosome 7 , SNP data were obtained from the Mouse Phenome Database ( http://phenome . jax . org/ ) , the Wellcome Trust Sanger Institute Mouse Genome Browser ( http://www . sanger . ac . uk/cgi-bin/modelorgs/mousegenomes/snps . pl ) , and the Center for Genome Dynamics ( http://cgd . jax . org ) . Physical map position was based on the genomic sequence from the NCBI Build 37/mm9 . Haplotype blocks were defined as three or more adjacent informative SNPs shared between the large infarct strains ( BALB/c , A/J and SWR/J ) which differed from the haplotype for the small infarct strains ( B6 and FVB ) [25] . The cytoplasmic tail peptide ( GQRRDIGMDQEERAGPGRL ) of Itgal-003 was synthesized and then used to generate an Itgal-003-specific polyclonal antibody ( Bethyl Laboratories ) . The rabbit antiserum was affinity purified and tested for immunoreactivity for Itgal-003 by immunoblotting . Each E18 . 5 embryonic brain was cut into small pieces , incubated in DMEM containing 10% fetal bovine serum and Collagenase type IV ( 0 . 2 mg/ml , Sigma ) for 30 min at 37°C and then passed through a 19G syringe to obtain a homogeneous cell suspension . Cells were washed with PBS supplemented with 0 . 5 mM EDTA and 0 . 5% BSA and incubated in 10 µl anti-CD131 ( Endothelial cells ) or anti-CD11b ( Macrophages ) antibody-conjugated magnetic beads ( Miltenyi Biotech ) for 15 min on ice . These cells were applied to MACS MS separation columns on magnetic stands and washed with 1 . 5 ml PBS supplemented with 0 . 5 mM EDTA and 0 . 5% BSA . The column was removed from the magnetic stand and magnetically labeled cells were isolated by flushing out fractions . Isolated cells were used to extract total RNA using Trizol ( Invitrogen ) . Quantification of RT-PCR products were measured by examining the increase in fluorescence that was induced by SYBR green binding to dsDNA ( Applied Biosystems ) . The reaction was analyzed on an ABI 7700 Sequence Detection system using the following conditions: 95°C for 10 minutes followed by 40 cycles of 95°C for 15 seconds and 60°C for 1 minute . All samples were run in triplicate and additional assays for endogenous controls ( Gapdh and/or Hprt1 ) were performed to control for input cDNA template quantity . Relative quantification was determined for each sample by calculating the mean Ct value using the 2−ΔΔCt method . Sequence detection software ( SDS version 2 . 1 . 1 , Applied Biosystems ) was used for this analysis . This approach requires at least one SNP in the transcript to distinguish the alleles of the two strains . PCR was performed using an appropriate dilution of cDNA generated from the cerebral cortex of F1 ( B6×BALB ) animals . Amplicons containing coding SNPs were amplified by conventional PCR , and 15 ul of PCR products were treated with 1 U exonuclease I ( New England Biolabs ) and 5 U of Shrimp Alkaline Phosphatase ( SAP ) ( Promega ) . Purified PCR products were used in combination with a conventional primer designed to sit at the nucleotide to the immediate 5′ position of a coding SNPs in the transcript . Cycling conditions were as follows: 40 cycles of 95C° for 10 s , 50C° for 5 s , and 60C° for 30 s . The primer is then labeled with a fluorescently tagged dideoxynucleotide through a single base pair extension ( SNaPshot kit from ABI ) . These products were treated with 1 U of SAP prior to running on an ABI 3130 sequencer , and peak heights were determined using Gene Mapper software ( ABI ) . To determine conditions under which the SNaPshot assays were quantitative , genomic DNA from the F1 animals was amplified and also analyzed by SNaPshot using the same extension . The expression ratio of each transcript allele was normalized to the ratio of the two alleles in the F1 genomic DNA . To reduce Itgal mRNA expression by RNA interference , the Itgal-specific siRNA was purchased from Thermo Scientific and used as recommended . For the ex vivo stroke experiments , siRNA was delivered to the cortical brain slices before the oxygen deprivation . Briefly , after explanting brain slices from B6 mice ( age P10 ) , plates were placed for recovery at 37°C for 30 min in a humidified incubator under 5% CO2 . 5 µM of non-target pooled ( D-001910-10-05 ) or Itgal-specific pooled ( E-046772-00-0005 ) siRNAs were introduced onto the brain slices and the slices were maintained at 37°C in a humidified incubator under 5% CO2 . Forty eight hours after introducing the siRNA , brain slices were deprived of oxygen using glucose-free , N2-bubbled artificial CSF . The slices were then incubated for an additional 24 hr . Mice were perfused with PBS and brains were dissected out and weighed . Brains were teased apart and digested for 1 hour at 37C with 2 mg/mL collagenase A ( Roche ) and 0 . 25 mg/mL DNase I ( Roche ) . Cells were strained and centrifuged in a 30% over 70% Percoll ( Invitrogen ) in PBS density gradient . Cells from the interface were isolated and red blood cells were lysed with ACK lysis buffer . Cells were counted and stained with the following antibodies: CD11a-PE or IgG2a-PE ( eBioscience ) ; CD11c-PE-Cy5 . 5 ( eBioscience ) ; CD45-PE-Cy7 ( eBioscience ) ; Ly6G-AF700 ( eBioscience ) ; IA-IE-Qdot655 ( eBioscience ) ; and CD11b-BV780 ( Biolegend ) . Flow cytometry was run on a LSR-II in the Duke Human Vaccine Institute Flow Research Facility . Analysis was done with FlowJo ( Treestar ) . Following anesthesia , the whole brains were isolated from mice , and were homogenized in cold lysis buffer ( 50 mM Tris/HCl [pH 7 . 4] , 1 mM EGTA , 1 mM DTT ) containing a protease and phosphatase inhibitor cocktail ( Thermo Scientific ) . After low speed centrifugation ( 1 , 000×g , 5 min ) , supernatants ( 30 ug ) were then electrophoresed in a 4–12% polyacrylamide gel and electro-blotted for 2 hr on PVDF ( Polyvinylidine Fluoride ) membranes at room temperature . The blot was incubated with the polyclonal anti-Itgal-003 ( 1∶1 , 000 ) or anti-alpha-Tubulin ( 1∶5 , 000 ) primary antibodies overnight at 4C° . To detect the level of apoptosis , six brain slices from each group were collected after OGD treatment . Anti cleaved-Caspase3 antibody ( 1∶3 , 000 , Cell Signaling Tech ) was applied to detect level of apoptotic cell death . The protein bands were visualized using the chemiluminescence reaction ( ECL detection kit ) . Results were represented as the mean±SEM . Statistical analysis of infarct volume , number of primary branch from MCA as well as collateral vessel number were performed using 1-way ANOVA or non-parametric Kruskal–Wallis test . For qRT-PCR and SNapShot allele-specific expression analyses , student's t-test was used to determine if the fold change was significantly different . | Stroke is the second leading cause of death and the most common cause of acquired adult disability worldwide . Ischemic stroke is caused by an interruption of blood flow in the cerebral arteries and results in neuronal damage ( infarct ) to the area of perfusion in the brain . Although significant progress has been made in the identification of genetic risk factors for stroke susceptibility , identification of genetic factors determining the severity of tissue damage has proven more challenging . By contrast , infarct volume varies widely among laboratory inbred mouse strains . Using a well-established mouse model of stroke and complex genetic analysis , we have exploited these differences and identified Itgal as one of the candidate genes . We further show that allelic variation in Itgal segregates with infarct volume among inbred mouse strains and deficiency of the gene increases ischemic neuronal cell death in stroke . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Natural Genetic Variation of Integrin Alpha L (Itgal) Modulates Ischemic Brain Injury in Stroke |
Peripheral insulin resistance contributes to the development of type 2 diabetes . TCF7L2 has been tightly associated with this disease , although the exact mechanism was largely elusive . Here we propose a novel role of TCF7L2 in hepatic glucose metabolism in mammals . Expression of medium and short isoforms of TCF7L2 was greatly diminished in livers of diet-induced and genetic mouse models of insulin resistance , prompting us to delineate the functional role of these isoforms in hepatic glucose metabolism . Knockdown of hepatic TCF7L2 promoted increased blood glucose levels and glucose intolerance with increased gluconeogenic gene expression in wild-type mice , in accordance with the PCR array data showing that only the gluconeogenic pathway is specifically up-regulated upon depletion of hepatic TCF7L2 . Conversely , overexpression of a nuclear isoform of TCF7L2 in high-fat diet-fed mice ameliorated hyperglycemia with improved glucose tolerance , suggesting a role of this factor in hepatic glucose metabolism . Indeed , we observed a binding of TCF7L2 to promoters of gluconeogenic genes; and expression of TCF7L2 inhibited adjacent promoter occupancies of CREB , CRTC2 , and FoxO1 , critical transcriptional modules in hepatic gluconeogenesis , to disrupt target gene transcription . Finally , haploinsufficiency of TCF7L2 in mice displayed higher glucose levels and impaired glucose tolerance , which were rescued by hepatic expression of a nuclear isoform of TCF7L2 at the physiological level . Collectively , these data suggest a crucial role of TCF7L2 in hepatic glucose metabolism; reduced hepatic expression of nuclear isoforms of this factor might be a critical instigator of hyperglycemia in type 2 diabetes .
Dysregulation of hepatic glucose metabolism is a major predicament for the development of type 2 diabetes . During insulin resistant conditions , physiological activation of Akt-dependent pathway under feeding is impaired , which results in the failure to suppress hepatic glucose production in part via prolonged transcriptional activation of gluconeogenesis [1] , [2] , [3] , [4] . Hepatic gluconeogenic gene expression is mainly controlled by two major transcriptional machineries , namely cAMP response element binding protein ( CREB ) Regulated Transcription Activator 2 ( CRTC2 , also known as TORC2 ) – CREB and Peroxisome Proliferation Activating Receptor Co-activator 1 alpha ( PGC-1α ) – FoxO1 . Under fasting conditions , cAMP-dependent protein kinase ( PKA ) is critical in activating both machineries . PKA-dependent phosphorylation of CREB at Serine 133 promotes the recruitment of CREB binding protein ( CBP ) /p300 [5] , [6] , [7] , [8] , [9] , [10] . Furthermore , PKA-dependent inhibition of AMP activated protein kinase ( AMPK ) and its related kinases ( AMPKRK ) results in the dephosphorylation and nuclear localization of CRTC2 , promoting active complex formation of CRTC2-CREB-CBP/p300 on the promoters of gluconeogenic genes such as phosphoenol pyruvate carboxykinase ( PEPCK ) and glucose 6 phosphatase catalytic subunit ( G6Pase ) [11] , [12] , [13] , [14] , [15] . Similarly , AMPK/AMPKRK-dependent signal activates FoxO1-driven transcription by increasing nuclear retention of this factor via a HDAC-dependent manner [16] . PGC-1α itself is transcriptionally activated by CRTC2-CREB-CBP/p300 , showing that PGC-1α-FoxO1 pathway is also under the control of the cAMP-dependent mechanism [17] , [18] . The role of individual contribution of each factor , however , is currently under the debate . Recent paper by Lu et al . [19] showed the data suggesting that insulin could regulate hepatic gluconeogenic gene expression via FoxO1-independent manner , contesting the current model regarding the critical role of this factor as a regulatory target of insulin signaling pathways in the liver . Similarly , two groups reported the contrasting results using the independent lines of knockout mice for CRTC2 [20] , [21] . These data collectively suggest that disruption of single transcriptional machinery might not be enough to affect hepatic glucose metabolism in vivo , and the transcriptional circuits are indeed tightly interwoven with each other for the fine tuning of glucose homeostasis . First identified as a member of the T-cell factor ( TCF ) family possessing HMG-box-containing DNA-binding domain , TCF7L2 ( also known as TCF4 ) has been known as a nuclear effector of Wnt/β-catenin pathway [22] , [23] , [24] , [25] . Activation of Wnt signaling promotes accumulation and nuclear entry of β-catenin , enabling an association between this factor and TCF7L2 to promote target gene expression . Wnt/β-catenin signaling plays a crucial role in many developmental processes as well as in some adult mammalian tissues that are active in self-renewing processes such as proliferating crypt precursors and differentiated villus cells in the intestinal epithelium , epidermal stem cells in the hair follicle , hematopoietic stem cells , osteoblasts , and several types of cancer cells ( reviewed in [26] , [27] ) . Recent evidences also indicated a role of this pathway in type 2 diabetes . Extensive genome-wide association ( GWA ) studies revealed that TCF7L2 is a strong candidate for a type 2 diabetes gene , and several studies indicated that the presence of certain common single nucleotide polymorphisms ( SNPs ) in this gene might increase the incidence of this disease in human [28] , [29] , [30] , [31] , [32] , [33] , [34] , [35] . Indeed , incretin hormone GLP-1 is induced by TCF7L2 in the intestinal endocrine L cells , and GLP-1-dependent pancreatic beta cell proliferation and insulin secretion also require TCF7L2 , suggesting that alteration in its expression in certain target tissues might display glucose phenotypes in affected individuals [36] , [37] . The functional role of TCF7L2 in hepatic glucose metabolism , however , has not been clearly stated to date . Here we propose that TCF7L2 is critical in mediating transcriptional control of hepatic glucose production . We found that hepatic expression of medium and short isoforms of TCF7L2 was specifically reduced in mouse models of insulin resistance . Acute depletion of TCF7L2 in the liver resulted in higher blood glucose levels that were associated with increased glucose intolerance and up-regulation of gluconeogenic genes , while ectopic expression of nuclear TCF7L2 in C57BL/6 mice with diet-induced obesity ( DIO ) improved glucose tolerance . TCF7L2 was shown to bind to the promoters of PEPCK and G6Pase , thereby interfering with the association of both CRTC2 and FoxO1 on their cognate recognition sites on the chromatin . Furthermore , mice with global haploinsufficiency of TCF7L2 exhibited higher glucose levels and impaired glucose tolerance compared with the littermate control , and adenovirus-mediated two-fold expression of TCF7L2 almost completely reversed the phenotype . Taken together , we suggest that TCF7L2 would be a critical player in regulating glycemia in mammals by modulating hepatic gluconeogenic gene expression .
Although TCF7L2 has been regarded as one of the major candidate genes for inducing type 2 diabetes , the exact role for this factor in hepatic glucose metabolism has not been well documented . To investigate the potential role for TCF7L2 in the liver , we firstly measured the expression level of TCF7L2 in livers of mice with various dietary conditions . Interestingly , overnight fasting or high-fat diet invoked reduced protein levels of only medium and short isoforms of TCF7L2 ( designated as M and S , respectively ) compared with control , while no change was shown in the expression levels of long isoforms ( designated as E ) ( Figure 1A and Figure S1A ) . Furthermore , decreased expression of medium and short isoforms was also pronounced in the livers of db/db mice compared with control , suggesting that hepatic insulin resistance might be correlated with the disappearance of smaller isoforms of TCF7L2 in the liver ( Figure 1A ) . While both medium and short isoforms of TCF7L2 primarily resided in the nucleus , a majority of long isoforms were found in the cytoplasm ( Figure S1B ) . Since the expression of TCF7L2 was up-regulated under feeding , we wanted to further delineate the potential signaling cascades that are involved in this phenomenon . Unlike our expectations , treatment of insulin alone did not provoke changes in expression of TCF7L2 in primary hepatocytes , showing only a slight induction of both mRNA and protein expression with 24 h-treatment ( Figure S1C ) . Addition of forskolin , a cAMP agonist , resulted in the reduction of TCF7L2 expression both at the mRNA and protein levels , suggesting that the disappearance of glucagon/cAMP signaling pathway , rather than the activation of insulin signaling pathway under feeding conditions , might be involved in the regulation of TCF7L2 expression ( Figure S1D ) . To explore the causal role of TCF7L2 in hepatic glucose metabolism , we generated an adenovirus expressing shRNA for TCF7L2 ( Ad-TCF7L2 shRNA ) and injected into the tail vein of C57BL/6 mice . Knockdown of all isoforms of hepatic TCF7L2 resulted in higher glucose levels with a slight increase in plasma insulin levels under both fasting and feeding conditions . No changes were observed in body weight , plasma and liver triacylglycerol ( TG ) levels , as well as plasma non-esterified fatty acid ( NEFA ) levels between mice injected with either Ad-TCF7L2 shRNA or control Ad-US virus , excluding a potential non-specific effect ( Figure 1B , and Figure S2A-S2D ) . Glucose intolerance was observed in TCF7L2-knockdown mice compared with control , suggesting that insulin signaling might be perturbed with acute depletion of TCF7L2 in mice ( Figure S2E ) . Excluding a change in insulin signaling , the rate of insulin-dependent clearance of blood glucose was not different between two groups as evidenced by the insulin tolerance test ( Figure S2F ) . Since TCF7L2 is a transcription factor that could potentially affect glucose metabolism at the transcriptional level , we attempted to measure the relative expression levels of genes involved in glucose and glycogen metabolism between two groups ( control vs . TCF7L2-knockdown ) by PCR array analysis . Interestingly , expression levels of genes that are involved in gluconeogenesis were increased upon TCF7L2 knockdown ( PEPCK , G6Pase , Fructose 1 , 6-bisphosphatase 1 ( Fbp1 ) , and Fructose 1 , 6-bisphosphatase 2 ( Fbp2 ) ) in mouse liver ( Table 1 ) . As well , genes encoding Fumarase ( FH1 ) and Malate dehydrogenase ( Mdh1b ) , two enzymes that are critical in providing malate for gluconeogenesis from the mitochondrial TCA cycle , and pyruvate dehydrogenase kinase 4 ( PDK4 ) , which functions to reduce the formation of acetyl CoA and block the TCA cycle , were also significantly induced with depletion of TCF7L2 in the liver . Indeed , we were able to confirm the significant induction in the expression of gluconeogenic genes in the livers of TCF7L2-knockdown mice compared with that of control by Q-PCR , suggesting that hepatic gluconeogenic potential is specifically enhanced upon depletion of TCF7L2 in the mouse liver ( Figure 1C and 1D ) . As hinted by the result from the insulin tolerance test , knockdown of TCF7L2 did not alter the phosphorylation status of key enzymes in the hepatic insulin signaling ( Figure 1E and Figure S3A ) , suggesting that the changes in the expression level of TCF7L2 per se might not be directly linked to the fluctuation in the insulin signaling pathway in the liver . Similar results were also obtained in primary hepatocytes using Ad-shTCF7L2 , further supporting the direct role of TCF7L2 in the regulation of hepatic gluconeogenic gene expression ( Figure S3B–S3E ) . Depletion of hepatic TCF7L2 promoted higher glucose levels , suggesting that reduced expression of certain isoforms of TCF7L2 under insulin resistance might be in part responsible for the hyperglycemia in that setting . To test this hypothesis , we generated adenoviruses expressing various isoforms of TCF7L2 ( Ad-TCF7L2 M , Ad-TCF7L2 S , and Ad-TCF7L2 E ) , and tested their effects on expression of gluconeogenic genes in primary hepatocytes . TCF7L2 M and S , nuclear isoforms that displayed reduced expression in livers of insulin resistant mice , were more effective in inhibiting expression of gluconeogenic genes than the cytosolic TCF7L2 E , suggesting that the effect of TCF7L2 might occur largely in the nucleus ( Figure S4A ) . We thus chose to utilize adenovirus expressing TCF7L2 M , a widely used isoform for various studies , for our in vivo experiments . Indeed , adenovirus-mediated expression of TCF7L2 M diminished fasting blood glucose levels without changes in body weight in DIO mice ( Figure 2A , 2B , and Figure S4B ) . No changes were observed in plasma TG and NEFA levels between mice injected with either Ad-TCF7L2 M or control Ad-GFP ( Figure 2C ) . Neither insulin tolerance nor plasma insulin levels was changed with expression of TCF7L2 , suggesting that global insulin signaling might not be affected by Ad-TCF7L2 M infection ( Figure S4C and S4D ) . Mice with Ad-TCF7L2 M displayed reduction in gluconeogenic gene expression , showing that indeed TCF7L2 could be linked to the regulation of glucose homeostasis by inhibiting expression of gluconeogenic genes ( Figure 2D ) . On the other hand , glucose tolerance was significantly improved in mice expressing TCF7L2 compared with control , and hepatic insulin signaling appeared to be slightly improved by TCF7L2 overexpression in the liver as evidenced by increased tyrosine phosphorylation of IRβ and serine phosphorylation of AKT , GSK3β , and FoxO1 , presumably due to the secondary effect that was associated with improved glycemia in DIO mice ( Figure 2E and 2F ) . Indeed , we did not observe changes in hepatic insulin signaling with Ad-TCF7L2 infection in lean mice , suggesting that TCF7L2 might not directly regulate insulin signaling in the physiological context ( data not shown ) . Next , we wanted to verify whether TCF7L2 is directly involved in the transcriptional control of gluconeogenic genes . Indeed , we were able to recapitulate the inhibitory effect of TCF7L2 on glucose production in primary hepatocytes without changes in insulin signaling pathways , ruling out the potential involvement of other organs or cell types upon adenoviral delivery in vivo ( Figure S4E and S4F ) . Furthermore , reporter assay revealed that both PEPCK and G6Pase promoter activities were inhibited by ectopic expression of TCF7L2 ( Figure S4G ) , providing an evidence for the involvement of direct binding of TCF7L2 on the promoters of gluconeogenic genes . Careful investigation of promoter sequences revealed the presence of putative TCF binding elements ( TBEs ) that is adjacent to the CREB/CRTC2 binding site ( cAMP response element , CRE ) and the FoxO1 binding site ( insulin response element , IRE ) on both PEPCK and G6Pase promoters ( Figure 3A ) . Consistent with the proposed role of TCF7L2 in inhibiting gluconeogenic gene expression under feeding conditions , we observed the reciprocal and mutually exclusive binding of TCF7L2 or CRTC2/FoxO1 onto the promoters of gluconeogenic genes under fasting and feeding conditions . By chromatin immunoprecipitation ( ChIP ) assay , we detected an increase in occupancy of TCF7L2 and a decrease in occupancy of CRTC2/FoxO1 over PEPCK or G6Pase promoter under feeding , while increased binding of CRTC2/FoxO1 and decreased binding of TCF7L2 onto these promoters were evident under fasting conditions in mouse liver ( Figure 3B ) . We speculated that the reduced expression of TCF7L2 under fasting conditions might in part contribute to the increased occupancy of CREB/CRTC2 or FoxO1 over gluconeogenic promoters . Mutations in TBE site blunted inhibitory effects of TCF7L2 on activity of gluconeogenic promoters in cultured cells ( Figure S5A ) . To further provide the evidence for the importance of the ability of TCF7L2 to bind DNA in inhibiting gluconeogenic gene expression , we generated two types of mutants; TCF7L2 Δβ-catenin mutant , which contains an intact DNA binding motif but lacks a β-catenin interaction domain , and TCF7L2 ΔHMG mutant , which retains a β-catenin interaction domain but lacks a DNA binding motif ( Figure S5B and S5C ) . In line with this result , mutations on DNA binding motif ( ΔHMG ) , but not on the beta-catenin binding motif Δβ-catenin ) , completely impaired the ability of TCF7L2 to inhibit gluconeogenic gene expression ( Figure 3C ) . These data suggest that while binding to β-catenin is dispensable , the ability to bind to the gluconeogenic promoters is essential for the inhibitory function of TCF7L2 . ChIP assay also revealed that ectopic expression of TCF7L2 WT or Δβ-catenin , but not of ΔHMG , inhibited the occupancy of CRTC2 or FoxO1 on the cognate binding sites of the gluconeogenic promoters ( Figure 3D ) . Instead , increased binding of TCF7L2 to the adjacent putative TCF binding element ( TBE ) on the chromatin was observed ( Figure S5D ) , suggesting that TCF7L2 would inhibit transcription of gluconeogenic genes by binding to the promoter and inhibiting the formation of active transcription factor complex in hepatocytes . To further assess the potential involvement of β-catenin , a known co-activator for TCF7L2 , in the TCF7L2-dependent inhibition of gluconeogenic gene expression , we generated adenovirus for β-catenin expression , and tested in primary hepatocytes . We found that overexpression of β-catenin did not promote the inhibitory effect of TCF7L2 on the expression of G6Pase , PGC1α , or Lipin1 , known targets for FoxO1 and CREB/CRTC2 ( Figure 4A–4C ) . Furthermore , knockdown of β-catenin rather reduced the forskolin-induced expression of G6Pase and PEPCK in the absence of TCF7L2 , suggesting that β-catenin and TCF7L2 did not function in concert at least for the regulation of gluconeogenic genes in the liver ( Figure 4D–4F ) . To ascertain whether chronic depletion of TCF7L2 in the liver might play a causal role in the promotion of hyperglycemia , we obtained knockout mice for TCF7L2 gene in C57BL/6 background from Sanger Institute . As in the case of previously generated lines , we were not able to obtain viable TCF7L2 homozygous knockout mice . Thus , we bred heterozygous null mice ( TCF7L2 +/− ) to produce TCF7L2 heterozygous null mice ( TCF7L2 +/− ) and their littermates ( TCF7L2 +/+ ) for the subsequent study ( Figure S6A ) . In accordance with the effect of the acute depletion of TCF7L2 in mice , TCF7L2 +/− mice displayed higher blood glucose levels with no significant changes in plasma insulin levels compared with their littermates under fasting ( Figure 5A , Figure S6B and S6C ) . In addition , TCF7L2+/− mice also displayed pyruvate intolerance that was accompanied with increased hepatic expression of gluconeogenic genes , suggesting that chronic depletion of TCF7L2 might promote increased glucose production from the liver ( Figure 5B and 5C ) . Similar results on blood glucose levels , plasma metabolites levels , and gluconeogenic gene expression were also obtained using TCF7L2 +/− mice under feeding conditions ( Figure 5D and 5E ) . Glucose intolerance was also apparent in TCF7L2 +/− mice compared with control ( Figure S6D , top ) . Excluding a potential involvement of pancreatic beta cells , we were not able to observe a difference in glucose-induced insulin levels between two groups of mice ( Figure S6D , bottom ) . Hepatic glycogen levels were reduced in TCF7L2 +/− mice compared with control , suggesting that glycogen metabolism might be affected by haploinsufficiency of TCF7L2 in mice ( Figure S6E ) . To evaluate the potential changes in whole body insulin sensitivity , we performed hyperinsulinemic-euglycemic clamp studies . Compared with the control , we observed increased glucose production from TCF7L2+/− mice , although the statistical significance was only observed at the basal period ( Figure S6F ) . However , no specific changes were observed in whole body glucose metabolism during the clamp period between TCF7L2 +/+ mice and TCF7L2 −/− mice , even in the presence of mild reduction in body weight and muscle mass upon TCF7L2 haploinsufficiency , suggesting that haploinsufficiency of TCF7L2 might not invoke changes in peripheral insulin signaling pathway at least under the normal chow diet conditions ( Figure S6F and S6G ) . In accordance with this phenomenon , we were not able to observe differences in phosphorylation status of key insulin signaling enzymes in the liver , pancreas , adipose tissues , or skeletal muscle between wild type and TCF7L2+/− mice ( Figure S7A–S7D ) . To analyze the liver-specific effect of chronic depletion of TCF7L2 , we prepared primary hepatocytes from either TCF7L2+/− mice or TCF7L2+/+ mice . Chronic haploinsufficiency of TCF7L2 indeed displayed higher levels of gluconeogenic gene expression and increased glucose production in primary hepatocytes , without impairment of normal insulin signaling ( Figure 6A–6C ) . Similar to the clamp studies in vivo , we were able to observe the increased glucose production from the TCF7L2 +/− hepatocytes compared with control . Again , insulin was able to repress the forskolin-induced glucose production from hepatocytes of both genotypes , showing insulin signaling itself was not perturbed by haploinsufficiency of TCF7L2 . Furthermore , increased occupancy of endogenous CREB , CRTC2 , or FoxO1 , with concomitant decrease in the occupancy of endogenous TCF7L2 , on the gluconeogenic promoter was apparent in TCF7L2+/− hepatocytes compared with control cells ( Figure 6D ) . These data once again suggest that binding of TCF7L2 and CRTC2/FoxO1 on the promoters of gluconeogenic genes might be mutually exclusive , and that the haploinsufficiency of hepatic TCF7L2 is indeed critical in promoting dysregulation of hepatic glucose production . To further ascertain that the effects of TCF7L2 on the hepatic gluconeogenic gene expression function by direct inhibition of CRTC2 and FoxO1 activities , we performed knockdown of both factors in primary hepatocytes from TCF7L2 +/− mice . Increased mRNA levels of PEPCK and G6Pase by haploinsufficiency of TCF7L2 were indeed greatly normalized by knockdown of CRTC2 and FoxO1 , showing that TCF7L2-dependent regulation of hepatic gluconeogenic gene expression directly modulated activities of these transcriptional machineries ( Figure S8A and S8B ) . To further support the hypothesis that impaired glucose metabolism in global haploinsufficiency of TCF7L2 in mice is largely due to the problems in the liver , we used adenovirus expressing TCF7L2 M to restore the expression of TCF7L2 specifically in the liver . We did not detect expression of TCF7L2 M expression in other insulin sensitive tissues such as pancreatic islet , skeletal muscle , or adipose tissues upon adenoviral infection ( data not shown ) . Restoration of TCF7L2 expression in the liver of TCF7L2 +/− mice slightly reduced fasting glucose levels with reduction in expression levels for gluconeogenic genes that were largely comparable with those of wild type mice , without promoting changes in plasma insulin , NEFA , and TG levels ( Figure 7A–7C ) . Glucose intolerance that was associated with global haploinsufficiency of TCF7L2 was almost completely abolished by hepatic expression of TCF7L2 ( Figure 7D and 7E ) . These data collectively suggest that the glucose phenotype that is associated with TCF7L2 +/− mice might be in part due to the dysregulation of glucose metabolism in the liver .
Common SNPs of TCF7L2 such as rs7903146 and rs12255372 are associated with type 2 diabetes . Indeed , several studies indicated that patients carrying these SNPs might have the increased risk for the development of this disease [38] , [39] , [40] , [41] . The observed SNPs , however , are localized in the intronic regions of TCF7L2 gene , and several attempts to correlate the presence of the intronic SNPs with changes in expression of this gene in various tissues such as adipose tissue , skeletal muscle , and pancreatic islets have been largely inconclusive [40] , [42] , [43] , [44] , [45] . Recent study provided the direct evidence against this hypothesis by showing no correlation between type 2 diabetes-associated SNPs and relative expression of this gene in adipose tissue from 159 obese individuals [46] . Rather , they suggested the possibility that tissue-specific expression of specific isoforms might be important for the functional consequences of TCF7L2-dependent signaling . In this study , we have provided the evidence for differential expression of long verses medium or short isoforms of TCF7L2 under the nutritional stress in mouse liver . Under insulin resistance , expression levels of the medium and short isoforms of TCF7L2 , which reside mostly in the nucleus , are specifically reduced while no such change is observed on that of long isoforms of TCF7L2 in mouse liver . The medium and short isoforms of TCF7L2 lack CtBP binding domain as well as auxiliary DNA binding domain termed C-clamp motif , and have shown to bind to the previously defined TBE sequence [47] . Interestingly , we located putative TBEs at or near the cAMP response element ( CRE ) or insulin response element ( IRE ) on the promoters of gluconeogenic genes such as PEPCK and G6Pase ( Figure 3A ) , and found that binding of TCF7L2 inhibited the recruitment of CREB , CRTC2 , or FoxO1 on the promoter under feeding conditions in mouse liver or in hepatocytes ( Figure 3B and Figure 6D ) . TCF7L2 per se might not directly affect insulin signaling in the liver , since we did not observe any changes in phosphorylation status of key enzymes in hepatic insulin signaling upon knockdown or knockout of TCF7L2 , at least under normal chow diet . Rather , we suspected that reduced expression of nuclear TCF7L2 by insulin resistance might be in part responsible for the enhanced hepatic glucose production , providing a potential mechanism for the hyperglycemic phenotype that is induced by DIO or genetic insulin resistance in mammals ( Figure 7F ) . We found that cAMP treatment could reduce expression of TCF7L2 in primary hepatocytes . Interestingly , glucagon/cAMP signaling pathway was known to be induced by insulin resistance in the liver . Further study is necessary to elucidate the potential regulation of TCF7L2 expression or activity by cAMP signaling pathway that is critical in glucose homeostasis in vivo . While we were preparing our manuscript , a new study by Nobrega's group was published suggesting that alterations in TCF7L2 expression would promote changes in glucose metabolism [48] . Surprisingly , they found the seemingly the opposite phenotype on their TCF7L2 null allele compared with our results , in that the TCF7L2 knockout mice displayed hypoglycemia that was associated with reduced plasma insulin levels . As well , systemic overexpression of TCF7L2 rather promoted hyperglycemia in their BAC transgenic models . We suspected the differences between two mouse lines might stem from the fact while we used the C57BL/6 mice for our transient/chronic models , they chose to use CD-1 mice that were rarely utilized for metabolic studies . In addition , while our knockout strategy produced a non-functional protein without the critical DNA binding domain as shown in our study ( Figure 3 ) , the null mice designed by Nobrega's group still produced a chimeric protein containing both DNA binding domain and β-catenin binding domain , making it difficult to assess the potential non-specific effect in the cellular signaling pathway driven by the chimeric protein . Furthermore , we employed the hyperinsulinemic-euglycemic clamp techniques to directly measure the endogenous hepatic glucose production as well as whole body glucose metabolism , and directly provided the evidence for the role of TCF7L2 in hepatic glucose production , while they only performed the glucose tolerance test without the further assessment of the role of other tissues that might affect the glucose homeostasis in their mice . Indeed , the role of TCF7L2 in reducing hepatic glucose production in the transformed hepatic cell line was also recently reported [49] , supporting our in vivo data that alterations in hepatic TCF7L2 expression might be critical in glucose production in the mammalian liver . Given the fact that changes in gluconeogenic gene expression per se might not be enough to invoke changes in hepatic glucose production [50] , TCF7L2 might affect yet to be identified pathways to invoke changes in glucose metabolism in vivo . Unbiased systemic approaches might be useful to identify potential transcriptional targets of TCF7L2 in this regard . In summary , we have provided the evidence for the influence of insulin-resistance on the isoform-specific expression of TCF7L2 in the liver , which contributes to the increased glucose production and the resultant hyperglycemia in mammals . A combination of DIO and genetic heterozygous mutations is considered a critical risk factor for the development of type 2 diabetes . DIO-mediated or genetic haploinsufficiency of TCF7L2 promotes hyperglycemia and insulin resistance in mouse models , suggesting that dysregulation of TCF7L2 expression in the liver might be a critical contributor for the insulin resistance and hyperglycemia in humans . Further study is necessary to provide the link between the differential expression patterns for TCF7L2 in the liver and the progression of diabetes in the affected patients .
Full-length sequence of TCF7L2 was PCR-amplified from pYX-mouse TCF7L2 ( Invitrogen ) , and was subcloned into pcDNA3-FLAG . TCF7L2 isoforms ( TCF7L2 M , S , and E ) , TCF7L2 M mutants ( Δβ-catenin and ΔHMG ) , and β-catenin were generated using site-directed mutagenesis . To generate pU6-TCF7L2 RNAi , palindromic sequences corresponding to nucleotides 773–798 from mouse TCF7L2 coding sequence ( 5′-CCA CAG CGC TGA CAG TCA ACG CAT CT-3′ ) were linked to human U6 promoter in the pBluescript KS vector ( Stratagene ) . hG6Pase ( −1227/+57 ) Luc and PEPCK Luc were generated based on the previous report [51] . Adenoviruses expressing GFP only , nonspecific RNAi control ( US ) , and CRTC2 were described previously [12] . Adenovirus expressing TCF7L2 isoforms , TCF7L2 mutants , TCF7L2 RNAi , FoxO1 , FoxO1 RNAi , β-catenin , or β-catenin RNAi were generated by homologous recombination between adenovirus backbone vector pAD-Easy and linearized transfer vector pAD-Track as described previously [52] . For animal experiments , viruses were purified on a CsCl gradient , dialyzed against PBS buffer containing 10% glycerol , and stored at −80°C . Male 4 or 7-week-old C57BL/6 mice were purchased form ORIENT BIO . TCF7L2 heterozygous null mice ( TCF7L2+/− ) were obtained from EUCOMM consortium and were backcrossed with C57BL/6 for 5 times before being used for the experiment . Mice were housed in a specific pathogen-free animal facility at the Sungkyunkwan University School of Medicine ( 12∶12 h light-dark cycle ) . To induce obesity and insulin resistance , male 4-week-old mice were fed a high-fat diet ( 60 kcal % fat diet: D12492 of Research Diets ) for 8–10 weeks . For animal experiments involving adenoviruses , mice were tail vein-injected with recombinant adenovirus ( 0 . 1–0 . 5×109 pfu per mice ) . Adenovirus-mediated expression was exclusively detected in the liver tissues , but not in other insulin sensitive tissues ( data not shown ) . In addition , plasma ALT and AST levels were not significantly different between mice among the same experimental groups that were injected with various adenoviruses ( data not shown ) . To measure fasting blood glucose level , animals were fasted for 16 h or 6 h with free access to water . For glucose tolerance test ( GTT ) and pyruvate tolerance test ( PTT ) , 16 h-fasted mice were injected intraperitoneally with glucose ( 2 g/kg of body weight for chow diet and 1 . 5 g/kg of body weight for high-fat diet ) . For insulin tolerance test ( ITT ) , 6 h-fasted mice were injected intraperitoneally with 1 unit/kg ( chow diet ) or 1 . 5 unit/kg ( high-fat diet ) body weight of insulin . Blood glucose levels were measured from tail vein blood collected at the designated times . All procedures were approved by the Sungkyunkwan University School of Medicine Institutional Animal Care and Use Committee ( IACUC ) . Primary hepatocytes were isolated from 200 g of Sprague Dawley rats or 8-week-old male C57BL/6 mice by collagenase perfusion method [12] . Briefly , 1×106 cells were plated in 6-well plates with medium 199 ( Sigma ) supplemented by 10% FBS , 10 units/ml penicillin , 10 µg/ml streptomycin , and 10 nM dexamethasone for 6 h . After attachment , cells were infected with adenovirus for 24 h ( for adenovirus expressing GFP , TCF7L2 M , TCF7L2 S , TCF7L2 E , CRTC2 , or FoxO1 ) or 48 h ( for adenovirus expressing US , TCF7L2 RNAi , β-catenin RNAi , CRTC2 RNAi , or FoxO1 RNAi ) . Subsequently , cells were maintained in medium 199 without 10% FBS for 18 h , and were treated with 10 µM forskolin for 2 h or 100 nM insulin for 24 h ( for RNA ) and 15 min ( for protein ) . To measure glucose production , cells were incubated in serum-free media for 16 h , and then were stimulated with 10 µM forskolin and 1 nM dexamethasone and/or 100 nM insulin in Krebs-ringer buffer containing gluconeogenic substrates ( 20 mM lactate and 2 mM pyruvate ) for 8 h . Glucose concentrations were measured using a Glucose Assay Kit ( Cayman Chemical ) . Total RNA from either primary hepatocytes or liver tissue was extracted using Easy-spin total RNA extract kit ( iNtRON biotechnology , Inc . ) . 1 µg of total RNA was used for generating cDNA with amfiRivert reverse transcriptase ( GenDEPOT ) , and was analyzed by quantitative PCR using SYBR green PCR kit and TP800 Thermal Cycler Dice Real Time System ( TAKARA ) . PCR array for glucose metabolism was purchased from Qiagen , and was used according to the manufacturer's instructions . All data were normalized to expression of ribosomal L32 in the corresponding sample . Human hepatoma HepG2 cells were maintained with Ham's F12 medium supplemented with 10% FBS , 10 units/ml penicillin , and 10 µg/ml streptomycin . For transfection , TrnasIT-LT1 Reagent ( Mirus Bio Corporation ) was used according to the manufacturer's instructions . Each transfection was performed with 200 ng of luciferase construct , 50 ng of β-galactosidase plasmid , and 2 . 5–10 ng of expression vector for TCF7L2 M , TCF7L2 S , TCF7L2 E , CRTC2 , or FOXO1 . After 24 h , cells were serum starved for 18 h , and then were stimulated with either 10 µM forskolin or DMSO vehicle for 4 h . Western blot analyses of whole-cell extracts were performed as described [53] . The specific primary antisera for TCF7L2 M , S , and E were produced from GenScript . Antibodies for TCF7L2 , AKT , phosphor-AKT , phosphor-GSK3β , FOXO1 , and phosphor-FOXO1 were from Cell Signaling Technology . Antibodies for HSP90 , insulin receptor , and GSK3β were obtained from Santa Cruz , antibodies for α-tubulin , β-actin , and flag-M2 were provided from Sigma-Aldrich , antibody for CRTC2 was from Calbiochem , and antibody for phospho-insulin receptor ( Tyr1162/1163 ) was from Millipore . The specific signals were amplified by addition of horseradish peroxidase-conjugated secondary antibodies ( Abcam ) , and were visualized by using an enhanced chemiluminescence system ( Abfrontier ) . Nuclear isolation , cross-linking , and chromatin immunoprecipitation assays on mouse primary hepatocyte samples were performed as described previously ( Jaeschke and Davis , 2007 ) . Precipitated DNA fragments were analyzed by PCR using primers against relevant mouse promoters . Blood glucose levels were determined from tail vein blood using an automatic glucose monitor ( One Touch; LifeScan , Inc . ) . Plasma TG and NEFA were measured by colorimetric assay kits ( Wako ) . Plasma insulin was measured by Mouse Insulin ELISA Kit ( U-Type; Shibayagi Corp . ) . Plasma IGFBP1 was measured by Mouse IGFBP-1 ELISA Kit ( Immuno-biological Laboratories , Inc . ) . Hepatic glycogen level was measured by EnzyChrom Glycogen Assay Kit ( BioAssay Systems ) . Total liver lipids were extracted with chloroform-methanol ( 2∶1 , v/v ) mixture as described previously [54] . Seven days prior to the hyperinsulinemic-euglycemic clamp studies , indwelling catheters were placed into the right internal jugular vein extending to the right atrium . After an overnight fast , [3-3H]glucose ( HPLC purified; PerkinElmer ) was infused at a rate of 0 . 05 mCi/min for 2 h to assess the basal glucose turnover . Following the basal period , hyperinsulinemic-euglycemic clamp was conducted for 120 min with a primed/continuous infusion of human insulin ( 84 pmol/kg prime , and 12 pmol/kg/min infusion ) ( Eli Lilly ) . Blood samples ( 10 ml ) were collected at 10–20 min intervals , plasma glucose was immediately analyzed during the clamps by a glucose oxidase method ( GM9 Analyzer; Analox Instruments ) , and 20% dextrose was infused at variable rates to maintain plasma glucose at basal concentrations ( 6 . 7 mM ) . To estimate insulin-stimulated whole-body glucose fluxes , [3-3H]glucose was infused at a rate of 0 . 1 mCi/min throughout the clamps as previously described [55] , [56] . Blood samples ( 10 ml ) for the measurement of plasma 3H activity were taken at the end of the basal period and during the last 45 min of the clamp . Glucose flux was calculated as described previously [55] , [56] . Results of Q-PCR and promoter assay were shown as mean ± SD . Values of metabolites were shown as mean ± SEM . The comparison of different groups was performed using two-tailed unpaired Student's t test . In all statistical comparisons , p value<0 . 05 were considered statistically significant and reported as in legends . | Previous genome-wide association studies revealed that TCF7L2 is a strong candidate for a type 2 diabetes gene . However , the direct involvement of TCF7L2 on hepatic glucose metabolism has been elusive to date . Here we show that TCF7L2 is critical in mediating transcriptional control of hepatic glucose production . We found that hepatic expression of nuclear isoforms of TCF7L2 was reduced in mouse models of insulin resistance . Acute depletion of TCF7L2 in the liver promoted glucose intolerance and up-regulation of gluconeogenic genes , while ectopic expression of TCF7L2 in DIO mice improved glucose tolerance . TCF7L2 was shown to bind to the gluconeogenic promoters , thereby interfering with the promoter occupancies of both CREB/CRTC2 and FoxO1 on their cognate sites . Furthermore , TCF7L2 haploinsufficiency promoted higher glucose levels with impaired glucose tolerance and increased hepatic glucose production in mice , and adenovirus-mediated TCF7L2 expression in the liver reversed the phenotype . We propose that TCF7L2 is a critical player in regulating glucose homeostasis in mammals by modulating hepatic glucose production . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"carbohydrate",
"metabolism",
"diabetes",
"mellitus",
"type",
"2",
"diabetic",
"endocrinology",
"dna",
"transcription",
"animal",
"models",
"model",
"organisms",
"metabolic",
"pathways",
"gene",
"expression",
"endocrinology",
"diabetes",
"and",
"endocrinology",
"biology",
"mouse",
"molecular",
"biology",
"biochemistry",
"molecular",
"cell",
"biology",
"metabolism"
] | 2012 | TCF7L2 Modulates Glucose Homeostasis by Regulating CREB- and FoxO1-Dependent Transcriptional Pathway in the Liver |
Minor histocompatibility ( H ) antigens are allogeneic target molecules having significant roles in alloimmune responses after human leukocyte antigen–matched solid organ and stem cell transplantation ( SCT ) . Minor H antigens are instrumental in the processes of transplant rejection , graft-versus-host disease , and in the curative graft-versus-tumor effect of SCT . The latter characteristic enabled the current application of selected minor H antigens in clinical immunotherapeutic SCT protocols . No information exists on the global phenotypic distribution of the currently identified minor H antigens . Therefore , an estimation of their overall impact in human leukocyte antigen–matched solid organ and SCT in the major ethnic populations is still lacking . For the first time , a worldwide phenotype frequency analysis of ten autosomal minor H antigens was executed by 31 laboratories and comprised 2 , 685 randomly selected individuals from six major ethnic populations . Significant differences in minor H antigen frequencies were observed between the ethnic populations , some of which appeared to be geographically correlated .
Minor histocompatibility ( H ) antigens are polymorphic peptides that are presented on the cell surface by major histocompatibility complex class I or II molecules [1] . These peptides are derived from allelic cellular proteins encoded by autosomal genes or by genes of the Y chromosome [2] . In general , minor H antigen-encoding proteins are biallelic , encoding an immunogenic and a nonimmunogenic allele . Individuals homozygous for the nonimmunogenic minor H allele can develop immune responses to cells expressing the immunogenic minor H allele . Minor H antigen alloimmune responses readily occur in the setting of human leukocyte antigen ( HLA ) –matched allogeneic solid organ and stem cell transplantation ( SCT ) [3 , 4] . The role of minor H antigen disparities is most thoroughly analyzed in the setting of HLA–matched SCT , wherein powerful minor H antigen alloimmune reactivities have been documented in the graft-versus-host direction ( for review see [5] ) . Minor H antigen responses are detected during the development of graft-versus-host disease ( GvHD ) and during the curative graft-versus-tumor ( GvT ) responses after HLA-matched SCT ( for review see [6] ) . This seemingly conflicting role of the minor H antigens in the two GvH ( i . e . , GvHD and GvT reaction ) responses was dissected by the differential tissue distribution of minor H antigens . In vitro studies showed differential modes of expression of the various minor H antigens being either ubiquitously expressed or with expression restricted to the hematopoietic system [7 , 8] . Ubiquitously expressed minor H antigens are present on many cells and tissues , including the GvHD target organs: skin , liver , and gut . These minor H antigens are therefore particularly relevant in the development of GvHD . Hematopoietic system–restricted minor H antigens are only expressed by hematopoietic cells and their progenitors with no expression on nonhematopoietic cells . Some of these minor H antigens can also be expressed by tumor cells [9–12] . Minor H antigens with tissue expression limited to cells of the hematopoietic system and/or tumor cells are especially relevant for the GvT activity . For example , cytotoxic T cells specific to the hematopoietic-restricted minor H antigens HA-1 and HA-2 eradicate circulating leukemic cells and leukemic progenitors cells in vitro [13 , 14] and in an in vivo translational mouse model [15] and can be isolated in the course of remission of donor lymphocyte infusion–treated patients after HLA-matched SCT [16 , 17] . Currently , hematopoietic system- and tumor-specific minor H antigenic peptides are being applied in clinical phase I and II vaccination studies to boost the GvT response ( reviewed in [6] ) . Although scarcely analyzed , minor H antigens are likely to function as potential risks for graft rejection of HLA-matched solid organ transplants [18] . The apparent need of the organ transplant recipient for lifelong immunosuppressive drugs supports this notion . Interestingly , a beneficial effect of minor H antigen mismatching on long-term HLA-matched renal allograft survival has been recently described [19] . Namely , HA-1 specific T regulator cells were present in a patient who had discontinued immune suppression over 30 y ago while sustaining normal kidney function . Herewith a novel characteristic for human minor H antigens has been disclosed with potential impact for decrement of immunosuppressive therapy in solid organ grafting . Minor H antigen alloimmune responses also occur in the physiological setting of pregnancy . Pregnancy leads to a mutual flow of cells between mother and child . The exposure to either foreign maternal or paternal minor H antigens during pregnancy drives the generation of minor H antigen specific T cells in mutual direction [20 , 21] . To date , it is clear that minor H antigenic responses cannot be ignored in the nonphysiological situation of transplantation and in the physiological setting of pregnancy , where female donors with a history of pregnancy and cord blood are used as sources for SCT . Moreover , various minor H antigen identification systems recently became available , facilitating their molecular identification ( for overview see [2] ) . It is therefore timely to estimate the potential impact of the minor H antigens molecularly identified today in the different settings described above . Furthermore , since the first minor H antigen–based clinical vaccination trials in SCT protocols have begun , it is of great importance to estimate their potential applicability in different ethnic groups . Hereto , information on the phenotype frequencies of minor H antigens is required . Population frequency analyses of minor H antigens have so far been performed for a limited number of minor H antigens and are mainly limited to the White population [22] . Additional but scarce information on some of the minor H antigen genotypes is provided by the HapMap project [23] . The present multicenter study reports on a worldwide analysis providing for the first time the global phenotype frequencies of the ten autosomally encoded minor H antigens identified at the initiation of the study . Using these data , geographic patterns were additionally analyzed . Moreover , the theoretical clinical usefulness in HLA-matched SCT of each individual minor H antigen was estimated within six different ethnic populations .
Genomic typing data of ten autosomally encoded minor H antigens were obtained from 2 , 685 individuals from five different continents comprising 16 different countries and five different ethnic groups . The backgrounds of the five ethnic groups were: ( 1 ) 305 Asian/Pacific Islanders , 11 . 4%; ( 2 ) 162 Black , 6 . 0%; ( 3 ) 2 , 011 White , 74 . 9%; ( 4 ) 119 Mestizo Mexican , 4 . 4%; and ( 5 ) 88 specified as “others , ” 3 . 3% . The Asian/Pacific Islander group could be subdivided into Southeast Asian/Southern Chinese ( 10 . 8% of the Asian/Pacific Islander group ) , Asian Indian ( 67 . 9% ) , Japanese ( 20 . 3% ) , and Asian not otherwise specified ( 1 . 0% ) . The Black population consisted of 102 African Americans ( 63 . 0% of the Black population ) , 54 African Black both parents born in Africa ( 33 . 2% ) , three South or Central American Black ( 1 . 9% ) , and three were Black not otherwise specified ( 1 . 9% ) . The White population originated from Europe ( 82 . 1% ) , Australia ( 0 . 6% ) , South Africa ( 0 . 5% ) , Southern America ( 6 . 1% ) , and Northern America ( 10 . 7% ) . The group “others” comprised a major population of Cape Colored individuals from South Africa ( n = 65; 2 . 4% of the study population ) and a Mulatto population from Brazil ( n = 23; 0 . 9% ) . Both groups comprise individuals of mixed ethnicity . Linkage analyses of the ACC-1/ACC-2 haplotypes in Whites who are homozygous for ACC-1 and/or ACC-2 , yield a correlation coefficient r2 of 0 . 91 between the alleles of ACC-1 and ACC-2 , indicating the existence of immunogenic and nonimmunogenic ACC-1/ACC-2 haplotypes ( Table 4 ) . In the other ethnic populations , the correlation is much weaker , and clearly absent in the Black population ( r2 = 0 . 18 ) . Due to the density of the number of geographical locations and of individuals from the identical ethnic background per location , reliable geospatial analysis could only be executed on the White population . The information on the phenotype frequencies subsequently allows calculation of the chance of having a minor H antigen disparate donor/recipient pair in the HLA-matched related ( sib ) and unrelated ( matched unrelated donor , MUD ) SCT settings . Frequencies of HLA molecules also vary among the ethnic populations ( http://www . allelefrequencies . net ) . Consequently , the theoretical estimations of the minor H antigen phenotypic disparities in HLA-matched pairs are corrected for the frequency of the relevant HLA restriction molecule in each ethnic population .
The present global study demonstrates the genotype and phenotype frequencies of all ten autosomally encoded minor H antigens identified at the initiation of the study , in 2 , 685 individuals of six different ethnic populations . Based on this information , the geospatial frequency distribution of the currently known minor H antigens is also described . Moreover , the phenotype frequencies provide for the first time a theoretical estimation for the relevance of ten minor H antigens for their potential use in minor H antigen–based immunotherapeutic protocols in SCT . Small but significant differences in the genotype and phenotype frequencies in the various ethnic populations were observed for all autosomally encoded minor H antigens . The highest variation among the ethnic populations was observed for UGT2B17 ( Figure 2C ) . The UGT2B17 frequency data in the White population are in line with previous studies [22] . As mentioned earlier , the UGT2B17 immunogenic phenotype results from homozygous deletion of the entire UGT2B17 gene . Interestingly , homozygous UGT2B17 gene deletion seems to be tolerated differently in the various ethnic populations . In the present study , we also addressed the issue of linked alleles by analyzing ACC-1 and/or ACC-2 homozygous individuals . The two SNPs responsible for the ACC-1 and ACC-2 minor H antigens are both located on the BCL2A1 gene [25] . The distance between these two SNPs is only 189 bp . Cellular recognition assays indicated that , within the White population from the Centre d'Etude du Polymorphisme Humain , the immunogenic alleles of ACC-1 and ACC-2 are linked [26] . The same study , however , also reported on Japanese individuals positive for only one of the two minor H antigens , thus providing evidence that the ACC-1 and ACC-2 are defined by distinct SNPs , which was confirmed by the characterization of the two relevant minor H peptides . In the White population , a strong linkage between the two alleles of ACC-1 and ACC-2 was indeed observed ( Table 4 ) . However , in the other ethnic populations , haplotypes containing an immunogenic allele for one and a nonimmunogenic allele for the other BCL2A1 polymorphism occur relatively frequently . These observations have important implications for minor H antigen typing , particularly in non-White individuals , where genotyping for both the ACC-1 and the ACC-2 minor H antigen may be essential . The availability of the large dataset on the phenotype frequencies of ten minor H antigens in the White population ( 2 , 011 White individuals from 22 different geographical locations ) allowed for the performance of geospatial analyses . Geographical correlations were observed for HA-8 and UGT2B17 . A distinct north–south gradient for the minor H antigen HA-8 was found , with high frequencies for the immunogenic phenotype in the northern parts of Europe and lower frequencies in the Mediterranean regions . It would be of interest to specifically analyze more samples from both the Scandinavian and the Mediterranean region to further confirm these observations . North-to-south gradients have been suggested to be a more general phenomenon for SNPs . A recent study demonstrated clustering of northern and southern European populations via SNP analyses [27] . However , since none of the other SNP-based minor H antigens demonstrated a north-to-south frequency gradient , we cannot confirm this suggestion . The geospatial analyses of the White population on the European continent indicate a strong west-to-east correlation between the geographic location and the phenotype frequency of UGT2B17 . With 58 . 7% in India , 32 . 1% in Taiwan , and 30 . 0% in Japan , the current Asian dataset suggests that the west-to-east gradient of UGT2B17 extends further eastwards in Eurasia . Additional data from regions such as Eastern Europe and the Middle East are required to strengthen this supposition . Apart from HA-8 and UGT2B17 , no geographical correlation could be detected for the other eight minor H antigen phenotype frequencies . A previous study analyzing geospatial distribution observed a north-to-south gradient for HA-1 within Europe [28] . The data included in the latter meta-analysis match our data and our extrapolation to Spain and Greece . Indeed , if only samples from the same regions ( countries: The Netherlands , Germany , and Italy ) had been analyzed using our data , a similar conclusion could have been drawn . However , we observed a large variation in the frequency of the immunogenic HA-1 phenotype in central Europe , leading to an absence of a clear north–south gradient for HA-1 in our study . The limited number of locations and the fact that these locations were oriented along a north-to-south axis , thereby skewing the results towards a similar orientation of the gradient , may have influenced the conclusions of Di Terlizzi , et al . [28] . Minor H antigens are peptides arising from genomic polymorphisms . Interestingly , most minor H antigens display a unidirectional immune recognition pattern , where the one allele is immunogenic and the other nonimmunogenic . Various mechanisms have been reported to be responsible for this phenomenon of “one allele immunogenicity . ” One of these mechanisms described is via differential antigen processing of HA-3 by the proteasome [29] or by TAP translocation of HA-8 [30] . The lack of a nonimmunogenic counterpart can also have a genetic basis . These cases involve mechanisms such as gene deletion of UGT2B17 [31] , frame-shift due to nucleotide insertion for LRH-1 [32] , and introduction of a translation termination codon for PANE1 and ECGF1 [12 , 33] . As discussed above , the minor H antigen alloimmune responses are generally unidirectional . Consequently , the immune responses between minor H antigen disparate HLA-matched sibs and MUD pairs are similarly unidirectional . The obtained information on the global phenotype frequencies of the minor H antigens known to date allows estimations on the disparity rates between putative HLA-matched stem cell donor and recipient pairs . The clinical relevance of minor H antigens currently lies in their curative effect following SCT . Minor H antigens qualifying for usage in the GvT effect of SCT are those with well-defined restricted cell and tissue expression . Earlier , dissection of minor H antigens into broadly expressed and expression restricted to the hematopoietic system has been described [7] . Interestingly , the hematopoietic minor H antigens HA-1 and ACC-1/ACC-2 show additional expression on carcinomas [9–11] . The calculations of the current clinically relevant minor H antigens are thus based on both their unidirectional immunogenicity and their hematopoietic—and in case of HA-1 and ACC-1/ACC-2 , hematopoietic and solid tumor—expression profiles . The minor H antigen disparity rate calculations require additional correction of the HLA restriction molecules that are also differentially distributed among the various ethnic populations . For correct estimations of the current minor H antigen clinical potential , all of the above characteristics have been taken into account . The results can be summarized as follows . The high phenotype frequency of HLA-A2 , in combination with a relatively high disparity rate for HA-1 in all ethnic populations studied , currently marks HA-1 as the most favorable minor H antigen for clinical application in all ethnic populations . Similar results , although limited to only one or two ethnic groups , were obtained for the minor H antigens ACC-1 , ACC-2 , and HB-1Y . The ACC-1 minor H antigen is restricted to the Asian/Pacific population , with 6 . 5% disparity in sib pairs and 12 . 1% in MUD pairs . Similarly , ACC-2 can only be applied in a significant proportion of the pairs in the White and Mulatto population , as counts for HB-1Y . Finally , estimating the chance of having at least one mismatch for a hematopoietic system–restricted minor H antigen out of the eight hematopoietic minor H antigens studied herein , revealed 21 . 2% and 33 . 6% for White sib and MUD pairs respectively , and 7 . 4% and 11 . 9% for the Black population , respectively . In all of these estimations , the HLA allele frequencies of the respective populations were taken into account . The estimations of the minor H antigen disparity rates between HLA-matched pairs strongly differ among the various ethnic populations , raising the question of whether some populations display more phenotypic diversity than others . To address this issue , we subjected the mean genetic disparity rates per population to a paired t-test . The overall genetic disparity is lowest in the Black population ( Figure 2 ) . It remains unclear which mechanism forms the basis for this phenomenon . The Black population analyzed in this study comprised two major subpopulations , i . e . , the African Black and the African American population . In contrast with the African Black population , the African American population has considerable Caucasoid admixture ( E . D . du Toit , personal communication ) . We therefore performed a separate analysis of the minor H antigen phenotype frequencies in these two subpopulations for all minor H antigens . Although a tendency towards White minor H antigen phenotype frequencies was observed for the African American population , the differences didn't reach statistical significance ( unpublished data ) . Similarly , a subanalysis of the Asian/Pacific population yielded no major regional differences except for the UGT2B17 minor H antigen ( unpublished data ) . The clinical potential of the current set of identified minor H antigens is clearly limited and points to enlargement of the pool of clinically relevant minor H antigens . Reverse immunology has recently been applied for identifying new minor H antigens [34] . This approach uses computer algorithms such as SYFPEITHI [35] and Bimas [36] to predict binding of polymorphic peptides to the HLA allele of choice . Accordingly , the UGT2B17 gene could be analyzed for new minor H peptides binding to frequently occurring HLA alleles in the various ethnic populations ( Table S2 ) . Similar analyses could be executed for HB-1H for its relevance in the Mexican Mestizo population , and HA-2 and PANE1 in the Brazilian Mulatto population . Finally , to enlarge the pool of minor H antigens potentially applicable in a stem cell–based protocol for solid tumors , the reverse immunology approach can be helpful in searching for minor H antigens with restricted expression on carcinoma cells [9–11] . Here , the search is specifically focused on SNPs in genes involved in tumorigenesis [5] . In conclusion , we here report differences in minor H antigen genotype and phenotype frequencies among and within the major ethnic populations . Some of these differences are geographically correlated , while others seem to be more randomly distributed . Moreover , the potential clinical application of hematopoietic specific minor H antigens as immunotherapeutic tools in SCT for hematological malignancies was estimated . The theoretical chance of having a difference for at least one hematopoietic minor H antigen in the whole study group , thus corrected for the HLA frequencies , is 11 . 9% to 33 . 6% in HLA-matched unrelated pairs .
The study population consisted of 2 , 685 locally available randomly selected healthy subjects and transplant donors . All research samples and data were collected according to the subjects' guidelines and protocols of the local institutional review boards . Individuals with an unknown ethnical background were excluded from the analyses . DNA was isolated using the local standard procedures used for HLA typing . Genotyping of ten autosomally encoded minor H antigens and of HY was performed using the PCR-SSP technique ( Invitrogen , http://www . invitrogen . com ) . The details of the minor H antigen typing methodology have been described before [37] . Previously typed cells from the Centre d'Etude du Polymorphisme Humain-Human Genome Diversity Project Cell Line Panel ( CEPH panel ) and from the reference DNA samples from the University of California , Los Angeles , International DNA Exchange Program were used as controls . All PCR-SSP samples were analyzed on agarose gel . If the internal control PCRs failed for one of the two allele of a particular minor H antigen , the typing for that minor H antigen was marked as incomplete and data were excluded from the analyses . The alleles of each minor H antigen were reported as the amino acid resulting from the relevant SNP . Typing data and sample parameters were submitted to a central online database ( http://www . lumc . nl/ihiw14 ) . Typing data were analyzed at three levels: ( 1 ) the allele frequency , ( 2 ) genotype frequency , and ( 3 ) phenotype frequencies . Allele and genotype frequencies except those of UGT2B17 were calculated by direct counting . Since the minor H antigen UGTB17 is the result of gene deletion , the applied methodology did not allow detection of UGT2B17 heterozygosity . Therefore the UGT2B17 allele frequencies were estimated as for the nonimmunogenic allele and 1 − for the immunogenic allele , where “aa” is the frequency of the nonimmunogenic phenotype . The heterozygosity frequencies ( Aa ) for this minor H antigen were estimated by 2 × ( 1 − ) × ( ) , assuming a Hardy-Weinberg equilibrium . Observed genotype frequencies were tested against Hardy-Weinberg equilibrium . The phenotype frequencies have been deduced from the genotype frequencies . Individuals who were homozygous or heterozygous positive for the immunogenic allele were regarded as phenotypically positive . Individuals carrying two nonimmunogenic alleles were marked as phenotypically negative . Since the HB-1 minor H antigen has been demonstrated to be immunogenic in either direction , phenotype frequencies have been calculated separately for the HB-1H and the HB-1Y allele as immunogenic alleles . The statistical significance of differences in allele , genotype , and phenotype frequencies between the various populations and geographic regions were determined by Chi-square analysis . The SPSS statistical software system ( SPSS 11 . 0 , http://www . spss . com ) was used for these analyses . p-values lower than 0 . 05 were considered statistically significant . Linkage of the alleles of ACC-1 and ACC-2 was analyzed with the case-control haplotype inference ( Chaplin ) software [34] . Individuals who were heterozygous for both ACC-1 and ACC-2 were classified as noninformative and excluded from the analysis . The disparity rate is defined as the chance of a putative transplant recipient with a minor H antigen immunogenic phenotype having an HLA-identical related or HLA-matched unrelated stem cell donor being homozygous for the nonimmunogenic allele of the same minor H antigen encoding gene . In the case of a putative HLA-matched unrelated donor selection , the disparity rate is the frequency of homozygous immunogenic ( AA ) and heterozygous ( Aa ) individuals multiplied by the frequency of homozygous nonimmunogenic individuals ( aa ) : ( ( AA ) + ( Aa ) ) × ( aa ) . Sib transplantation disparity rates were estimated by 3/16 ( Aa ) 2 + 1/2 ( Aa ) ( aa ) . For both related and unrelated situations , the rates were multiplied by the phenotype frequency of the relevant HLA restriction molecule . To investigate the level of disparity within each population , two-tailed Wilcoxon signed rank tests were performed . Spatial frequency distribution maps for each of the immunogenic minor H antigen phenotype frequencies were constructed for the White population samples using the Kriging procedure [38] with the Surfer 8 software ( Golden Software , http://www . goldensoftware . com ) . For these analyses the geographical area between 29 . 000/−28 . 000 ( latitude/longitude ) and 72 . 000/49 . 000 , covering the European continent , was split into a 1 , 000 × 550 squares grid using the standard software settings . Cartesian coordinates of the participating institutes were obtained from the Getty Thesaurus of Geographic Names Online ( http://www . getty . edu/research/conducting_research/vocabularies/tgn ) .
The National Center for Biotechnology Information ( NCBI ) Gene database ( http://www . ncbi . nlm . nih . gov/entrez/query . fcgi ? db=gene ) accession numbers for the minor H antigen genes used in the study are: HMHA1 ( HA-1 ) , 23526; MYO1G ( HA-2 ) , 64005; AKAP13 ( HA-3 ) , 11214; KIAA0020 ( HA-8 ) , 9933; HMHB1 ( HB-1 ) , 57824; BCL2A1 ( ACC-1 and ACC-2 ) , 597; SP110 ( SP110 ) , 3431; CENPM ( PANE1 ) , 79019; UGT2B17 ( UGT2B17 ) , 7367 . | Peptides from cellular proteins evoking alloimmune responses in human leukocyte antigen–identical transplantation are called minor histocompatibility ( H ) antigens . Upon hematopoietic stem cell transplantation ( HSCT ) for hematological malignancies or for solid tumors , responses to minor H antigens may have detrimental effects , e . g . , graft-versus-host-disease and graft rejection , but can also significantly contribute to the eradication of the tumor cells . We designated the latter antigens as “tumor” minor H antigens . Current clinical trials aim at using these tumor minor H antigens to boost the graft-versus-tumor response . So far , it is unclear how frequently the HSCT recipient and donor differ in their minor H antigens and thus how many cancer patients are eligible for minor H antigen-based treatment . Therefore , worldwide 31 laboratories joined forces to determine the genotype and phenotype frequencies of ten autosomally encoded minor H antigens in six ethnic populations . The frequencies vary depending upon ethnic background and geographic location of the population , implying that the potential applicability of the tumor minor H antigens differs from one population to another . Depending on the population , the current theoretical percentages of clinical application of the eight tumor minor H antigens in HLA-matched combinations is estimated at 12% to 34% . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"oncology",
"homo",
"(human)",
"hematology",
"immunology"
] | 2007 | Phenotype Frequencies of Autosomal Minor Histocompatibility Antigens Display Significant Differences among Populations |
Zoonotic disease ( ZD ) pose a serious threat to human health in low-income countries . In these countries the human burden of disease is often underestimated due to insufficient monitoring because of insufficient funding . Quantification of the impact of zoonoses helps in prioritizing healthcare needs . Kyrgyzstan is a poor , mountainous country with 48% of the population employed in agriculture and one third of the population living below the poverty line . We have assessed the burden of zoonoses in Kyrgyzstan by conducting a systematic review . We have used the collected data to estimate the burden of ZDs and addressed the underestimation in officially reported disease incidence . The estimated incidences of the ZDs were used to calculate incidence-based Disability Adjusted Life Years ( DALYs ) . This standardized health gap measure enhances comparability between injuries and diseases . The combined burden for alveolar echinococcosis , cystic echinococcosis , brucellosis , campylobacteriosis , congenital toxoplasmosis , non-typhoidal salmonellosis and rabies in Kyrgyzstan in 2013 was 35 , 209 DALYs [95% Uncertainty interval ( UI ) :13 , 413–83 , 777]; 576 deaths [95% UI: 279–1 , 168] were attributed to these infections . We estimate a combined median incidence of ZDs of 141 , 583 cases [95% UI: 33 , 912–250 , 924] in 2013 . The highest burden was caused by non-typhoidal Salmonella and Echinococcus multilocularis , respectively 14 , 792 DALYs [95% UI: 3 , 966–41 , 532] and 11 , 915 DALYs [95% UI: 4 , 705–27 , 114] per year . The health impact of zoonoses in Kyrgyzstan is substantial , comparable to that of HIV . Community-based surveillance studies and hospital-based registration of all occurrences of zoonoses would increase the accuracy of the estimates .
Zoonoses are diseases in humans , which are naturally transmissible directly or indirectly from vertebrate animals . Of 1415 species of infectious organisms know to be human pathogens , 61% are zoonotic [1] . The Food and Agriculture Organization of the United Nations ( FAO ) , the World Health Organization ( WHO ) and the World Organisation for Animal Health ( OIE ) have underlined the important socioeconomic impact of these diseases , yet in many low income countries the burden of zoonoses remains unknown [2] . The lack of information often results in a vicious circle of underestimation and limited incentive to quantify the problem [3 , 4] . Kyrgyzstan is a country in Central Asia , neighboured by China in the west , Kazakhstan in the north , and Tajikistan and Uzbekistan in the southeast ( Fig 1 ) . Because of a poor functioning veterinary and sanitation system , emerging zoonoses are an increasing problem [5] . Since independence in 1991 , veterinary services deteriorated , causing an increase in zoonotic disease ( ZD ) [6 , 7] . At particular risk are the 64% of the inhabitants who live in rural areas , where livestock farming plays an important role . Seventy-six percent of these rural dwellers are considered to be poor [8] . The small-scale farming , which is often nomadic , allows intensive contact between humans and animals [9] . Furthermore , in Kyrgyzstan , an estimated 100 DALYs per 100 , 000 were lost due to inadequate hygiene in 2012 [10] , which ranks the country in the 61th place out of 146 low/middle-income countries on disease burden due to poor hygiene . The combination of poor healthcare , poverty , inadequate hygiene , and the close interaction between humans , livestock and other animals , leaves a large share of the population at risk of being infected with zoonoses . Another difficulty in assessing the burden of the diseases , is the low scientific output from Kyrgyzstan which is often published in Russian [11 , 12] . The World Bank and the OIE have advised Kyrgyzstan to develop national animal disease control strategies [2] . A quantification of the impact of zoonoses helps prioritizing these diseases . The aim of this study was to quantify the burden of ZD in Kyrgyzstan using disability-adjusted life years ( DALYs ) a standardized approach to increase comparability of disease impact [13–19] . In this review , we have assessed the available data on zoonoses in Kyrgyzstan with special attention to the potential underreporting using stochastic disease modelling . We have comprehensively summarised the burden of the most important zoonoses that are endemic in Kyrgyzstan and addressed the underestimation in officially reported cases .
The ZDs described in this systematic review ( Table 1 ) are regarded as the most important in terms of socioeconomic impact based on the WHO report on neglected tropical diseases , the World Bank report on Kyrgyzstan of 2011 and other systematic reviews of neighbouring or overlapping regions [2 , 5 , 20 , 21] . We assembled all the available evidence regarding prevalence or incidence of the selected ZDs in Kyrgyzstan since the country became independent . Therefore , the time period for the search was January 1991-January 2016 . Both formal , peer-reviewed scientific literature , and informal sources , grey literature , were considered . A full list of sources can be found in S1 Supporting Information . We conducted a systematic review by following guidelines of the Preferred Reporting Items for Systematic reviews and Meta-Analyses ( PRISMA guidelines , Moher , Liberati , Tetzlaff , Altman , & The PRISMA Group , 2009 ) [31] ( S2 Supporting Information– Prisma checklist ) . A list of synonyms for the ZDs was constructed using the pathogen’s name and alternative ( popular ) names of the disease . The computer search was constructed by combining these terms with the Boolean OR and the term ‘Kyrgyzstan’ with the AND Boolean . The databases of PubMed , Google Scholar , Web of Science , OVID , Scopus , the WHO Global Health Library , Food and Agriculture Organization of the United Nations ( FAO ) and ProMED-mail were searched using English search terms and Google Scholar using Russian search terms . For each database , the search construct was adapted to the specific modus operandi of the search engine . S1 Supporting Information lists the search terms as well as the search constructs for the different databases . Furthermore , we searched the internet for published reports on demographic surveillance sites in the English and the Russian language . Data from government sources was contributed by the co-authors . Following retrieval , studies were selected by critically appraising the titles and abstracts . A study was excluded when it did not address prevalence or incidence for the specific disease , when it was not from the defined period , when it did not address the disease in humans or when it did not address Kyrgyzstan . Secondly , the full text was screened and for each retrieved result the list of references was inspected for additional sources ( backward searching ) . Forward searching was performed by entering the titles in google scholar using the ‘cited by’ function . Searches were executed until no new results were found . Additional results were screened according to the same methodology . Finally , the selected studies were summarized based on study design , study area , disease measure ( prevalence/incidence ) and the reported margin of error ( S3 Supporting Information ) . Each study was critically appraised on methodology , selectivity in reporting and assumptions made by the authors . Fig 2 displays a flow diagram of the used selection strategy . The Disability-Adjusted Life Year ( DALY ) was used as the burden-of-disease metric . It is a health gap measure which quantifies health loss . DALY calculation is a standardized method developed by the World Bank , Harvard School of Public Health and the World Health Organization for the Global Burden of Disease and injury ( GBD ) study and the global burden of foodborne diseases [13 , 32–35] . It allows the comparison of health conditions across countries and across diseases [13 , 32] . In this study , an incidence-based DALY calculation was applied . This allows us to include all sequelae resulting from infection [36 , 37] . The DALYs are calculated as the sum of the healthy years lost to disability ( YLD ) and the years of life lost due to premature death ( YLL ) . The YLD is the sum of the different outcomes that result in disability , where an outcome is defined as sequelae of the disease or another categorisation of the disease , e . g . chronic vs . acute . The YLD per outcome is the product of the duration , the incidence , and the disability weight of the outcome . The YLL is the residual life expectancy at the age of death . YLLs were calculated based on the life table from [38] . YLDs with a lifelong duration were calculated based on a local life table from the WHO for Kyrgyzstan for 2013 [39] . Based on the recommendations and methodology of the GBD 2010 and FERG [33 , 36] , we have used a non-discounted and non-weighted approach in calculating DALYs . If a disease and its outcomes were quantifiable , a corresponding disease outcome model was constructed based on literature . When data on the incidence , the outcome of a disease , or other parameters for the DALY calculation were missing , these gaps were filled using data from neighbouring countries or overlapping regions . The disease model or outcome-tree model was constructed per ZD using health outcomes with an evidence-based causal relationship between infection and outcome . Disagreement in inputs of the disease model between different sources was modelled using distributions ( pert , triangular , and uniform ) accounting for this uncertainty . A full description of the disease models , the input parameters and its uncertainties used to calculate the DALYs can be found in S3 and S4 Supporting Information . Where available , disability weights from the GBD 2010 study were used . Age distributions of outcomes were , when possible , based on data collected from Kyrgyzstan . Furthermore , the total population size , age and sex distribution was obtained from census data by the National Statistical Committee of the Kyrgyz Republic and the United Nations Demographic Yearbook [40 , 41] . The uncertainty in the estimates was modelled using Monte Carlo analysis . We generated in this simulation 10 , 000 draws from the probability distributions . All analyses were performed using R version 3 . 2 . 2 ( R Foundation for Statistical Computing , Vienna , Austria ) [42] . Additional information on the analyses and disease models is provided in S4 Supporting Information . The burden of disease was calculated for the reference year 2013 , a result of a trade-off between data availability and as recent as possible . The data collected by the Department of State Sanitary and Epidemiological Supervision of the Ministry of Health of the Kyrgyz Republic provided the officially reported cases for notifiable diseases which includes a number of zoonoses . This is available online [41 , 43] . We have used the data retrieved from literature to evaluate these reported figures and assess the level of underestimation . As reported in S3 Supporting Information , the availability of published disease data is scarce in Kyrgyzstan; we were not able to assess the burden of anthrax since not enough was known about the outcome of the cases . In 2013 , 16 human cases of anthrax were reported . The majority of the cases were the cutaneous form , Zoldoshev reviewed 217 cases of cutaneous anthrax with no fatalities [44] . For alveolar echinococcosis ( AE ) , cystic echinococcosis ( CE ) , brucellosis and rabies incidence data for Kyrgyzstan was available from the Department of State Sanitary and Epidemiological Supervision of the Ministry of Health of the Kyrgyz Republic [41 , 43] . Prevalence data on AE , CE and brucellosis was used to address the underestimation in the official data; to address the potential underestimation in rabies we have used data from overlapping regions ( Eurasia ) [27] . The incidence of congenital toxoplasmosis was not formally recorded , but Minbaeva et al . ( 2013 ) provided estimates for Kyrgyzstan [45] . For campylobacteriosis and non-typhoidal salmonellosis , no specific disease prevalence or incidence data was recorded for Kyrgyzstan . The incidence estimates are based on the number of acute gastrointestinal infections reported in 2013 [41] and the assumed etiological fraction as described by [21 , 22] . This conservative estimate was used as the lower limit for the number of cases; the upper limit was formed by the etiologic proportion of the diarrhoea incidence multiplied by the gastroenteritis incidence from the European region based on Walker et al . and Lanata et al . [24–27] . Invasive non-typhoidal salmonellosis ( iNTS ) forms an important outcome of non-typhoidal salmonellosis infection since mortality is much higher compared to the gastro-enteric manifestation of the disease [35 , 46] . However limited data is available on the true incidence of iNTS since only few population-based incidence studies have been conducted . Therefore , we have used the ratio between iNTS:NTS as described by Ao et al . [46] who classified Kyrgyzstan in the Asia/Oceania region where the proportion iNTS:NTS was 1:3 , 851 compared to 1:7 in European region which included Russia; the global average ratio was 1:28 [46] .
We identified 438 unique citations and excluded 411 by title and abstract screening . Of the remaining 28 potential eligible citations with relevant abstracts , 10 were eligible for full text review . The PRISMA flowchart summarizing the data collection process is presented in Fig 2 . Reports published during January 1991-January 2016 were searched . The last search was performed on 19-02-2016 . All collected data are summarised in S3 Supporting Information . In 2013 seven ZDs were quantifiable in Kyrgyzstan . AE , brucellosis , campylobacteriosis , CE , congenital toxoplasmosis , NTS and rabies . These were responsible for an estimated total of 141 , 583 [33 , 912–250 , 924] new cases resulting in 35 , 209 [13 , 413–83 , 777] DALYs and 576 [279–1 , 168] deaths ( Table 2 , Fig 3 ) . Both Rabies and AE contribute a large number of DALYs per case , 70 . 1 [10 . 0–90 . 0] DALYs/case and 50 . 3 [20 . 7–78 . 3] DALYs/case respectively , due to high mortality ( Fig 4 ) . Campylobacteriosis and NTS had relatively low mortality but a high incidence; most of the mortality was due to the sequelae Guillain Barre Syndrome ( GBS ) and invasive non-typhoidal salmonellosis ( iNTS ) respectively , see S4 Supporting Information . Infections with salmonellosis and AE were responsible for the majority of deaths , respectively 254 [66–571] and 236 [153–466] . Although only 5 . 1% ( 11/216 ) of the cases of congenital toxoplasmosis was fatal , the DALY/case is high ( 6 . 69 ) due to early onset of the sequelae and the lifelong duration . Table 2 provides the estimates for the number of cases , the DALY , the number of deaths and the DALY per case per disease and their 95% uncertainty range . Fig 3 displays a graphical representation of the per annum burden per disease and its uncertainty , plotting the DALY estimates from Table 2 per disease . Table 3 displays the sensitivity analysis with different iNTS:NTS ratios . Fig 4 provides the percentage of YLD and YLL per disease . Premature mortality , or YLL , ( in blue in Fig 4 ) contributes for all diseases most to the DALY , ranging from 67% for CE to 100% for Rabies .
This work provides a first attempt at quantifying the burden of ZD in Kyrgyzstan . It underlines the lack of published data on many zoonoses in this region . However , the estimates of the impact of the ZDs help to break the vicious circle of underreporting by providing estimates of the true incidence and burden of these diseases . Because of the scarcity of data we did not exclude information based on methodology; we analysed it using conservative assumptions and stochastic modelling to handle uncertainty [47] . We have used official Kyrgyz data and addressed its underestimation . The total burden of the seven quantified ZDs ( 35 , 209 [13 , 413–83 , 777] DALYs in 2013 ) is slightly less than the yearly burden of HIV , which was attributable for 38 , 870 [21 , 261–64 , 297] DALYs in 2010 in Kyrgyzstan [48] . This burden is based on prevalence based DALYs , as used in the GBD 2010 studies [33] . Forty-three percent of the estimated burden of zoonoses or 14 , 967 [6 , 213–32 , 319] DALYs in 2013 , in Kyrgyzstan is caused by echinococcosis ( both AE and CE ) . Torgerson et al . estimated in 2010 that in China 16 , 629 new cases of AE per year arose among 22 . 6 million people at risk [17] . In Kyrgyzstan we estimate for 2013 that 236 cases arose among 5 . 7 million people . However AE is characterized by a clustered distribution and some regions have a much higher incidence rate [49] . The officially reported incidence of AE has increased since 2004 at an alarming rate . Where before 2004 only 0–3 cases per year were reported , in 2013 148 cases were officially reported [49 , 50] . We assume that , corrected for underestimation , the incidence is likely to be approximately 236 [153–466] cases in 2013 . Although the goal of this study was to provide an estimate of the burden of zoonoses in 2013 , the collected data allows us to reflect on temporal trends . Raimkylov & Kuttubaev , and Usubalieva et al . describe an increasing trend over the last decade in the incidence of echinococcosis [49 , 50] , although increased awareness might lead to more diagnoses . The yearly incidence of brucellosis , on the other hand , seems to decline over time ( S3 Supporting Information ) . The application of the ocular Rev-1 vaccination over several years has most likely resulted in this decreased incidence [51] . Over time , the improvement of diagnostics and the application of novel treatments may cause changes in the outcome of the disease and thus the DALY per case . For example , in our analysis , we assumed that all cases of AE are eventually fatal due to insufficient treatment . However , as illustrated in Switzerland , adequate treatment of the disease will lead to an increased survival [52] . This illustrates the need for periodic updating of the burden assessment . We choose an incidence-based approach in the DALY modelling because it allows us to include all sequelae resulting from infection . However , one of the consequences of using the incidence-based DALY approach , is that deaths in the future are attributed to the year of the infection . Careful interpretation of the mortality rate is therefore advised . For example , AE did not cause the reported number of deaths in 2013 since its incidence is increasing and it has a long latency; the deaths that will be caused by the infections diagnosed in 2013 are attributed to that year . For a disease with a short incubation time and a relative constant incidence rate , such as rabies , the difference is not so striking . Using a prevalence based DALY approach in diseases with a trend over time and a long latent phase or incubation period , might lead to under or overestimation as it reflects past infection rather than a present day event [37] . Another limitation is the assumption , in common with other burden studies , that the outcome of diseases can be extrapolated to different countries . Regional differences in pathogens might change the tropism of the causative agent or cause a shift towards certain sequelae . The same holds true for other spatially fluctuating factors , such as co-infection; The incidence of iNTS , for example , is correlated with malaria and HIV infection [46] . This underlines the importance of not only reporting incident cases , but also of documenting disease outcome . Other factors such as ethnicity might also have an influence on disease outcome , as for example has been postulated in tuberculosis and Plasmodium falciparum malaria [53 , 54] . Even more striking are the vast differences in treatment according to region , as illustrated for AE . Likewise , brucellosis , with inadequate treatment is more likely to become chronic or relapse [55] which increases the burden . Underestimation of disease is caused by under-ascertainment and underreporting of cases . A disease might not be severe enough for the patient to visit a medical facility . In addition , patients might have limited access to care or the disease are not accurately diagnosed . Underreporting is the result of incomplete registration of cases . Even in countries with a high standard of medical care , such as WHO high-income countries , reported cases form only the tip of the iceberg of the true incidence . For example , it is estimated that only 1/30 . 3-1/86 cases of campylobacteriosis are reported in the USA [56 , 57]; the estimate in the European Union is that on average 1/47 cases of campylobacteriosis are reported [58] . CE is often substantially underreported . In Uzbekistan the official case numbers appear reported were 1 , 435 cases reported in 2000 and 819 cases reported in 2001 [59] . However , Nazirov and others undertook a detailed study of hospital records throughout Uzbekistan and found a total of 4 , 430 cases in 2000 and 4 , 089 cases in 2001 . Likewise in Chile official notifications between 2001 and 2009 were a mean of 311 cases per annum , whilst a detailed audit of hospital records revealed a mean of 1 , 009 cases per annum [60] . The assumption we made on the underestimation of the incidence of ZDs is conservative . For most ZDs we have used either a uniform or a pert distribution and included the officially reported incidence as minimum and the with a multiplication factor corrected value as maximum . The assumption for the mode in AE , CE and brucellosis are also conservative . The multiplication factor we used to correct for underestimation in brucellosis ( 4 . 6 ) lies close to the mean multiplication factor ( 5 . 4 [1 . 6–15 . 4] ) Kirk et al . used in [35] . Hampson et al . reviewed the global burden of Rabies and estimated for 2010 and estimated 14 rabies deaths in Kyrgyzstan contributing to 887 DALYs [27] . Since rabies is a fatal disease , which often affects the young , it is possible that some cases go unreported in Kyrgyzstan . We believe that our estimate and its 95% uncertainty range represent the true incidence . The burden consists only of the estimated fatal cases , and not the disability caused by dog bites and the burden of the treatment . This indirect burden is highest in countries where crude nerve-tissue vaccines are used [61] , which is not the case in Kyrgyzstan . Other carnivores than dogs , are assumed not be relevant in contributing to the transmission risk [62] , however , there is a steep increase described in the wolf population and an increasing contact rate between humans and these wild carnivores in mainly in the south of Kyrgyzstan [63] . In this study we have explored the proportion of diarrhoea attributable to Campylobacter and non-typhoidal Salmonella in Kyrgyzstan . We assumed etiologic proportion of diarrhoea of both pathogens based on literature [64] . Close inspection of the reported incidence of acute intestinal infections , reveals an approximate two-fold increase in cases between 2004 and 2007 . However , the change was likely because the funding of hospitals was modified to a case-based system [65] . This illustrates that the variance in reported data does not always represent epidemiological change; it can be merely a reflection of an alteration in policy . The estimates we present are based on conservative extrapolates from overlapping regions . However , more accurate incidence data on salmonellosis and campylobacteriosis in Kyrgyzstan are lacking . Estimates of overlapping regions often lacked nuance and tend to group heterogeneous countries . Our median incidence estimates for both NTS ( 1 , 101/100 , 000 ) and campylobacteriosis ( 1 , 305/100 , 000 ) are higher than the estimated yearly incidence by Havelaar et al . of campylobacteriosis ( 802/100 , 000 cases ) and of NTS ( 318/100 , 000 cases ) in the EUR B region [66] . However earlier estimates by the same author are higher; ranging from 1 , 800–11 , 800 cases/100 . 000 for salmonellosis and 2 , 240–13 , 500 cases/100 . 000 for campylobacteriosis [58] . The data presented in [58] shows a correlation between Gross Domestic Product ( GDP ) and both salmonellosis and campylobacteriosis; both diseases have a higher estimated true incidence in countries with a lower GDP . There seems to be no clear relation between the quantity of consumed protein ( egg , chicken , and pork ) according to the FAO and the estimated true incidence of the two diseases in the different European countries ( EU-27 ) [58] . We believe that although chicken , egg and pork consumption in Kyrgyzstan are lower than in EU-27 countries , the lower GDP and the lower hygiene standard in Kyrgyzstan justify our estimates . To obtain more reliable burden estimates of both campylobacteriosis and salmonellosis , it would be advisable to undertake a community-based incidence study in Kyrgyzstan . Both diarrhoea incidence and aetiology are important inputs to narrow the uncertainty around our estimates . Furthermore , a longitudinal study on the aetiology of febrile illness might provide a reliable estimate of the burden of different zoonoses or sequelae ( brucellosis , iNTS , listeriosis , Q-fever , leptospirosis ) . A small scale investigation in Bishkek showed that part of the undiagnosed febrile illness was due to brucellosis and Q-fever [67] . To date , the exact burden of leptospirosis in Kyrgyzstan is unknown . Although occurrence of leptospirosis in cattle in Kyrgyzstan has been reported [68] , no data on the occurrence of this ZD in humans in Kyrgyzstan is available . Torgerson et al . estimated that the burden of leptospirosis in Kyrgyzstan was 927 [355–1629] DALYs per year [69] . Costa et al . clearly illustrate a lack of data on the occurrence of leptospirosis in Central Asia; the estimates of incidence for Kyrgyzstan were based on extrapolation using a multivariable regression model [70] . Likewise , the role of cryptosporidium and giardia as causative agent for ZD in Kyrgyzstan has not been established . These parasites have zoonotic potential [71] , however the incidence of the disease caused by these parasites has not been investigated in Kyrgyzstan , nor has the role of animals in the transmission of these ZDs been quantified . Therefore , burden assessment at this moment is not feasible . Only sequelae that have a solid proven causal relationship with the pathogen have been included in the disease models we used . Reactive arthritis , irritable bowel syndrome and GBS are evidence based sequelae of campylobacteriosis [72] . However , we followed the conservative assumption of Kirk et al . that the relation between some sequelae were not sufficiently proven in middle and high-mortality countries [35] . Most of the burden of salmonellosis is due to YLLs , mainly deaths caused by iNTS . The sensitivity analysis ( Table 3 ) illustrates the influence of the proportion of iNTS:NTS . This underlines the importance of investigating the incidence of iNTS in Kyrgyzstan and is in line with the findings of Ao et al . [46] conclude that there is a lack of population-based incidence data on iNTS . In a limited-means setting such as Kyrgyzstan it is inevitable for policy makers to prioritize health care needs . The DALY provides one tool to do so , but is by itself not sufficient [73] . In the application of the DALY by healthcare legislators , it is important to look at the presented figures in a wider context . DALYs should be combined with for example , economic parameters in cost-utility analyses [74] . It is also important to realize that the DALY might not capture the full effect of the disease and that a disease might have bigger impact than just on the ones directly affected [75] . Especially ZDs often cause economic loss in livestock production as well [21 , 76] . Where in this paper we have only quantified the human burden , it makes sense to extend the work with the assessment of the economic impact of the disease in both humans and animals . Furthermore , it is advised to conduct an integrated approach in disease intervention and prevention where both veterinary and human health officials work together [3] . | Zoonoses are diseases transmitted from vertebrate animals to humans . They can cause a variety of symptoms ranging from mild gastrointestinal complaints to debilitating illness and even death . Especially in low-income countries where animals play an important role for many , the burden of these diseases can be substantial . However , there is often little attention for these diseases , thus they remain under-researched and underfunded . In this review , we present estimates of the burden of the most important zoonotic diseases in Kyrgyzstan for the reference year 2013 . We estimated the burden by calculating the incidence-based disability adjusted life years ( DALYs ) , allowing comparison between diseases and injuries . Disease frequency data is scarce and hospital-based incidence data often underestimates the true incidence of the disease . By addressing the underestimation in officially reported incidence using data from our systematic review , we estimated the true incidence of the most important zoonoses in Kyrgyzstan . We quantified the substantial impact these diseases have on the wellbeing of people in Kyrgyzstan in 2013 . The results underline the need for more intensive monitoring and surveillance of zoonotic diseases . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"parasitic",
"diseases",
"salmonellosis",
"brucellosis",
"bacterial",
"diseases",
"rabies",
"neglected",
"tropical",
"diseases",
"veterinary",
"science",
"campylobacteriosis",
"echinococcosis",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"veterinary",
"diseases",
"zoonoses",
"helminth",
"infections",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases"
] | 2016 | The Burden of Zoonoses in Kyrgyzstan: A Systematic Review |
Foci of the HIV epidemic and helminthic infections largely overlap geographically . Treatment options for helminth infections are limited , and there is a paucity of drug-development research in this area . Limited evidence suggests that antiretroviral therapy ( ART ) reduces prevalence of helminth infections in HIV-infected individuals . We investigated whether ART exposure and cotrimoxazole preventive therapy ( CTX-P ) is associated with a reduced prevalence of helminth infections . This cross-sectional study was conducted at a primary HIV-clinic in Lambaréné , Gabon . HIV-infected adults who were ART-naïve or exposed to ART for at least 3 months submitted one blood sample and stool and urine samples on 3 consecutive days . Outcome was helminth infection with intestinal helminths , Schistosoma haematobium , Loa loa or Mansonella perstans . Multivariable logistic regression was used to assess associations between ART or CTX-P and helminth infection . In total , 408 patients were enrolled . Helminth infection was common ( 77/252 [30 . 5%] ) . Filarial infections were most prevalent ( 55/310 [17 . 7%] ) , followed by infection with intestinal helminths ( 35/296 [11 . 8%] ) and S . haematobium ( 19/323 [5 . 9%] ) . Patients on CTX-P had a reduced risk of Loa loa microfilaremia ( adjusted odds ratio ( aOR ) 0 . 47 , 95% CI 0 . 23-0 . 97 , P = 0 . 04 ) , also in the subgroup of patients on ART ( aOR 0 . 36 , 95% CI 0 . 13-0 . 96 , P = 0 . 04 ) . There was no effect of ART exposure on helminth infection prevalence . CTX-P use was associated with a decreased risk of Loa loa infection , suggesting an anthelminthic effect of antifolate drugs . No relation between ART use and helminth infections was established .
Globally , more than 2 billion people are estimated to be infected with soil-transmitted helminths , and the geographical distribution of these infections overlaps considerably with regions of high HIV sero-prevalence [1 , 2] . Helminth infections have been hypothesized to be factors driving the HIV-epidemic in Africa [3 , 4] , which may be due to their effects on the host immune system , as demonstrated by an increased susceptibility to HIV infection and progression to AIDS [3] . However , the immunological interaction between the two infections is complex , and others have found results that are conflicting with this hypothesis [5] . Although treatment of intestinal helminth infections and schistosomiasis is relatively simple and cheap , current options are limited to a few drugs , and emergence of resistance is anticipated [6 , 7] . Currently , no really effective and safe drug is available for the treatment of filariases . Diethylcarbamazine ( DEC ) is only moderately effective and has to be administered under supervision , due to its toxicity . DEC or ivermectin treatment may cause serious adverse reactions due to microfilarial disintegration triggering a cytokine release [8] . This underscores the need for new drugs for the treatment of helminth infections . Data on helminth co-infections in patients receiving ART is scarce , as well as information on possible ART effects on helminth infections . Studies investigated patients on ART and compared them to non-treated or HIV-negative controls , usually looking at all intestinal parasites [9–13] . Although the common theme is that parasite infections are reduced in patients on ART , the underlying mechanisms remain unclear . Naturally , improved cellular immunity is often mentioned to explain these findings [9 , 10] , especially for protozoal infections such as cryptosporidiosis . However , several authors speculate on the contribution of drug effects , both from ART as well as from cotrimoxazole preventive therapy ( CTX-P ) , on the reduction of parasite burden [9 , 11 , 13] . Whilst most studies were cross-sectional or used historic controls , one Ethiopian study in patients with newly diagnosed tuberculosis found a significant reduction in helminth infections over time only in HIV-positive as compared to HIV-negative individuals [11] . Authors speculate on the possible effects of ART or CTX-P to explain these findings . A cross-sectional study in Rwanda found a reduced risk of T . trichiura infection in patients using stavudine in combination with lamivudine and nevirapine ( OR , 0 . 27; 95% CIs , 0 . 10–0 . 76; p<0 . 05 ) compared to those on zidovudine , lamivudine and nevirapine [14] . Unfortunately , these studies were not designed to address a specific ART effect on helminth infections . Anthelminthic drugs of the benzimidazole class appear to act on the mitochondrion , supposedly on its tubulin structure [15] . Interestingly , mitochondrial toxicity seems to be a common adverse effect of many anti-retroviral drugs [16] , with stavudine exhibiting the highest degrees of mitochondrial toxicity [17] . This toxic effect is mediated through multiple mitochondrial pathways including inhibition of gamma DNA polymerase [18] . Whether CTX is effective against helminths is less clear . It consists of sulfamethoxazole and trimethoprim which interfere with folate biosynthesis and metabolism . CTX is active against various bacteria , Pneumocystis jirovecii , several parasites , such as Toxoplasma gondii and Plasmodium spp . [19 , 20] , and the intestinal parasites Isospora belli and Cyclospora cayetanensis [20] . The aims of this study were to systematically assess the prevalence of helminth infection among HIV-patients in Lambaréné , Gabon , and to investigate whether ART or CTX-P use is associated with a reduced risk of helminth infections . We hypothesized ( i ) that ART may be associated with a reduced risk of helminth infection , possibly due to mitochondrial toxicity of the ART on the worms , and that ( ii ) CTX-P may reduce the risk of helminth infection through intervention with the folic acid metabolism of the worms .
Between October 2012 and June 2014 , a cross-sectional study was conducted in Lambaréné , a town of 25 , 000 inhabitants situated within the Central African rainforest area of the Moyen Ogooué province , Gabon . Patients were recruited at the Centre de Traitement Ambulatoire ( CTA ) , which is the main clinic for HIV-care in Lambaréné . At this clinic , patients were followed up every 3 months , with an additional follow-up visit after 2 weeks if a patient was started on ART . ART was initiated if patients had CD4 counts <350 cells/μL or were symptomatic ( World Health Organization ( WHO ) stage 3 or 4 [21] ) . Patients received CTX-P if they had CD4 counts <200 cells/μL , in accordance with the Gabonese national guidelines valid at the time of the study . Cotrimoxazole was stopped after two subsequent measurements of CD4 count above 200 cells/uL . Diagnostics for intestinal parasites were not routinely done . However , patients did receive a single dose of albendazole on a 3 monthly basis . Parasitologic diagnostic procedures were performed at the Centre de Recherches Médicales de Lambaréné ( CERMEL ) . The study design capitalized on the limited data available; namely a prevalence of helminth infections of 40–50% in pregnant women and 70% in school-going children [22 , 23] . Filariases and schistosomiasis are endemic in the region [23 , 24] . S . haematobium is endemic in Gabon , whereas S . mansoni is not . All hookworm infections are N . americanus; A . duodenale is not endemic in the study area . HIV seroprevalence is estimated at 3 . 9% in Gabon [25] . Ethical clearance was obtained from the Institutional Research Board of CERMEL . All adult subjects provided written informed consent . Any consenting HIV-infected adult ( age >18 years ) attending the HIV clinic was invited to participate . Patients were divided into two groups: ART naïve , or taking ART for at least 3 months . Patients who started ART within 3 months before recruitment were excluded , to avoid potential effects from previous ART and CTX-P on helminths . Inclusion rates and CD4 counts of patients not enrolled were documented to avoid selection bias . Sample size was calculated based on a power of 80% , and an estimated 25% difference in prevalence between respective groups . We assumed that 65% of patients in the non-exposed group would be ‘helminth-infected’ , based on available data [22–24] . The calculated sample size was 125 patients . However , an interim analysis revealed lower helminth infection prevalences ( ART naïve 32% , on ART 19% ) . The sample size was amended accordingly to 352 patients , with some additional patients to compensate for patients submitting incomplete samples . Basic demographic data ( age , sex , salary , profession , educational level , residence , pregnancy ) were obtained and clinical data were collected using patient files ( ART and CTX-P history , WHO stage [21] , history of opportunistic infections , anthelminthic treatment ) and laboratory registers from the HIV clinic ( CD4 counts and hemoglobin ) . Three stool and urine samples were collected on consecutive days . Where possible , patients submitted stool samples on the day of collection . However , as many patients were living far away from the clinic , we used a maximum time period for acceptance of samples of 24 hours . Stool samples were examined for presence of eggs in smears prepared by the Kato Katz method [26] . Stool samples were analysed for the presence of larvae of S . stercoralis and N . americanus using the modified agar-plate culture technique [27] . Stool was incubated at 25°C on an agar plate . The culture supernatant was checked for larvae with microscopy after 3 and 7 days . Urine was filtrated and analysed for presence of eggs by microscopy . In addition , 2 mL of blood was collected by venipuncture into EDTA around noon ( 11 am—1 pm ) . EDTA blood was analysed by direct microscopy . In addition , red blood cells were lysed using Saponin solution , after which cells were centrifuged . The pellet was analysed for presence of microfilariae by microscopy . If microfilariae were present , samples were stained with methylene blue to identify Loa loa or M . perstans [28] . All positive samples were confirmed by a senior laboratory technician . Outcome was helminth infection , defined as at least one positive stool , urine or blood sample for intestinal helminths ( T . trichiura , A . lumbricoides , N . americanus , S . stercoralis ) , S . haematobium or microfilaria ( Loa loa or M . perstans ) . Factors assessed for their association with helminth infection prevalence were ART and CTX-P use . As pre-defined risk factors were considered age , sex , educational level and rural versus urban residence . The following factors were considered as potential confounders: income , pregnancy , WHO stage , body mass index ( BMI ) , CD4 count , hemoglobin and use of anthelminthic treatment <12 weeks prior to participation . The outcome for the primary analysis was the diagnosis of any helminth infection . Patients were included if they provided at least 2 stool and urine samples and one blood sample . Secondly , sub-analyses were done for different groups of helminth infections; intestinal helminths ( T . trichiura , A . lumbricoides , N . americanus or S . stercoralis ) ; S . haematobium; and the filariae Loa loa and M . perstans . Patients were included for sub-analyses if they provided at least 2 stool samples ( intestinal helminths ) , 2 urine samples ( S . haematobium ) or 1 blood sample ( filariases ) . The distribution of pre-defined risk factors , potential confounders and exposure to ART or CTX-P of patients who were diagnosed with any helminth infection were compared to those who were not infected . We used the χ² test for categorical data , the Student’s t-test for normally distributed continuous data and the Mann-Whitney-U test for non-normally distributed data . Data were assessed for completeness . If for a certain factor , >10% of data were missing , patient characteristics for the group with missing data were compared to those with complete data . Multivariable logistic regression analysis was used to assess the odds of helminth infection associated with the main exposures ( ART and CTX-P ) . Odds ratios were adjusted for pre-defined risk factors . Potential confounders were assessed for their interaction with the main exposures ( ART and CTX-P ) . Factors causing an odds ratio change of >10% were considered as confounders and included in the final model , avoiding multi-collinearity . CD4 count was excluded from the primary analysis because it is a key parameter in decision-making on ART initiation . CTX-P use is inevitably linked with ART use ( patients on CTX-P will as well qualify for ART , unless ART naïve and treated for another opportunistic infection before starting ART ) . Therefore , we performed a sub-group analysis including only patients on ART to assess the odds of helminth infections associated with CTX-P exposure . In this same sub-group , the median time on ART was compared between patients diagnosed with helminth infections and those without using the Mann Whitney U test . We also included the time on ART in the multivariate analysis . Analyses were done using SPSS Statistics Version 21 ( IBM , Chicago , IL , USA ) .
A total of 803 patients were screened and 408 patients were recruited ( Fig 1 ) . Two-hundred fifty-two patients submitted at least 2 stool and urine samples and one blood sample , and were included in the main analysis . Baseline characteristics of the study population are given in Table 1 . The mean age was 41 years ( standard deviation ( SD ) 12 yrs ) ; the majority was female . The majority of patients lived in a semi-urban setting , and had a highest educational level of secondary school . The median CD4 count at the time of study participation was 356 cells/μL ( interquartile range [IQR] 186–526 ) . Around 60% of patients had been on ART for at least 3 months . Almost half the study population was receiving CTX-P . For most determinants , completeness of data was >90% . However , there was >10% missing data for hemoglobin ( 92/252 , 36% ) , the self-reported use of anthelminthic treatment ( 101/252 , 40 . 1% ) , and use of CTX-P ( 30/252 , 11 . 9% ) . Patient characteristics for missing data are given in S1 Table . ART use was reported less often for patients with missing data for all 3 variables . Patients with missing data on the use of CTX-P were more likely to have any infection and to have an infection with intestinal helminths . Fig 2 displays the prevalence of helminth infections . The overall prevalence was 77/252 ( 30 . 5% ) . Sub-analyses showed that filariases were most prevalent ( 56/310 , 18 . 1% ) ; 19/323 ( 5 . 9% ) individuals were infected with S . haematobium; 35/296 ( 11 . 8% ) subjects carried one or more intestinal helminths . Patients carrying one or more helminths were more frequently male , lived in rural areas and were less educated ( Table 1 ) . There were no differences in income , WHO stage , pregnancy or BMI ( S2 Table ) . Patient characteristics for patients infected with intestinal helminths or Loa loa versus those not infected are given in S3 Table . Patients diagnosed with intestinal helminths were more often living in rural areas as compared to those with no diagnosis of intestinal helminths . Loa loa , the most prevalent infection in this cohort , was found more frequently in male patients , and ART and CTX-P use were reported more often in the non-infected patient group . In multivariable logistic regression , there was no evidence of an association between ART use and the risk of having any infection , infection with intestinal helminths , or Loa loa ( Table 2 ) . In contrast , CTX-P was associated with a reduced risk of Loa loa microfilaremia , in the whole population as well as in the subgroup analysis of patients on ART . Female sex was associated with a 3-fold decreased risk of any infection and loiasis . Lower education level was associated with a 2 . 5-fold increased risk of having any infection , and rural residence was associated with an almost 4-fold increased risk of infection with one or more intestinal helminths . There was no evidence of an association between time on ART and the risk of having any infection , infection with intestinal helminths , or Loa loa ( Table 2 ) . Also , there was no difference in median time on ART of patients diagnosed with one or more helminth infections compared to those not infected ( S4 Table ) .
This study assessed the prevalence of helminth infections in HIV-infected adults in Lambaréné , Gabon , and the association of ART and CTX-P with the prevalence of helminth infections . ART use did not alter the risk of harbouring any of the studied infections; intestinal helminths ( T . trichiura , A . lumbricoides , N . americanus , S . stercoralis ) , S . haematobium , or microfilaria ( Loa loa , M . perstans ) . In contrast , CPT use reduced the risk of Loa loa microfilaremia . The overall infection prevalence was 30 . 5% , which compares favorably to earlier studies in the same region [22–24] , yet is well in line with similar studies from other settings [9–13] . The recruitment of individuals who may have a different risk for helminth infections , like pregnant women or school children , hampered direct study comparisons . Furthermore , many studies investigated only intestinal parasites but included protozoa [9–14] . In this study , systemic helminth infections were also investigated , with filarial infections being most prevalent ( 18 . 1% ) followed by S . haematobium ( 5 . 9% ) . Intestinal helminth infections were found in only 11 . 8% , in line with infection rates observed in other studies [13 , 14] . However , higher rates for A . lumbricoides and S . stercoralis have been reported , especially in the pre-ART era [9] . This study found microfilaremia more frequently in males . This finding is not new [24 , 29] . One possible explanation is behavioral factors , such as male patients working in the forest and therefore being more exposed during day time [24 , 29] . Animal models have shown potential effects of gonadal hormones [30] . Rural residence is a known risk factor for helminth infection [31 , 32] and was associated with infections in this study . In general , factors associated with poverty are associated with an increased risk for helminth infection [31 , 32] . CD4 counts were not correlated with the risk of harboring helminth infections . No significant difference in CD4 count distribution was observed between helminth-infected and uninfected patients ( Table 1 ) . Only one hyper-infection syndrome with S . stercoralis was encountered in a HTLV-1 co-infected patient [33] . Data on the interplay of immune depression due to HIV and helminth infections are conflicting . A recent study reported higher co-infection risk with higher CD4 counts [31] . In agreement with this , a negative association of CD4 counts and risk of helminth infection was reported in HIV-infected patients in Uganda [34] . However , conflicting data have been reported from other settings [32] . None of these studies documented CTX-P as a potential confounder or causal factor for the altered prevalence of helminth infection in patients with lower CD4 counts . Although several studies addressed the effect of deworming on HIV viral load and CD4 counts as reviewed elsewhere [35 , 36] , little is known about the opposite effect of HIV-related treatments such as ART and CTX-P on helminths . In this study , ART use did not alter the risk of any helminth infections . However , although ART was not associated with a risk reduction of helminth infections when adjusted to pre-defined risk factors and confounders , the rather low number of infections may have masked a more discrete effect . In contrast , CTX-P reduced the risk of Loa loa microfilaremia in the whole population as well as in the subgroup of patients on ART ( Table 2 ) . Antifilarial effects of dihydrofolate reductase ( DHFR ) inhibitors seem to exist for lymphatic filariae [37] , but no data are available for Loa loa and M . perstans . The present study suggests CTX-P may have antifilarial effects on Loa loa microfilaria , most likely by acting as a DHFR inhibitor thereby inducing apoptosis of the microfilaria [37 , 38] . The low number of cases of M . perstans and S . haematobium infection did not allow assessment of this association , and further exploration seems warranted . CTX-P might be a safer alternative compared to the currently available agents , DEC and ivermectin , either as curative agent , or used for reducing worm burden prior to administration of one of the classical drugs to reduce the risk of adverse reactions . Interestingly , a potentially reduced risk of infection with intestinal helminths in patients receiving CTX-P was not observed . One explanation could be the lower trimethoprim affinity to intestinal helminths’ DHFR , while lower bioavailability and drug concentrations at the site of intestinal worm infection may be alternative explanations . In fact , CTX is very well absorbed and mainly excreted in urine [39] and thus levels in the gut may be too low to be effective against helminths . However , CTX is used successfully to treat several intestinal infections , caused by bacteria and protozoa . They might be more sensitive even to lower concentrations , considering that their DHFR has a much higher affinity to trimethoprim than intestinal helminths . This study has several strengths . To increase specificity of the parasitologic diagnosis , at least 2 negative stool or urine samples were necessary to qualify as a negative test result . The Saponin method for detection of microfilariae has been found superior to a thick blood smear [40] . However , the absence of microfilaraemia does not exclude infection with Loa loa or M . perstans . Previously reported risk factors were found also in this study , thereby assuring the external validity of the data reported . However , there are also limitations . Although recommended by the WHO , the sensitivity of the Kato-Katz method is not optimal [41] . The modified agar-plate culture method is relatively sensitive for detection of S . stercoralis larvae , but less so for hookworm larvae [42] . It should be noted that not all studies use the same methodology for the detection of parasites in stools , which may make direct comparisons difficult , especially at low concentrations of eggs . By using more sensitive methods , possibly more cases could have been identified , although it appears doubtful that this would have changed the overall results . Secondly , not all patients submitted sufficient samples to be included in the final analysis . Therefore , the statistical power to detect certain associations was lower than expected even though the calculated sample size was reached . The use of anthelminthic treatment was not directly observed and self-reported data available on its use during the 12 weeks prior to study participation were limited . There was no clearly reduced infection prevalence in patients who reported prior anthelminthic treatment . This could be explained by the fact that the anthelminthic treatment which was administered at the time of the study was single-dose albendazole , of which the efficacy has been shown to be limited in this setting [43] . Recall bias may have played a role in data availability on self-reported anthelminthic treatment . Prevalences of helminth infections did not differ among patients with incomplete data on the use of antihelminth treatment versus those with complete data , although they were more likely to have certain risk factors for infection ( S1 Table ) . Drugs with less mitochondrial toxicity might have been used in the treatment scheme , thus masking any anti-helminthic effect of ART based on mitochondrial toxicity . The design of this study did not permit assessment of associations of helminth infection with different antiretroviral drugs . In the absence of randomisation between treatment regimens , analyses of treatment effects in observational data are possible using causal methods such as inverse probability weighting , for which our sample was too small . Also , we did not collect data on compliance to ART and cotrimoxazole . Lack of compliance in certain patient groups may have masked anti-helminthic effects of both ART and cotrimoxazole . There were no reliable data available on the duration of CTX-P . Therefore a sensitivity analysis of the effect of the duration of CTX-P on the risk of helminth infections was not possible . In conclusion , the main findings of this study are a high prevalence of Loa loa microfilaremia in HIV-patients in Gabon , and a decreased risk of Loa loa infection in patients using CTX-P , suggesting an anthelminthic effect of antifolate drugs . No association of ART use and helminth infections was established . Additional studies are needed to further assess the effects of CTX on blood-dwelling microfilariae and other helminths , as this might be a safe , cheap and effective alternative to ( few ) existing treatment options . | The geographical distribution of helminth infections , which are highly prevalent in many areas , overlaps considerably with regions of high HIV sero-prevalence . The highest burden of infection is found in resource-poor settings , making it unattractive for the pharmaceutical industry to invest . Limited available treatment options and drug-resistance are increasing problems for soil-transmitted helminths , whereas for some other helminth infections , such as for the blood-dwelling microfilariae , effective and safe treatment options are still far from being optimal . Limited evidence suggests antihelminthic effects of antiretroviral therapy ( ART ) in HIV-infected individuals . We aimed to investigate whether ART or cotrimoxazole preventive treatment ( CTX-P ) reduces prevalence of helminth infection in HIV-infected individuals attending a primary HIV clinic in a semi-rural area in Gabon . The most important finding of our study was that the use of CTX-P was associated with a reduced prevalence of Loa loa microfilaremia . ART use was not associated with a reduced prevalence of helminth infections . Additional studies are needed to assess the effects of CTX on helminth infections , as this might be a promising safe and effective drug adding to the limited repertoire of anthelminthic drugs . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Impact of Anti-Retroviral Treatment and Cotrimoxazole Prophylaxis on Helminth Infections in HIV-Infected Patients in Lambaréné, Gabon |
Mouse Zfy1 and Zfy2 encode zinc finger transcription factors that map to the short arm of the Y chromosome ( Yp ) . They have previously been shown to promote meiotic quality control during pachytene ( Zfy1 and Zfy2 ) and at the first meiotic metaphase ( Zfy2 ) . However , from these previous studies additional roles for genes encoded on Yp during meiotic progression were inferred . In order to identify these genes and investigate their function in later stages of meiosis , we created three models with diminishing Yp and Zfy gene complements ( but lacking the Y-long-arm ) . Since the Y-long-arm mediates pairing and exchange with the X via their pseudoautosomal regions ( PARs ) we added a minute PAR-bearing X chromosome derivative to enable formation of a sex bivalent , thus avoiding Zfy2-mediated meiotic metaphase I ( MI ) checkpoint responses to the unpaired ( univalent ) X chromosome . Using these models we obtained definitive evidence that genetic information on Yp promotes meiosis II , and by transgene addition identified Zfy1 and Zfy2 as the genes responsible . Zfy2 was substantially more effective and proved to have a much more potent transactivation domain than Zfy1 . We previously established that only Zfy2 is required for the robust apoptotic elimination of MI spermatocytes in response to a univalent X; the finding that both genes potentiate meiosis II led us to ask whether there was de novo Zfy1 and Zfy2 transcription in the interphase between meiosis I and meiosis II , and this proved to be the case . X-encoded Zfx was also expressed at this stage and Zfx over-expression also potentiated meiosis II . An interphase between the meiotic divisions is male-specific and we previously hypothesised that this allows meiosis II critical X and Y gene reactivation following sex chromosome silencing in meiotic prophase . The interphase transcription and meiosis II function of Zfx , Zfy1 and Zfy2 validate this hypothesis .
Historically the realisation that there were spermatogenic factors on the human and mouse Y chromosomes distinct from the testis determinant came from the study of Y deletion variants [1] , [2] . However , it was not until the search for the testis determinant that Y-encoded genes began to be identified; amongst these were the human and mouse Y genes encoding zinc finger transcription factors cloned in the late 1980s [3]–[5] . Subsequent progress in assigning spermatogenic gene functions to mouse Y-encoded genes was thwarted by a failure to disrupt Y gene functions using the emerging gene targeting techniques that had proved successful in disrupting X and autosomal gene functions , compounded by the paucity of genomic sequence data for the mouse Y chromosome . To circumvent these problems the Mitchell and Burgoyne labs established a collaboration with the aim of identifying mouse Y gene functions using a Y ‘transgene rescue’ strategy whereby Y genes were added to Y deletion variants with defined spermatogenic failure . In the context of Y genes mapping to the short arm ( Yp ) , three XO male mouse models with diminishing Yp gene complements were utilised ( Figure 1 ) : XSxraO in which the X carries the Yp-derived sex-reversal factor Tp ( Y ) 1CtSxr-a that provides an almost complete Yp gene complement [6] , XSxrbO males where the X carries an Sxra derivative Tp ( Y ) 1CtSxr-b in which a 1 . 3Mb deletion ( ΔSxr-b ) has removed the majority of the Yp gene complement [6] , [7] , and XOSry males in which the only Yp gene present is an autosomally located Sry transgene [8] . The latter two Yp-deficient models have a marked block in spermatogonial proliferation , and in 2001 we reported that this block could be circumvented by the addition of Eif2s3y; this Y-linked gene encodes a protein almost identical to that encoded by the X-linked gene Eif2s3x - a subunit of the essential translation initiation factor EIF2 [8] . Paradoxically , in both Eif2s3y rescue models the majority of spermatocytes complete meiosis I , whereas in the XSxraO ‘control’ there is a very efficient apoptotic elimination of spermatocytes at the first meiotic metaphase ( MI ) [9]–[11]; this apoptosis is assumed to be triggered by an MI spindle assembly checkpoint ( SAC ) response to the univalent X at MI [12] . This suggested that a Yp gene that was deleted or inactivated in Sxrb was necessary for an efficient apoptotic response to the univalent X , although a markedly reduced apoptotic response remained . To identify the Yp gene that promoted the MI spermatocyte apoptosis , transgenes were tested by adding them to XOSry males that carried an X-linked Eif2s3y transgene ( here denoted as XEOSry ) , but none of Yp genes completely removed by ΔSxr-b ( Figure 1C ) reinstated the apoptotic response . Focus then shifted onto Zfy1 and Zfy2 because the ΔSxr-b deletion breakpoints lie within these two genes , creating a transcribed Zfy2/Zfy1 fusion gene with the encoded protein almost identical to that encoded by Zfy1 [7] , [13] . Introducing an X-linked Zfy2 transgene into XEOSry males reinstated the apoptotic response but addition of Zfy1 had no discernible effect [11] . Further studies of the Eif2s3y rescue models XEOSry and XESxrbO revealed that although most primary spermatocytes evaded the apoptotic response and completed meiosis I to form diploid secondary spermatocytes that entered interphase ( “interphasic secondary spermatocytes” ) , very few secondary spermatocytes recondensed their chromosomes and underwent meiosis II [14] . We can envisage three factors that individually , or in combination , could be responsible for the meiosis II impairment: ( 1 ) The triggering of the MI SAC by the univalent X , ( 2 ) the reduced apoptotic response , and ( 3 ) the lack of a Yp gene or genes that promotes meiosis II . It is assumed that the apoptotic elimination is in some way a consequence of the prior triggering of the MI SAC [12] , but with as yet no information on the molecular link between Zfy2 expression and the apoptotic response , factors ( 1 ) and ( 2 ) are confounded . We therefore sought to check for a Yp gene requirement in a situation where the MI SAC and apoptotic response are circumvented . We have previously shown that the apoptotic elimination of MI spermatocytes in XSxraO males can largely be circumvented by adding a minute X chromosome derivative ( denoted Y*X for historical reasons ) comprising a complete PAR , an X PAR boundary , a very short X-specific region and an X centromere [15]–[19] ( Figure 1E ) . In the majority of MI spermatocytes this Y*X mini-chromosome and XSxra had formed a sex bivalent ( indicative of prior PAR synapsis and crossing over ) and thus evaded the MI SAC/apoptotic elimination [19] . In the present study we therefore added this chromosome to the XEOSry , XESxrbO and XSxraO models in order to assess if the near complete Yp gene complement of Sxra promotes meiosis II more effectively than the two depleted Yp gene complements . This proved to be the case so we then proceeded to use Yp transgene addition to identify the Yp genes responsible for meiosis II completion .
We will abbreviate the three models with Y*X used in this study as XY*XSxra , XEY*XSxrb and XEY*XSry . Their Yp gene complements are as shown in Figure 1B–D , except for the addition of the Eif2s3y transgene to the X of the latter two models ( denoted XE ) . For comparison with the published data on ploidy frequency of post-meiotic cells in the XESxrbO and XEOSry models [14] we have processed the Y*X complemented males at 6 weeks of age . We used a combination of centromere ( CREST ) and chromosome axial element ( SYCP3 ) immunostaining , which allows PAR-PAR synapsis to be distinguished from associations between the X-derived Y*X centromere and the X centromere ( Figure 2 ) . This revealed an average of 74 . 9% PAR-PAR synapsis with no significant difference between the three Y*X complemented models and the remainder of cells either having the X and Y*X PARs unpaired , or lacking an identifiable Y*X ( Table 1 ) . In the latter case it is likely that the tiny Y*X chromosome was lost during cell spreading . To assess the efficiency of the meiotic divisions , we analyzed the ploidy of spermatids in SYCP3 and DAPI-stained spermatogenic cell spreads . The DAPI nuclear morphology of diploid spermatids is indistinguishable from that of interphasic secondary spermatocytes , but the latter have a characteristic SYCP3 staining pattern [11] and were excluded . It is important to bear in mind when assessing the consequences of the Y*X additions that in an average of 25 . 1% of MI cells from all three models the X fails to achieve PAR synapsis with the Y*X ( Table 1 ) ; these cells will be subject to efficient apoptotic elimination when Zfy2 is present , but not when Zfy2 is absent . Our strategy was therefore to adjust the haploid frequencies of all the Zfy2-negative males carrying Y*X by ‘removing’ the products of the 25 . 1% of MI cells that did not achieve PAR-PAR synapsis ( see Table S1 ) . The adjusted frequencies are presented in Figure 3; the unadjusted frequencies are available in Table S2 . Strikingly , there was no significant increase in haploid frequency in XEY*XSry ( 17 . 4% ) relative to XEOSry ( 11 . 4% ) ; in marked contrast the haploid frequency had significantly increased ( P = 0 . 00059 ) in XEY*XSxrb ( 54 . 4% ) relative to XESxrbO ( 5 . 2% ) ( Figure 3A ) . This was an unexpected result because there was no indication from the two XO models that Sxrb potentiated meiosis II; indeed XESxrbO had a lower haploid spermatid frequency than XEOSry . We conclude that meiosis II is not potentiated by the formation of a sex bivalent per se , but there is genetic information in Sxrb that in the context of a sex bivalent promotes the completion of meiosis II . Sxrb has a very depleted Yp gene complement so we next wanted to assess the consequences of the Y*X addition in the context of Sxra , which provides a near complete Yp gene complement . This proved to have a much more potent effect than in the Sxrb context with the haploid frequency increasing to 96% ( P = 0 . 00082 ) ( Figure 3A ) . We conclude that there is genetic information on mouse Yp that promotes meiosis II when a sex bivalent is formed , and this is provided more effectively by Sxra than Sxrb . The protein-coding gene content of Sxrb is thought to be limited to a few copies of Rbmy , two copies of H2al2y , Sry and a Zfy2/1 fusion gene spanning the Sxrb deletion breakpoint [7] , [20] , [21] . Because interphasic secondary spermatocytes are a very transient cell type in normal testes , there is no published information on expression of these genes at this stage . However , by RNA in situ analysis Rbmy transcripts are not detected beyond early pachytene [22] and H2al2y does not appear until step 6 round spermatids ( Figure S1 ) , so they are unlikely to be transcribed in interphasic secondary spermatocytes . The Sry transcripts present in the adult mouse testis are circular transcripts that are thought to be untranslated [23] , [24] . Our initial focus was therefore on the Zfy2/1 fusion gene , which is known to be transcribed during early prophase , is presumed to be silenced by meiotic sex chromosome inactivation ( MSCI , [25] ) at the beginning of pachytene , as are Zfy1 and Zfy2 in normal males , but is transcribed post-meiotically [7] , [26] . The Sxrb deletion breakpoint is located within a 95 bp region of sequence identity between intron 5 of Zfy2 and Zfy1 , and the protein encoded by the fusion gene is predicted to be identical to that encoded by Zfy1 except for the 16th amino acid where a leucine replaces a phenylalanine [7] , [11] . Because the Zfy2/1 fusion gene encodes a protein nearly identical to that of Zfy1 we first added a Zfy1 transgene to XEY*XSry to see if this mimicked the effect of Sxrb in promoting the second meiotic division in the presence of Y*X . This proved to be the case in that the proportion of haploid spermatids increased significantly ( P = 0 . 01219 ) from 17 . 4% to 47 . 2% ( Figure 3B ) . Based on their DNA sequences , Zfy1 and Zfy2 are expected to produce transcription factors that will bind to the same target genes . We therefore also generated XEY*XSry males that were transgenic for Zfy2 , and this addition increased the haploid frequency from 17 . 4% to 78 . 2% ( P = 0 . 00011 ) , which is significantly higher ( P = 0 . 00665 ) than that achieved with the Zfy1 transgene ( 47 . 2% ) . Thus the Zfy2 transgene promotes meiosis II more effectively than the Zfy1 transgene . Both transgenes are single copy and inserted on the X chromosome , but we cannot assess relative transcript levels in interphasic secondary spermatocytes because of our inability to adequately purify this rare cell type . However we have previously established by qRT-PCR that the transcript level for the Zfy1 transgene is higher than that for the Zfy2 transgene in testes from 17 . 5 day-old XEOSry carriers [11] , so we would expect a similar excess of Zfy1 transcripts in interphasic secondary spermatocytes . We were therefore surprised that the Zfy2 transgene had a markedly greater effect . With the addition of both transgenes the frequency of haploid spermatids increased to 87 . 6% ( Figure 3B; Table S2C ) . These results point to the combined activity of Zfy1 and Zfy2 as important for promoting meiosis II . Transcription of Zfy1 and Zfy2 is reportedly testis specific , at least post-natally [27]–[29] . Recently we have shown that in the adult testis this transcription is limited to germ cells , starting in leptotene spermatocytes , with more robust transcription in zygotene spermatocytes , followed by silencing in pachytene spermatocytes as a consequence MSCI; there was no resumption of transcription prior to MI , but transcription was shown to have resumed in ( Y-bearing ) round spermatids [7] . However , no data are available for interphasic secondary spermatocytes [14] . To assess transcription of Zfy1 and Zfy2 in these cells we used Zfy1 and Zfy2 RNA-FISH on DAPI- and SYCP3-stained testis cell spreads from XY males and confirmed the presence of the Y chromosome using Zfy1 and Zfy2 DNA-FISH . Zfy1 and Zfy2 transcription was detected in 45% and 27% , respectively , of the Y-bearing secondary spermatocytes ( Figure 4; Table 2 ) . As expected , Zfy1 and Zfy2 DNA-FISH signals were not observed in half of the secondary spermatocytes ( X-bearing ) and these also lacked Zfy1 and Zfy2 RNA-FISH signals . However , 92% of these X-bearing secondary spermatocytes were transcribing the related X-linked gene Zfx ( Figure 4; Table 2 ) . Our finding that Zfx is also expressed in interphasic secondary spermatocytes raised the question as to whether Zfx also promotes the second meiotic division . We had available a Zfx transgenic line with 7 copies of a Zfx genomic BAC inserted on an autosome . As expected for an autosomally located X-chromosome-derived transgene it was exempt from MSCI , and was expressed in pachytene cells ( Figure S2A , B ) ; like the endogenous Zfx gene it was expressed in interphasic secondary spermatocytes ( Figure S2C ) . We added this transgene to XEY*XSry males and the proportion of haploid spermatids increased from 15 . 0% ( in non-transgenic siblings ) to 79 . 9% , showing that Zfx can promote meiosis II ( Figure 5A , B; Table S2D ) ; we therefore consider it likely that the endogenous Zfx also has a minor role in promoting meiosis II . We were struck by the much more potent effect of Zfy2 as compared to Zfy1 in promoting meiosis II . Based on an in vitro assay it was previously reported that Zfy2 encodes a protein with a much more potent transactivation ( TA ) domain than that of Zfx , but Zfy1 was not assayed at that time [30] . We have therefore used a similar in vitro assay to compare the transactivation domains of mouse Zfx , Zfy1 , Zfy2 and the autosomal Zfa ( originating from a retroposed X transcript [31]–[35] ) , and have also compared these with the transactivation domains of human ZFX and ZFY ( Figure 6 ) . We confirmed the expression of all the ZF-Gal4 fusion proteins by western blot analysis ( Figure S3 ) . The assay revealed that the TA domain of mouse Zfy1 has a similar activity to human ZFX and ZFY . Strikingly , the mouse Zfy2 TA domain is 5 . 5-fold more active than that of mouse Zfy1 , and is ∼10-fold more active than that of mouse Zfx . The TA domain of the putative ZFA protein proved to have a very weak TA activity . A single nucleotide deletion near the beginning of the ZFY/ZFX open reading frame of Zfa actually makes it very unlikely to translate a protein that includes the zinc finger DNA binding domain , which would preclude binding to target genes . Zfa is now flagged as a pseudogene in Genbank ( accession no . NR_037920 ) .
Our previous study of meiotic progression in the three XO male models with varying Yp gene complements revealed that the majority of spermatocytes in the Yp gene deficient models XEOSry and XESxrbO reached the interphase that precedes meiosis II; however , they failed to recondense their chromosomes to enable completion of meiosis II and instead formed diploid round spermatids [14] . Formally , the failure to undergo meiosis II could be a consequence of the prior triggering of the MI SAC by the univalent X , the reduced apoptotic response due to the absence of Zfy2 , or the lack of a Yp gene or genes that promotes meiosis II ( Figure 7A ) . The aim of the present study was to check specifically for a Yp gene requirement by circumventing the MI SAC and apoptotic responses; for this we added a minute PAR-bearing X chromosome derivative ( Y*X ) to all three XO models to enable formation of a sex bivalent without altering the Yp gene complement . This established that there is genetic information present in Sxra and Sxrb that promotes meiosis II , but Sxra was more effective ( Figure 7B ) . Yp transgene additions to XEY*XSry males then identified Zfy1 and Zfy2 as the Yp genes responsible with Zfy2 having the more potent effect ( Figure 7C ) . We attribute the difference in potency between Sxra and Sxrb to the presence of Zfy1 and Zfy2 in Sxra whereas Sxrb only has the Zfy2/Zfy1 fusion gene that encodes a protein almost identical to Zfy1 [7] , [11] ) . However , we were struck by the fact that Sxrb did not promote meiosis II in the absence of Y*X; this implies that the triggering of the MI SAC , and/or the reduced apoptotic response , impairs progression through meiosis II . In order to distinguish between these possibilities it is informative to consider what happens in XO females where there is an MI SAC response but no apoptotic response . XO female mice are fertile and produce XO ( and XX ) daughters , so that some XO oocytes must complete meiosis I and meiosis II . Furthermore , although some X univalents achieve bipolar attachment to the spindle ( which is expected to satisfy the MI SAC ) , this is not a prerequisite for the completion of meiosis I [36] . This is in agreement with accumulating data for female mice showing that the MI SAC does not maintain arrest until all kinetochores have achieved appropriate attachments to the spindle - anaphase can proceed in the presence of one ( or a few ) univalents [37]–[41] . Thus XO oocytes can complete meiosis I to generate MII oocytes with either an XX sex chromosome complement ( i . e . two X chromatids ) or lacking an X chromosome , both of which should be able to complete meiosis II without triggering an MII SAC response . This suggests that in the three XO male models the triggering of the MI SAC per se would not impair meiosis II . We therefore favour the view that the addition of the Y*X to the XESxrbO model increases the haploid spermatid frequency from 5 . 2% to 54 . 4% because the formation of an XSxrb/Y*X bivalent avoids the reduced apoptotic response . We envisage that the reduced DNA damage at MI as a consequence of the reduced apoptotic response is usually insufficient to trigger elimination at MI , but is sufficient to trigger a G2/M DNA damage checkpoint ( reviewed in [42] ) at the post meiosis I interphase and block progression to MII . The arrested interphase cells then enter spermiogenesis as diploid spermatids . Unfortunately we cannot use these models to assess the ultimate fate of the diploid spermatids , because in the absence of the Y long arm there is marked over-expression of X and Y genes due to the absence of the repressive effects of Sly , which is present in >50 copies on the Y long arm [43] , [44] , and this ( together with Yp gene deficiency ) results in severely perturbed spermiogenesis [14] . Figure 8 summarises how we see these MI and G2/MII checkpoint responses operating in males with a normal ‘Zf’ gene complement ( XSxraO , XY*XSxra and XY ) . The finding that both Zfy2 and Zfy1 promoted meiosis II was surprising because at MI only Zfy2 promotes the robust apoptotic elimination of spermatocytes with a univalent X chromosome [11] . What is the basis for the resurrection of Zfy1 function in the short interval between MI and meiosis II ? Our previous RNA FISH analyses of nascent nuclear transcripts on spread spermatogenic cells from normal XY males [7] have established that Zfy1 and Zfy2 are transcribed in all mid-late zygotene nuclei , but this ceased in pachytene nuclei – an expected consequence of meiotic sex chromosome inactivation ( MSCI – reviewed by [25] ) , and remained undetectable right through MI . The role of Zfy2 in the apoptotic response to univalence at MI must therefore be a consequence of transcriptional changes mediated by ZFY2 translated from these zygotene transcripts . Zfy1 and Zfy2 have the same predicted DNA target sequences , so if both are robustly expressed during zygotene , why is the apoptotic role limited to Zfy2 ? A plausible explanation is provided by our finding that during the pre-pachytene phase of transcription , alternative splicing of Zfy transcripts leads to ∼81% of Zfy1 transcripts lacking exon 6 with the encoded protein lacking transactivation ( TA ) activity , whereas ∼96% of Zfy2 transcripts have exon 6 and thus a functional TA domain [7] . Our current TA domain analysis further demonstrates that the few full length Zfy1 transcripts that are produced during zygotene generate a protein with a much less potent TA domain than that of Zfy2 . In view of this it is reasonable to conclude that Zfy1 function in meiosis II is based on the de novo transcription in interphasic secondary spermatocytes ( and that this also applies to Zfy2 ) . It also implies that at this stage there is a greater preponderance of Zfy1 transcripts with exon 6 that encodes the TA domain; this is supported by our previous finding of a 3 . 7-fold increase in such transcripts in pubertal testes between 20dpp and 27dpp , which covers the period when transcripts from interphasic secondary spermatocytes and round spermatids should progressively increase as a proportion of the total testis RNA [7] . Given that Zfx , Zfy1 and Zfy2-encoded transcription factors are predicted to bind the same target sequences , it is to be expected that in tissues where they are all expressed they will regulate the transcription of the same genes . The extent to which they transactivate target genes will be dependent on the relative potency of their TA domains ( Zfx<Zfy1<Zfy2 ) ; the protein encoded by Zfy1 lacking exon 6 would be expected to bind but not transactivate , and could thus function as a competitive inhibitor of the three full length ZF proteins [7] . In XY males , all three sex-linked ‘Zf’ genes are transcribed in zygotene spermatocytes with a predominance of Zfy1 transcripts lacking exon 6 ( Figure S4 and [7] ) , and in interphasic secondary spermatocytes ( Figure 4 and Table 2 ) in which full length Zfy1 transcripts are thought to be more prevalent . Does the Zfx transgene addition support the expectation that Zfx will contribute to ‘Zf’-mediated functions at MI ( indirect ) and during meiosis II ( potentially direct ) ? We have already presented the data showing a marked promotion of meiosis II by the Zfx transgene ( Figure 5 ) and there is also a marked promotion of the apoptotic response to X univalence at MI ( Figure S5 ) , which clearly demonstrates that Zfx is able to contribute to these functions . [The marked promotion of these functions in both cases is unsurprising given that the transgene is present in 7 copies and that its autosomal location is associated with extension of transcription through to just prior to MI , together with exemption from the MSCI-dependent repression that affects the X and Y chromatin of interphasic secondary spermatocytes ( see below ) . ] Thus it is reasonable to conclude that the endogenous Zfx also contributes to these functions; indeed , in the XEOSry model there is some MI apoptosis [11] and in the XEY*XSry model there is some progression through meiosis II ( 17 . 4% haploid spermatids , Figure 3A ) . A role for the endogenous Zfx in spermatogenesis has also been suggested based on the reduced sperm count in Zfx knockout males , although this effect is confounded with severe growth deficiency [45] . Although Monesi reported transcription in interphasic secondary spermatocytes in the 1960s [46] , [47] , other than our finding that there is de novo transcription of the multi-copy mouse Y gene Sly in interphasic secondary spermatocytes [48] , we are not aware of any published data giving information on which genes are actively transcribed at this stage . It has previously been concluded based on Cot1 RNA FISH assessments of global transcription in interphasic secondary spermatocytes that the autosomal chromatin is actively transcribed , whereas the X and Y chromatin remains substantially repressed; the repression of the sex chromosomes has been shown to be dependent on the prior MSCI , and is carried through into round spermatids ( ‘post-meiotic sex chromosome repression’ ) [49]–[51] . This raises the possibility that the de novo transcription of the ‘Zf’ gene family represents a selective reactivation . As a first look at this issue we assessed de novo transcription of Mtm1 ( X-linked ) and Uty ( Y-linked ) in interphasic secondary spermatocytes and round spermatids , with Zfx serving as a positive control . This revealed that these two genes are also transcribed in interphasic secondary spermatocytes , but it is noteworthy that for all three genes the frequency of RNA FISH positive cells was higher in interphasic secondary spermatocytes ( Zfx 90%; Mtm1 67%; Uty 88% ) than in round spermatids ( Zfx 32%; Mtm1 36%; Uty 60% ) ( Table S3 ) . These preliminary data are consistent with: ( 1 ) there being a partial relaxation of sex chromosome silencing during the late diplotene–MI period , counterbalanced by the global transcriptional repression associated with the condensation of the metaphase chromosomes , ( 2 ) the decondensation of the chromosomes in interphasic secondary spermatocytes allowing strong transcription from the autosomes , but weaker transcription from the sex chromosomes because of the MSCI carry-over effect; and ( 3 ) further repression of the sex chromosomes in round spermatids due to the repressive chromatin changes driven by the multi-copy Y gene Sly [44] . The marked increase in transactivation activity of Zfy2 relative to Zfx and Zfy1 ( Figure 6 and [30] ) raises some interesting questions in an evolutionary context . The autosomal ‘Zf’ precursor of Zfy and Zfx is thought to have been added to the PAR after the separation of the eutherian and marsupial lineages 193–186 million years ago , and that with further PAR additions and rearrangements it became located in the non-recombining regions of the X-Y pair [52] , [53] . In eutherian mammals the X-linked genes with retained Y-linked homologues are typically exempt from X dosage compensation , suggesting a constraining dosage requirement in somatic tissues , and in most eutherian mammals this is known or is presumed to be true for Zfx and Zfy [54] , [55] . However , around 40–70 million years ago in the myomorph rodent lineage , Zfx became subject to X-dosage compensation and the Zfy-encoded proteins diverged [30] , [55]–[59] . Furthermore , the divergence in Zfy protein sequence is more marked in the highly acidic amino terminal TA domain that activates target genes , than in the carboxy terminal zinc finger domain that mediates binding to DNA . Here we have shown that in Mus musculus this divergence is associated with increased TA activity and that this is much more marked in Zfy2 than in Zfy1 . Given that in mature male mice expression of the Zfy genes has only been detected in testes [27]–[29] , specifically in the germ-line [7] , this implies that there was a strong selective force in spermatogenic cells for improved TA activity . This male germ-line specific selective force is likely to have been MSCI , which will have affected Zfx as well as Zfy1 and Zfy2 . For a zinc finger transcription factor needed for meiosis II that is dependent on transcription during the brief interphase between meiosis I and meiosis increasing the transactivation activity would be a major advantage . The TA domain of Zfx is likely precluded from responding to the selection because of a dosage sensitive role in somatic cells . On the other hand , the spermatogenic cell specific expression of the Y-encoded genes in the post natal testis allowed their TA domains to increase in activity , but the TA domain of Zfy2 has responded much more than that of Zfy1 . Given the importance of ‘Zf’ gene transcription during the interphase between meiosis I and meiosis II in male meiosis , it is intriguing that female mice ( and female mammals generally ) have no interphase between the two meiotic divisions . We previously hypothesized that the presence of an interphase between the two meiotic divisions in male mammals would be essential if there are meiosis II critical genes on the sex chromosomes , because they would have been transcriptionally silenced ( MSCI ) during the preceding ∼8 days [14] . However , the two meiotic divisions in females are dependent on RNAs produced and stored during oocyte growth [60] , and it may be this dependence on stored RNAs that has enabled female meiosis to dispense with the interphase . In conclusion , our present findings provide evidence for a specific requirement for Zfx and Zfy expression in the interphase between meiosis I and meiosis II , for meiosis II to be efficiently completed . We have also provided additional evidence for a marked divergence in the functionality of the three ‘Zf’-encoded transcription factors , with Zfx providing a dosage constrained somatic role with only a minor contribution to sex-linked ‘Zf’ gene function in spermatogenesis , Zfy1 developing a dual role in spermatogenesis via alternative splicing to produce activatory and repressive proteins ( see also [7] ) , and Zfy2 becoming a super-active transcription factor ( see also [30] ) to enable it to function in the face of the repressive effects of MSCI and the linked post-meiotic sex chromosome repression . There are undoubtedly further sex-linked ‘Zf’ gene functions to be discovered so the identification of the direct targets of the ‘Zf’-encoded transcription factors is a high priority . This has been thwarted by a failure to obtain specific antibodies for chromatin immunoprecipitation analyses , but transgenes encoding tagged versions of the proteins should provide a way forward . We have also presented a case for their being a G2/M DNA damage checkpoint operating in the interphase between meiosis I and meiosis II that prevents progression to MII if there is unrepaired DNA damage present; our XO mouse models together with the recent first report of a successful method for the targeted disruption of a single copy Y gene [61] , will be invaluable for investigating this further . Some intriguing recent data obtained with the XESxrbO model suggested that following the injection of diploid spermatids ( almost certainly together with interphasic secondary spermatocytes ) into eggs ( “ROSI” ) , a proportion of the cells completed the second meiotic division in the egg , thus avoiding triploidy which is lethal in early pregnancy [62] . The egg provides the cellular machinery for DNA damage repair by non-homologous end joining ( NHEJ ) [63] , which would be expected to release the proposed G2/M DNA damage checkpoint arrest in the XESxrbO model . However , this pathway of repair is inherently mutagenic [64] , which may have important ramifications for the use of ROSI with cells harboring such DNA damage if they were unintentionally used when treating human male infertility .
All animal procedures were in accordance with the United Kingdom Animal Scientific Procedures Act 1986 and were subject to local ethical review . Aside from the mice with Sxra or Sxrb attached to the Y*X chromosome ( see section ( 1 ) below ) , the mice in this study have an outbred MF1 ( NIMR colony ) background . The XY*X males with varying Yp gene complements ( Figure 1 ) were produced by either 1 or 2 below , and the Zfy and Zfx transgene additions to XEY*XSry males are described in 3 . Crosses 1 ( i ) -2 ( iii ) above generate XO males as well as the XY*X males with varying Yp complements . As a guide to the presence of Y*X we utilised PCRs for X-linked Prdx4 ( absent in Y*X ) , Amelx ( present in Y*X ) and Myog ( on chromosome 1 ) for normalisation . Two PCR reactions were used to detect the presence of Y*X and the number of X-chromosomes . An 82-bp Prdx4 and a 162-bp Amelx fragment were amplified using primers Prdx4-F and Prdx4-R together with primers Amelx-F and Amelx-R . The 162-bp Amelx fragment and a 246-bp Myog fragment were amplified using Amelx primers together with Myog primers Om1a and Om1b [72] . Primer sequences are described in Table S4 . The following conditions were used: 95°C for 5 min , followed by 28 cycles of 95°C for 30 sec , 60°C for 20 sec and 72°C for 30 sec , with a final extension at 72°C for 5 min . Products were separated on a 3 . 5% ( w/v ) agarose gel and the genotype inferred from the relative intensities of the PCR products: XO 1 Prdx4 + 1 Amelx + 2 Myog , XY*X 1 Prdx4 + 2 Amelx + 2 Myog , XX 2 Prdx4 + 2 Amelx + 2 Myog and XXY*X 2 Prdx4 + 3 Amelx + 2 Myog . For mice typed as Y*X positive that provided material for the present study the presence of the Y*X was confirmed either by examination of Giemsa-stained bone marrow chromosome spreads to check for the presence of the very small Y*X chromosome , by SYCP3 and CENT immunostaining of testis cell spreads ( see below ) , or by quantitative PCR using Prdx4-F , Prdx4-R , Amelx-F and Amelx-R primers with the following genotypes as controls: XX , XO and XY*X . Om1a and Om1b primers were used for normalisation . Copy number estimation was done by quantitative PCR as previously described in Royo et al 2010 with slight modification . A SacBII amplicon obtained using SacBII-F and SacBII-R primers ( Table S4 ) , match the backbone of the Zfx-bearing vector . A XZfy1/Uba1yY sample was used as a reference , because it bears a known transgene copy number of one ( Zfy1-7; [26] ) and the backbone of the Zfy1/Uba1y-bearing vector contains SacBII ORF . Reactions were normalised against amplification of the Atr gene . The difference in PCR cycles with respect to Atr ( ΔCt ) for a given experimental sample was subtracted from the mean ΔCt of the reference samples ( XZfy1/Uba1yY ) ( ΔΔCt ) . The transgene copy number was calculated as the mean of the power 2 ( ΔΔCt ) . Pairing efficiency between X and Y*X was assessed on surface-spread spermatogenic cells preparation from 6-week-old testes . Briefly , a portion of frozen testicular tissue ( approximately 10 mg ) was defrosted and macerated in 0 . 2 ml RPMI 1640 solution ( Invitrogen Corporation , Gibco ) to produce a thin cell suspension . One drop of cell suspension was applied on a pre-boiled microscope slide , mixed with five drops of 4 . 5% sucrose solution and allowed to stand for one hour in a humid chamber at room temperature . The cells were permeabilized by adding three drops of 0 . 05% Triton X-100 solution for 10 min , after which ten drops of 2% formaldehyde solution ( TAAB ) containing 0 . 02% SDS pH 8 . 4 were added for 30 min . The slides were then dipped briefly in distilled water and air-dried . After rehydration in PBS the slides were soaked in PBST-BSA ( PBS containing 0 . 1% Tween 20 and 0 . 15% BSA ) for 1 hour and incubated overnight at 37°C with rabbit polyclonal anti-SYCP3 ( 1∶300; Abcam ) and an anti-centromere ( CREST ) antibody ( 1∶500; Antibodies Inc . ) diluted in PBST-BSA . Slides were washed in PBST , incubated with chicken anti-rabbit Alexa 488 ( 1∶500; Molecular Probes ) and goat anti-human Alexa 594 ( 1∶500; Molecular Probes ) diluted in PBS for 1 h at 37°C and washed in PBST . Pairing efficiency was evaluated on a Leica microscope after staining the cell nuclei with 4′ , 6-diamidino-2-phenylindole ( DAPI ) diluted in the mounting medium ( Vectashield with DAPI; Vector ) . At least four mice per genotype were used and pairing efficiency was assessed for ∼50 pachytene spermatocytes that were identified based on their DAPI nuclear morphology and their full autosomal synapsis identified by the synaptonemal complex ( SYCP3 ) staining pattern . We classified the pairing of the Y*X in three categories: ( i ) clear PAR-PAR pairing of the X-chromosome with the Y*X chromosome , ( ii ) the Y*X chromosome clearly identifiable as a univalent chromosome , and ( iii ) no Y*X chromosome could be identified ( most likely lost during cell spreading ) . Nuclear DNA content was measured on surface-spread spermatogenic cells from 6 week old testes as described previously [11] , [14] using SYCP3 staining and DAPI fluorescence intensity measurements . Antibody against γH2AFX ( 1∶500; Upstate ) was used to identify the sex body of pachytene spermatocytes [73] . RNA-FISH for nascent nuclear transcripts from Zfy1 , Zfy2 and Zfx was performed as previously described [7] , [74] using spread testis cells from adult MF1 male mice . Zfx RNA FISH was also carried out on spread testis cells from XY , Zfx transgenics . The Zfy2-specific probe was BAC CITB-288D7 ( Research Genetics ) , the Zfy1-specific probe was a modified version of BAC RP24-498K8 ( CHORI ) from which we had removed the entire Uba1y gene by recombineering , the Zfx-specific probe was BAC BMQ-372M23 ( CHORI ) , the Mtm-specific probe was BAC RP24-287E17 ( CHORI ) and the Uty-specific probe was BAC CITB-246A22 ( Research Genetics ) . Zfy1 , Zfy2 and Zfx RNA FISH signals were confirmed with DNA FISH as described previously [74] . Antibody against SYCP3 ( 1∶100; Abcam ) was used to identify secondary spermatocytes as previously describe [11] . ZF TA domain-Gal4 fusion-protein constructs were made by inserting cDNA segments encoding the different acidic domains into the NcoI and SalI , or NdeI and SalI , restriction sites of the vector pGBK-CEN6 , a single-copy version of pGBKT7 ( Clontech ) , downstream of the Gal4 DNA-binding domain and the c-myc epitope tag of the vector . pGBK-CEN6 without an insert was included as a negative control . We used a low-copy origin because we had previously noticed that the expression of an acidic domain that strongly transactivates ( Zfy2 and Gal4 ) inhibits yeast growth [7] , and this has been described for the overexpression of Gal4 [75] . Validating our strategy , yeast transformed with the different acidic domain constructs , including Zfy2 , all showed similar growth rates ( as did the Gal4 acidic domain – data not shown ) . To create pGBK-CEN6 , we replaced the 2 µ high-copy origin of pGBKT7 with the ARS4/CEN6 low-copy origin from pDEST22 ( Invitrogen ) , by recombineering in the E . coli strain DY380 [76] . Acidic domains from human ZFY and mouse Zfy1 and Zfy2 were transferred to pGBK-CEN6 from pGBKT7 constructs as described previously [7] . The acidic domains from human ZFX and mouse Zfx were amplified from testis cDNAs and mouse Zfa was amplified from genomic DNA . PCR-amplified inserts were shown to be without error by sequencing recombinants . Primers used were Zfx: o4472/o4109 , with respectively NcoI and SalI adaptors , and ZFX: o4473/o4109 and Zfa: o4471/o4109 , with respectively NdeI and SalI adaptors ( Table S4 ) . One recombinant was selected for each construct and transformed into the S . cerevisiae strain Y187 , in which the β-galactosidase gene is under the control of the Gal4-responsive Gal1 promoter . Three single transformed colonies were picked from SD/-trp agar plates and grown separately in liquid culture to an OD600 of 0 . 9–1 . 26 in SD/-trp liquid minimal medium . The β-galactosidase assay was performed on 1 OD600 unit of the culture using the permeabilized cell assay [77] . For ploidy frequency the differences between genotypes were assessed by one tail student t-test assuming unequal variances after angular transformation of percentages , using Excel ( Microsoft ) software . For the transactivation assay one tail student t-test assuming unequal variances was performed on the β-Galactosidase activity . | The mouse Y chromosome genes Zfy1 and Zfy2 were first identified in the late 1980s during the search for the gene on the Y that triggers male development; they encode proteins that regulate the expression of other genes to which they bind via a ‘zinc finger’ domain . We have now discovered that these genes play important roles during spermatogenesis . Zfy2 proved to be essential for the efficient operation of a ‘checkpoint’ during the first meiotic division that identifies and kills cells that would otherwise produce sperm with an unbalanced chromosome set . Female meiosis , which does not have an equivalent checkpoint , generates a significant proportion of eggs with an unbalanced chromosome set . In the present study we show that Zfy2 also has a major role in ensuring that the second meiotic division occurs , with Zfy1 and a related gene , Zfx , on the X chromosome providing some support . In order to fulfil this function all three genes are expressed in the ‘interphase’ stage between the two divisions . In female meiosis there is no interphase stage between the two meiotic divisions but in this case essential functions during the divisions are supported by stored RNAs , so an interphase is not needed . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"meiosis",
"cell",
"biology",
"cell",
"cycle",
"and",
"cell",
"division",
"genetics",
"biology",
"and",
"life",
"sciences",
"cell",
"processes",
"molecular",
"cell",
"biology",
"gene",
"function"
] | 2014 | Mouse Y-Linked Zfy1 and Zfy2 Are Expressed during the Male-Specific Interphase between Meiosis I and Meiosis II and Promote the 2nd Meiotic Division |
The mismatch negativity ( MMN ) is a differential brain response to violations of learned regularities . It has been used to demonstrate that the brain learns the statistical structure of its environment and predicts future sensory inputs . However , the algorithmic nature of these computations and the underlying neurobiological implementation remain controversial . This article introduces a mathematical framework with which competing ideas about the computational quantities indexed by MMN responses can be formalized and tested against single-trial EEG data . This framework was applied to five major theories of the MMN , comparing their ability to explain trial-by-trial changes in MMN amplitude . Three of these theories ( predictive coding , model adjustment , and novelty detection ) were formalized by linking the MMN to different manifestations of the same computational mechanism: approximate Bayesian inference according to the free-energy principle . We thereby propose a unifying view on three distinct theories of the MMN . The relative plausibility of each theory was assessed against empirical single-trial MMN amplitudes acquired from eight healthy volunteers in a roving oddball experiment . Models based on the free-energy principle provided more plausible explanations of trial-by-trial changes in MMN amplitude than models representing the two more traditional theories ( change detection and adaptation ) . Our results suggest that the MMN reflects approximate Bayesian learning of sensory regularities , and that the MMN-generating process adjusts a probabilistic model of the environment according to prediction errors .
A key theme of contemporary neuroscience is the notion that the brain embodies a generative model of the environment , enabling inference on the causes of sensory inputs and predicting future events . This is also known as the “Bayesian brain hypothesis” ( for reviews , see [1] and [2] ) . This framework provides an abstract explanation of adaptive cognition and behaviour , which has been instantiated in schemes like predictive coding and hierarchical Bayesian message passing [3]–[5] , or , more recently , the free-energy principle [2] . Experimentally , an important paradigm for testing the implications of these theories in humans is the mismatch negativity ( MMN ) paradigm [6] . In this paradigm , electrophysiological methods such as electroencephalography ( EEG ) or magnetoencephalography ( MEG ) are used to measure event-related “mismatch potentials” in response to violations of expectancy or learned regularities . Traditionally , the MMN ( cf . Figure 1 ) is recorded during auditory oddball experiments or , more recently , during “roving” oddball paradigms . It can be defined operationally by subtracting the event-related potential ( ERP ) elicited by standards , i . e . stimuli that are predicted by an established regularity , from the ERP elicited by deviants , i . e . the same stimuli when they violate the regularity . The MMN is usually expressed most strongly at fronto-central electrodes , and its peak latency varies between 100 and 250 milliseconds after deviance onset , depending on the specific paradigm and type of regularity that is violated [7] , [8] . Previous EEG and fMRI studies suggest that the MMN originates from temporal generators ( A1 and STG ) and a prefrontal generator in the inferior frontal gyrus [9] , [10] . A major research theme has been the search for models of the neurophysiological and computational processes that underlie the MMN [7] , [11] , [12] . Such models would contribute to a better understanding of statistical learning in the brain and the prediction of future events . However , the neurocomputational processes that generate the mismatch negativity are still subject to debate [7] , [13]–[15] . Over the years , five major hypotheses have been formulated , which we compare in this article: So far there has been no objective procedure to conclude which MMN theory is best supported by a given dataset , because most theories of the MMN are of a qualitative nature and do not make quantitative predictions . Furthermore , the inferences that could be drawn were limited by the averaging inherent to standard ERP analysis: this destroys any information about the temporal dynamics of learning . The first goal of this study was to overcome both limitations by providing a modelling framework with which competing MMN theories can be formalized and objectively compared against one another by their capacity to explain single-trial MMN amplitudes . Here , the explanandum was not just the mismatch negativity per se , but also how its single-trial amplitude changes as the subject learns statistical regularities during the successive presentation of stimuli . The mismatch response to the same stimulus differs depending on the history of all preceding stimuli , and our models should be able to predict these changes . The ensuing modelling of single-trial MMN amplitudes and their progressive changes represents a novel approach , which emphasizes the sensory learning on which the MMN rests . Two related studies using a similar approach recently suggested that single-trial MMN and P300 amplitudes reflect the trial-wise degree of Bayesian and Shannon surprise , respectively [23] , [24] . Here , we extend this trial-wise approach and formalize the processes postulated by the five MMN theories introduced above in terms of specific process models; these are then subjected to Bayesian model comparison in order to assess how well each of them explains the variability of trial-wise MMN amplitudes . This formulation of detailed and quantitative models representing the 5 major contemporary MMN theories constituted the second goal of this paper . In constructing these models , the third goal was to show that the prediction error , model adjustment , and novelty detection theories of the MMN can be unified . Concretely , we propose that prediction errors , model adjustments and novelty are different manifestations of a common underlying process , namely variational free-energy minimization during perceptual inference and learning [2] . This paper is structured as follows . The Models and Methods section describes our roving oddball experiment , data acquisition and pre-processing , the extraction of the single-trial MMN amplitudes used in the subsequent analysis , as well as our modelling framework and its application to formalizing each of five MMN theories by a model family ( a set of models with a shared essence ) . The two final sections present and discuss the results obtained by fitting the ensuing models to empirical MMN responses and applying Bayesian model comparison to assess the relative plausibility of individual models and MMN theories ( model families ) .
The empirical data used in this study comprised trial-wise mismatch responses , acquired during a roving oddball experiment with electroencephalography ( EEG ) from eight healthy subjects in a previously published study [25] , [26] . Twelve healthy volunteers ( aged 24–34 , 4 female ) listened passively to a structured sequence of 1600 pure sine tones adapted from [27] . Subjects sat in front of a computer screen and were instructed to ignore the tones and press a button whenever there was a change in the luminance of the fixation cross . The structure of the stimulus sequences is illustrated in Figure 1 ( lower right panel ) . For each subject , the stimulus sequence was structured into approx . 250 trains of a varying number of identical tones , each of which was followed by a train of tones with a different frequency . In other words , the same tone was repeated several times and then changed to a new tone . This lead to two types of events: tone repetition and tone change . The probabilities of trains with zero to ten tone repetitions were 2 . 5% , 2 . 5% , 3 . 75% , 3 . 75% , 12 . 5% , 12 . 5% , 12 . 5% , 12 . 5% , 12 . 5% , 12 . 5% , and 12 . 5% . The tone frequencies were , and they occurred with equal probability in a pseudorandom order . Tones lasted for and were presented at a constant stimulus onset asynchrony of for 15 minutes using headphones . In this study , we quantified the MMN by subtracting the average of waveforms elicited by the sixth presentation of a tone ( the standard ) from the waveform elicited by its first presentation ( the deviant ) . In other words , we compared responses to physically identical stimuli presented in different contexts ( i . e . after different stimulus trains ) . This avoids confounding factors that would have arisen had we used a classical oddball or mismatch negativity paradigm [28] for our single-trial analysis ( e . g . , differences in physical stimulus properties between standards and deviants and differences in the degree to which the standard was expected [27] ) . This section introduces our modelling framework for single-trial responses . In terms of notation , we denote vectors by lower case bold letters , matrices by upper case bold letters , and scalars and functions by lower case italics ( except for variables like the free-energy for which there are notational conventions in the literature ) . Vector and matrix elements can be scalars , vectors , or matrices , and they are referred to via subscripts ( e . g . , denotes the tth element of vector , and denotes the jth element of the kth row of matrix ) . Models of single-trial responses can be cast in a general dynamic state-space framework that models the measurements as manifestations of internal states which cannot be observed directly . The internal states evolve according to an evolution function mapping an internal state and some sensory input to the ensuing state . The internal states generate neurophysiological signals in response to sensory input according to a response function . These are scaled and combined according to a linear observation model with regression coefficients and corrupted by Gaussian measurement noise . Both the evolution function and the response function may depend on parameters and have the following general form: ( 1 ) Together with the prior density , the evolution function and the response function define a generative model of the measurements: ( 2 ) This framework is based on [34] and enables inferences about ( hidden ) computational processes and representations from neurophysiological measurements . It is particularly powerful in conjunction with model comparison methods such as random-effects Bayesian model selection [35] and model space partitioning ( i . e . , inference on model families [36] ) . Given competing models of learning and inference , Bayesian model inversion and comparison can be used to infer the nature of the underlying process and its relationship to the measured responses . The resulting posterior model probabilities assess each model's relative explanatory power in a way that balances fit and complexity such that the comparison between any two models is valid irrespective of their relative complexity . We applied the framework introduced in the previous section to formalize five competing theories of the MMN by formulating thirteen models ( ) of measured trial-wise MMN amplitudes elicited by tone sequences . Each of the five theories summarized in the introduction ( predictive coding , novelty detection , model adjustment , change detection , and adaptation ) explains the MMN as originating from a particular process operating on some neural state or cognitive representation . We modelled these processes and representations as well as the resulting neural responses which we interpret as local field potentials . Since the EEG signal is a linear mixture of local field potentials , we use a general linear model to map predicted neuronal activity to MMN amplitude; this is expressed by Eq . ( 3 ) where are the unknown regression coefficients , and the trial-wise values of define the design matrix: ( 3 ) Note that this is an equation for a single electrode ( we generalize it to multiple electrodes in Eq . ( 13 ) ) . The 13 models are derived in detail below . After formalizing two traditional phenomenological MMN theories ( the change detection hypothesis and the adaptation hypothesis ) , we formalize three current theories of the MMN using Bayesian information processing models based on the free-energy principle . These models assume that the brain represents probabilistic beliefs about its environment whose evolution approximates Bayes optimal learning and perception according to the free-energy principle [37] . The predictive coding , the model adjustment , and the novelty detection theories were formalized by extending this core assumption by response models of different neural sub-processes of the belief updates prescribed by the free-energy principle . Overall , our model space is structured hierarchically , as shown in Figure 2 . First , our 13 models can be grouped into five model families that correspond to the five MMN theories introduced above: change detection ( ) , adaptation ( ) , prediction error ( ) , novelty ( ) , and model adjustment ( ) . The models within each family assume the same internal representation and the same evolution function , but differ in their response functions . Second , these model families can be grouped into two super-families: phenomenological models ( ) and information processing models ( ) . The latter are formulated within a Meta-Bayesian framework [34] and build upon the free-energy principle [37] . Table 1 summarizes all computational models , and the notation used to describe them is summarized in Table 2 . Above , we have derived 13 different models predicting the trial-wise MMN amplitudes during our roving oddball experiment . These models differ in numerous ways , conceptually and mathematically . For example , the evolution function of the change detection models has no free parameters whereas the evolution function of free-energy models has 3 free parameters ( see Table 2 ) . Critically , because model fit increases monotonically with model complexity , the relative plausibility of these models cannot simply be established based on how well they fit the data . Generally , the true desideratum of model comparison , the generalizability of a model , cannot be determined from fit measures alone; instead , model comparison needs to assess the trade-off between model fit and model complexity [65] , [66] . From a Bayesian perspective , this is provided by the ( log ) model evidence ( i . e . , the log probability of the data given a model ) which corresponds to the negative surprise about the data and represents a principled measure of the balance between model fit and model complexity . Here , we used a Bayesian model selection ( BMS ) procedure at the group level that treats models as random effects in the population and can successfully deal with population heterogeneity and outliers [35] . As input , this procedure requires the log-evidence of each model considered , for each subject separately . In the following , we describe how these log-evidences were obtained , detailing the likelihood function and priors that underlie the computation of the log-evidence for individual models and subjects . As EEG signals result from a linear superposition of local electrophysiological responses , one can use a general linear model to map the predictions of local field potentials ( in Table 2 ) to measured trial-wise MMN amplitudes . In each subject and for each model considered , we modelled the data matrix of trial-wise MMN amplitudes across all trials and across all selected electrodes as follows: Let denote the vector of MMN amplitudes at a selected electrode . We regard each as noisy observations of an electrode-specific linear mixture of evoked neuronal responses that reflect the trial-by-trial evolution of internal states . For each response model described above , we therefore apply the following multivariate Bayesian linear regression model with conjugate priors to each subject's data: ( 13 ) Here , denotes the design matrix that was created by replacing the non-constant columns of ( cf . Eq . 3 ) by their z-transforms , are the regression coefficients for the kth electrode , and is the standard deviation of measurement errors at the kth electrode . When inverting this model , we used uninformative Gaussian priors on the regression coefficients and uninformative Gamma priors on the error precisions; for details see Section 5 in Text S1 . Note that we are not interested in the regression coefficients but in each model's log-evidence . Given the likelihood function and priors described above , the log-model evidences were computed by Monte-Carlo integration ( see Section 5 in Text S1 for details ) . Based on the log model-evidences , we estimated the posterior probability of each model by a Bayesian random effects analysis at the group level [35] with a uniform prior on models . For comparing the model families described in Figure 2 Bayesian inference on partitions of model-space [36] was performed to compute the posterior probability of each model family , where denotes the data across all pre-defined electrodes and subjects . This approach can easily deal with families of different size ( i . e . , different numbers of models per family ) . In brief , unbiased family-level inference requires uniform ( flat ) priors over families , and this was achieved by setting each model's “prior count“ ( i . e . the parameters of the Dirichlet prior on model probabilities ) to 1 over the size of the respective model family; see [36] for details . Inference on model families used Gibbs sampling with two million samples per family . Finally , we computed the exceedance probability [35] for each model and model family , i . e . , the probability that this model ( family ) was more likely to have generated the data than any other model ( family ) .
In the Models and Methods section , we derived five classes of models describing how the MMN may reflect the computational processes that govern learning and perception during the roving oddball experiment . Three of the five model classes were derived from the free-energy principle and correspond to formal representations of three contemporary theories of the MMN; i . e . , predictive coding , novelty detection , and model adjustment . These models explain the MMN as arising from prediction error signals , surprise or adjustments to model parameters , respectively . Furthermore , we formalized two traditional theories of the MMN: the change detection and adaptation theory . The resulting model space comprised 13 models in five families ( see Figure 2 ) . In all models , we have connected the ( hidden ) processes of perception and learning to measured EEG responses via different response models and a linear electromagnetic forward model . In this section , we assess the relative plausibility of these models and model families using posterior model probabilities and exceedance probabilities computed by Bayesian model selection ( BMS ) as detailed above . The resulting posterior distributions will be presented as figures , and the main text will report inferences based on those distributions in terms of exceedance probabilities . Figure 4 shows the results of BMS in terms of the posterior probabilities of all models considered . First , note that our “null” model ( M1 , the first change detection model ) , the only model predicting the absence of trial-by-trial changes in MMN amplitudes , is not the best model . Contrary to the predictions of this model , the MMN amplitude appears to vary systematically over deviant trials . This suggests that the MMN is not simply a categorical response to regularity violation but context dependent , as predicted by trial-by-trial statistical learning . Notably , the best five models were all derived within the free-energy framework . Model M6 , which explains trial-wise changes in MMN amplitude as a manifestation of precision weighted prediction errors ( on the hidden tone category ) , was best supported by our data ( exceedance probability ) . It was followed by three “model adjustment” models ( M10 , M11 , M13 ) , each with exceedance probability . These models explain fluctuations in MMN amplitude as arising from a trial-wise adjustment of the parameters encoding posterior beliefs about the expected number of tone repetitions and the conditional transition probabilities . When examining the fit of the best model , we found that it accounted for 2 . 3% of the total variance of single-trial MMN amplitudes ( across all subjects ) . The amount of variance explained was significant in each and every subject ( p<0 . 01 in 6 subjects; p<0 . 02 in two subjects ) . To put this into perspective , this model-based explanation accounted for about 6 . 5 times as much variance as could be explained by a more conventional analysis , i . e . , a linear regression model considering recent stimulus history ( number of standards preceding the deviant ) . While the exceedance probability of the best model M6 was about five times as large as the exceedance probability of our “null” model M1 , this was too small to yield an acceptably low probability of model selection error [67] . As the bar plot shows , the probability mass is concentrated on two model families ( prediction error and model adjustment ) but distributed over several models . Thus , BMS at the level of model families was more appropriate than comparing individual models . From a statistical perspective , this trades a reduced resolution of the hypothesis ( model ) space for increased statistical power . In other words , we move from asking which specific model is best to asking which of the five general MMN theories best explains the data , irrespective of their precise implementations ( cf . Figure 2 ) . This comparison of the five model families is summarized in Figure 5a . The most plausible MMN theory was the model adjustment theory ( ) , followed by the prediction error theory ( ) . Finally , we used BMS to examine whether the free-energy principle based models provide , in general , better explanations of the variability of single-trial MMN amplitudes than phenomenological models . This means we are now comparing only two families ( Figure 2 ) : the family of free-energy based models ( predictive coding , novelty detection and model adjustment; ) and the family of more traditional phenomenological models ( change detection and adaptation , ) . Family-level BMS indicated that models based on the free-energy principle were considerably more convincing than phenomenological models; ( see Figure 5b ) . Finally , we asked which level of the processing hierarchy contributes most to the fluctuations in trial-wise MMN amplitudes . In other words , we examined whether response variations arise from lower auditory areas representing physical sound properties like frequency , or from higher areas that represent abstract temporal structure . For this purpose we re-partitioned the 13 models into two families according to whether they explain MMN generation in relation to a low-level auditory feature ( sound frequency ) or a high-level auditory feature ( temporal structure ) . For the models based on the free-energy principle models the two levels of representation map onto the two levels of the mental model: sensory inputs and hidden sequence of tone categories ( Figure 3 ) . We assigned the free-energy based models that relate the MMN elicited by changes in sound frequency to the representation of sound frequencies to the first model family and those that relate it to the represented sequence of tone categories to the second . Furthermore , both the adaptation model and the change detection theory are formulated explicitly with regard to stimulus frequencies and are therefore assigned to the first model family . Overall , this resulted in the following two model families: and . Comparing these two model families yielded an exceedance probability of for , suggesting that the auditory MMN is more closely related to a representation of high-level auditory features , such as temporal structure , than to a representation of low-level features , such as sound frequency . The models reported above were designed to predict the evolution of single-trial MMN amplitudes throughout the experiment . This was done to capture putative history-dependent effects . The models which did take into account such effects ( i . e . , free energy based models ) were found to have higher evidence than models which did not ( e . g . , the various change detection models ) . One may ask , however , as did one of our reviewers , whether our single-trial approach was really necessary or whether it would have been sufficient to analyse the average MMN amplitude as a function of the number of preceding standards and the change in frequency . Here we provide a conventional analysis of variance to demonstrate that our data did contain history-dependent effects that would have been removed by conventional averaging . By history-dependent effects we mean that the MMN amplitude evoked by a deviant following a given number of standards and a given frequency change will differ depending on the tones that preceded the current sequence of standards . The mere number of such tones is a minimal definition that ignores the effects of their statistical structure , some of which are captured by our models . However , it allows for a conservative test of history-dependence , i . e . , whether a 3-way analysis of variance ( ANOVA ) of trial-wise MMN amplitudes reveals interactions among three factors: ( i ) number of preceding standards , ( ii ) frequency difference , and ( ii ) time , i . e . , the number of preceding trial sequences . We found significant main effects for the number of preceding standards and for frequency difference ( Figure 6 ) . More importantly , however , we found highly significant interaction effects , indicating that the effect of the number of preceding standards on MMN amplitude did not only depend on the frequency difference between standard and deviant ( ) but also on the number of previous tone sequences ( ) . This demonstrates that the trial-wise MMN amplitudes we recorded do indeed show history-dependent effects that would be removed by conventional averaging procedures .
Our analyses suggested that stimulus history ( i . e . , previous tone sequences ) affects the MMN in intricate ways . This was not only demonstrated by a simple ANOVA of single-trial MMN amplitudes , but , more importantly , by our systematic model comparisons which favoured free-energy based Bayesian information processing models that capture history-dependent effects . In particular , these models explain the dependence of the MMN on interactions between previous tone sequences and the current tone sequence in terms of trial-by-trial learning of statistical structure . Trial-by-trial statistical learning implies that the probabilistic expectation evoked by a given tone sequence is different for every presentation , and that each difference reflects what has been learned since the previous presentation . While traditional MMN studies have ignored trial-specific effects by averaging responses across deviant events , several studies have addressed sequential changes in the MMN across trials [19] , [23] , [25]–[27] , [51]–[54] , [68]–[70] . However , only [23] and [68] have completely avoided averaging procedures altogether . The results of this study and [23] question the frequent assumption that the MMN amplitude is constant throughout an experimental condition ( i . e . , for given tones and following a given number of standards ) . Instead , our results suggest that trial-by-trial changes in MMN amplitude are highly history-dependent and represent an informative index of statistical learning as the recording session proceeds . It is pleasing that [23] reached a similar conclusion , even though they studied mismatch potentials in a different modality ( i . e . , somatosensory ) and with simpler models , but with source-reconstruction and a high temporal resolution . Thus , while averaging is a useful tool to increase the signal-to-noise ratio , single-trial data carry unique information about the processes of learning and perception that underlie the MMN . A number of previous studies reported that the MMN amplitude elicited by a change in sound frequency increases monotonically with the number of preceding standards [27] , [51]–[54] , [69] , [70] . By contrast , we found a non-monotonic effect of the number of preceding standards on deviant response amplitude ( see Figure 6a ) . The reason for this discrepancy may be that previous studies did not disentangle the contributions of the standard ERP and the deviant ERP ( cf . [71] ) . In contrast , in this study , we operationalized the MMN with respect to a fixed standard ERP ( see Models and Methods ) , so that changes in MMN amplitude reflected changes in the neural response to the deviant only . In summary , our results do not contradict previous findings on the relationship between the number of preceding standards and the MMN amplitude [27] , [51]–[54] , [69] , [70] but complement them . Furthermore , our models based on the free-energy principle can explain why Haenschel et al . [27] observed a monotonic decay of the standard response with the number of standard repetitions , and they predict how stimulus history determines the effect of preceding standards on deviant response amplitude . Our modelling results do not lend support to the adaptation hypothesis of the MMN [18] or the change detection interpretation of the memory trace hypothesis [72] . Instead , our results support explanations postulating that the brain maintains and constantly updates an internal model of its environment . For example , the model adjustment hypothesis [19] posits that auditory cortex maintains a model of the acoustic environment , and that stimulus-induced updates of this model are indexed by the MMN [20] . While the original proposal was of a conceptual nature , our present work formalizes this hypothesis by specifying how trial-wise changes in MMN reflect an approximation to Bayesian updating of a probabilistic mental model . The resulting models are consistent with the conclusion drawn by [23] that ( somatosensory ) mismatch potentials reflect perceptual learning . However , our analysis was more fine-grained in that it distinguished between three computational mechanisms that might underlie the perceptual learning that [23] indexed in terms of Bayesian surprise . Concretely , we distinguished between prediction error signalling , novelty detection , and model adjustment . Our results supported model adjustment and , to a lesser extent , prediction error signalling , but not novelty detection , even though it computes an approximation to ( Shannon ) surprise . We also distinguished between perceptual learning at the level of physical stimulus properties ( sound frequency ) and learning of abstract temporal structure and found strong evidence for the latter . In neurobiological terms , model adjustment might correspond to synaptic plasticity at top-down projections targeting pyramidal neurons in layers 2 and 3 ( “prediction error units” ) via NMDA receptors [3] ( see Section 6 in Text S1 ) . This would be consistent with the observation that pharmacological blockage of NMDA receptors diminishes the MMN [73]–[75] . Predictive coding formulations of free-energy minimization assign prediction errors a critical role in the update of posterior beliefs . When comparing all models individually , the best model was indeed one that explained trial-wise fluctuations in MMN amplitude as a function of precision weighted prediction errors ( model M6; Figure 4 ) . However , its superiority over other models was marginal , and model comparison at the family-level ( Figure 5a ) did not support the hypothesis ( proposed in [3] ) that the MMN solely reflects precision weighted prediction errors . This suggests that while prediction error signalling may be essential for the free-energy minimization process underlying the MMN , it is probably not the sole determinant of trial-wise MMN amplitudes . Alternatively , our failure to find stronger evidence for the hypothesis that ( precision weighted ) prediction errors alone determine trial-wise MMN amplitudes may be due to some of our simplifying assumptions , as discussed in the next section . Overall , one should bear in mind that our inferences are primarily about rather abstract models or classes of models . Our free-energy based models , in particular , consider the outcomes of neuronal computations rather than their process . This is a necessary constraint on models of discrete trial-by-trial variations in responses; as opposed to continuous time models that would consider the precise time-course of neural responses over peristimulus time . This means that we have to assume that there is some aspect of neuronal activity or excitability that encodes the posterior beliefs associated with each oddball trial . However , the relationship between biophysical quantities like synaptic activity or gain , on the one hand , and posterior beliefs , predictions , and surprisal , on the other hand , are not specified explicitly in this sort of model . This means that it is difficult to make any strong statements about the neurobiology that implements any Bayesian inference . Furthermore , our models make several simplifying assumptions that may turn out to be false . First , there is still no conclusive evidence about how prediction errors are represented at the level of single neurons . Second , the assumption of a linear relationship between the encoded quantity and the MMN amplitude is simplistic and ignores potential nonlinearities . Third , all of our models represent the MMN by a single number ( i . e . , its peak amplitude ) , rather than by its waveform , thereby ignoring its temporal dynamics and spatial topography . Fourth , each of our models relates trial-wise MMN amplitudes to a single computational variable , whereas it is known that the MMN scalp potential is a mixture of signals from several brain areas with ( presumably ) different functional characteristics [29]–[31] , [76] . Finally , while our results indicated that our neuronal adaptation model M4 is insufficient to explain single-trial variations in MMN , we have not tested the fresh-afferent theory [13] that is based on stimulus specific adaptation . In future work , it would be useful to formulate this theory as models of stimulus specific adaptation [12] , [13] , [42] under the present framework and compare it to the computational models presented in this paper . Our models based on the free-energy principle link the MMN to the neuronal encoding of posterior beliefs that is postulated by the Bayesian brain hypothesis . According to this hypothesis , the brain represents probabilistic beliefs , and updates them in an ( approximately ) Bayesian fashion . Previous work along these lines has assumed that the support of probability distributions is partitioned into small bins and that each bin's probability mass is represented by the firing rate of dedicated neurons [77] , [78] , or that probability densities are approximated by a linear combination of basis functions [79] . In contrast to these high-dimensional representations , we have implicitly assumed a much simpler , low dimensional fixed-form approximation to the posterior density . Our predictors of electrophysiological responses are simple functions of posterior expectations on log-frequency , tone category and transition probabilities . These posterior expectations might be encoded by the average activities of neuronal populations , and the precision parameters that determine the relative weight assigned to prior beliefs and sensory evidence could be encoded by the strength of the recurrent connections of prediction error units [80] ( see also Section 6 in Text S1 ) . This representation is not motivated by sparseness , but by computational efficiency: It replaces the problem of computing the ( potentially very high-dimensional ) posterior probability density by optimizing the free-energy with respect to a small set of sufficient statistics . This variational Bayesian optimization rests on free-energy minimization [37] and proposes the minimization of prediction error as an explanation for stimulus-evoked transient neuronal responses such as the MMN [3] , [63] , [81] . The work presented in this paper is a step towards linking models of probabilistic neural coding and inference to neuronal signals that can be measured non-invasively in humans . Our present results were based on a single “roving oddball” EEG experiment that was originally designed for comparing dynamic causal models of interactions among cortical areas during the MMN [25] . In the future , it would be interesting to apply the approach presented here to other types of MMN paradigms . Additionally , one could use our models in conjunction with recent advances in design optimization that maximize the sensitivity of Bayesian model selection [67] to create an experiment that is optimal for discerning between the models selected by our analysis . In addition , our modelling and model comparison framework could be applied to source-reconstructed mismatch potentials to characterize functional differences between the brain areas jointly generating MMN scalp potentials . Furthermore , the link between single-trial mismatch potentials , on the one hand , and statistical learning and perceptual inference , on the other hand , could be exploited to measure the temporal dynamics of how the brain learns the probabilistic structure of complex environments . This is an attractive prospect , given that the MMN is elicited not only in simple oddball experiments , but also in more complex experiments involving speech , language , music , and abstract features , as well as various other sensory modalities [14] , [71] , [82] , [83] . Our modelling framework could also be used to probe disturbances of perceptual inference and learning in psychiatric conditions , such as schizophrenia [84]–[86] . In addition , future studies might use the meta-Bayesian approach [34] for inferring , from single-trial MMN amplitudes , subjects' prior beliefs about hidden temporal structure , which constitute the inductive biases [87] that endow the brain with its remarkable ability to discover complex sequential regularities . | The ability to predict one's environment is crucial for adaptive and proactive behaviour . It requires learning a mental model that captures the environment's statistical regularities . A process of this sort is thought to be reflected by the mismatch negativity ( MMN ) potential , a non-invasive electrophysiological measure of the neural response to regularity violation by sensory stimuli . However , the exact computational processes reflected by the MMN remain a matter of debate . We developed a modelling framework in which competing hypotheses about these processes can be objectively compared by their ability to predict single-trial MMN amplitudes . We applied this framework to formalize five major MMN theories and propose a unifying view on three distinct theories which explain the MMN as a reflection of prediction errors , model adjustment , and novelty detection , respectively . We assessed our models of the five theories with EEG data from eight healthy volunteers . Our results are consistent with the idea that the MMN arises from prediction error driven adjustments of a probabilistic mental model of the environment . | [
"Abstract",
"Introduction",
"Models",
"and",
"Methods",
"Results",
"Discussion"
] | [
"computational",
"neuroscience",
"biology",
"neuroscience"
] | 2013 | Modelling Trial-by-Trial Changes in the Mismatch Negativity |
Strongyloides stercoralis is a soil-transmitted nematode that can replicate within its host , leading to long-lasting and potentially fatal infections . It is ubiquitous and highly prevalent in Cambodia . The extent of morbidity associated with S . stercoralis infection is difficult to assess due to the broad spectrum of symptoms and , thus , remains uncertain . Clinical signs were compared among S . stercoralis infected vs . non-infected participants in a cross-sectional survey conducted in 2012 in eight villages of Northern Cambodia , and before and after treatment with a single oral dose of ivermectin ( 200μg/kg BW ) among participants harboring S . stercoralis . Growth retardation among schoolchildren and adolescents was assessed using height-for-age and thinness using body mass index-for-age . S . stercoralis prevalence was 31 . 1% among 2 , 744 participants . Urticaria ( 55% vs . 47% , OR: 1 . 4 , 95% CI: 1 . 1–1 . 6 ) and itching ( 52% vs . 48% , OR: 1 . 2 , 95% CI: 1 . 0–1 . 4 ) were more frequently reported by infected participants . Gastrointestinal , dermatological , and respiratory symptoms were less prevalent in 103 mono-infected participants after treatment . Urticaria ( 66% vs . 11% , OR: 0 . 03 , 95% CI: 0 . 01–0 . 1 ) and abdominal pain ( 81 vs . 27% , OR: 0 . 07 , 95% CI: 0 . 02–0 . 2 ) mostly resolved by treatment . S . stercoralis infection was associated with stunting , with 2 . 5-fold higher odds in case of heavy infection . The morbidity associated with S . stercoralis confirmed the importance of gastrointestinal and dermatological symptoms unrelated to parasite load , and long-term chronic effects when associated with malnutrition . The combination of high prevalence and morbidity calls for the integration of S . stercoralis into ongoing STH control measures in Cambodia .
Strongyloides stercoralis , one of the most difficult to diagnose neglected tropical diseases , is an intestinal soil-transmitted parasitic nematode that occurs worldwide and is highly prevalent in warm regions with poor sanitation [1 , 2] . Its prevalence is largely underestimated due to the inability of simple coprological diagnostic techniques to detect S . stercoralis larvae [3 , 4] . Although numerous aspects of the epidemiology of S . stercoralis remain poorly documented , the parasite is very common , with prevalence rates in the tropics and subtropics exceeding 40% [1] . A rough estimate of 200–370 million cases worldwide has recently been put forward [3] . Because it can be life-threatening in immunocompromised patients , S . stercoralis is also known throughout the developed countries , with most literature originating from hospital-based case reports of severe strongyloidiasis among transplant recipients , travelers , and migrants . Commonly reported symptoms of uncomplicated strongyloidiasis include diarrhea , vomiting , abdominal pain , urticaria , and “larva currens” , while half of the cases are asymptomatic [5–7] . Larva currens is an intermittent urticarial linear , serpiginous eruption due to the migration of larvae under the skin . The high speed at which larvae travel ( 5 to 10 centimetres per hour ) and the location of lesions ( lower trunk , bottom and thighs ) make larva currens a highly specific symptom of S . stercoralis infection [5 , 8 , 9] . S . stercoralis can replicate within its host , permitting ongoing “autoinfection” . This leads to long-lasting infections and potential fatalities among immunosuppressed patients , such as those undergoing corticosteroid therapy or suffering certain concomitant diseases or malnutrition [2 , 5 , 10] . There is a paucity of studies investigating the health impacts of S . stercoralis in low-income tropical and sub-tropical countries , including its associations with malnutrition and growth retardation [2 , 4] . It is challenging to document the clinical signs and symptoms associated with specific soil-transmitted helminth ( STH ) infections in endemic poly-parasitic settings because of their non-specific clinical presentations and because of the difficulty accounting for co-infections ( due to the sub-optimal sensitivity of most direct diagnostic approaches ) [4] . Preventive chemotherapy , the strategy for controlling STH , would be feasible for S . stercoralis using a single oral dose of ivermectin ( 200 μg / kg body weight ( BW ) ) , which achieves high cure rates , comparable to that of a two dose regimen [11–16] . The evidence base for morbidity in endemic areas is small and must be improved in order to have S . stercoralis integrated into the WHO control strategy for reducing the global burden of soil-transmitted helminths [2–4] . The objective of this work was to quantify the morbidity associated with chronic S . stercoralis infection in a highly endemic setting in Cambodia . We compared infected vs . non-infected participants and assessed the degree of symptom resolution achieved with a single oral dose of ivermectin ( 200 μg / kg BW ) , while excluding or adjusting for co-infection with other helminth species or pathological protozoan parasites . Additionally , we investigated the association between growth retardation and S . stercoralis infection risk and intensity among schoolchildren and adolescents .
The study protocol was approved by the National Ethics Committee for Health Research , Ministry of Health , Cambodia; and by the Ethics Committee of Northeast and Central Switzerland . All participants were informed of the study purpose and procedures and written informed consent was obtained before enrollment . Morbidity associated with S . stercoralis infection was assessed in three sub-studies . First , symptoms associated with S . stercoralis infection were identified in a community-based cross-sectional study . Second , symptom resolution by standard treatment was quantified in a before-and-after treatment approach . Third , the association between malnutrition and S . stercoralis infection was assessed for children participating in the cross-sectional study . A community-based , cross-sectional survey was conducted between February and June 2012 in Rovieng district , Preah Vihear Province , Northern Cambodia , where S . stercoralis is highly endemic [17] . The survey was part of a larger two-year intervention study , described in detail elsewhere [13] . In brief , we included all households in the eight villages that had never received ivermectin treatment; all household members over the age of two were eligible . The resulting sample was used to assess the symptoms associated with S . stercoralis infection while adjusting for co-infection with any other diagnosed helminths or pathological protozoan parasites . All S . stercoralis cases were treated with a single oral dose of ivermectin ( 200μg/kg BW ) and all other diagnosed parasitic infections were treated in accordance with the national guidelines [18] . Symptom resolution after ivermectin treatment was investigated among S . stercoralis infected patients from two villages who were followed-up 21 days after treatment with a single oral dose of ivermectin ( 200μg/kg BW ) . Assessment was conducted both among patients with S . stercoralis mono-infection ( i . e . excluding co-infection with any other diagnosed helminths or pathological protozoan parasites ) and among all S . stercoralis infected patients ( i . e . all S . stercoralis cases whether they were co-infected with other parasites or not ) . Adjustments were made for those with co-infections . A time span of 21 days is long enough to maximize cure time and short enough to avoid reinfection . The single oral dose was chosen because it achieves a high cure rate and is appropriate in the framework of control efforts [13] . Post-treatment data were collected through parasitological assessment and medical interviews about symptoms experienced in the three days preceding follow-up . The growth retardation assessment was performed among children 5–19 years of age , residing in the eight villages . Demographic data ( sex , age , occupation and education level ) and history of anthelmintic treatment were collected from all participants using a pre-tested questionnaire . A clinical assessment of all participants was conducted by a medical doctor and included anthropometric measures , physical examination , and an assessment of signs and symptoms experienced in the two-weeks prior . Information on ownership of household-assets was obtained from heads of households . S . stercoralis was diagnosed using Koga Agar plate ( KAP ) culture and the Baermann technique on two samples collected on consecutive days [17 , 19 , 20] . KAP and Baermann centrifuged eluents were checked for species to prevent the misclassification of hookworm vs . S . stercoralis larvae . Other STH were diagnosed using a Kato-Katz thick smear on each of the two fecal samples and formalin-ether concentration technique ( FECT ) on one sample [21 , 22] . Protozoa were diagnosed with FECT performed on one stool sample [22] . S . stercoralis parasite load was estimated based on larvae counts from the Baermann test , using the following thresholds: low parasite load—a positive count , up to one larva per gram of stool ( LPG ) ; moderate parasite load—two to nine LPG; and high parasite load—more than 10 LPG [23 , 24] . Participants with positive KAP and negative Baermann tests were assumed to have low parasite loads . To keep all the positives in the sample and to maximize the specificity , the infection status of participants in any of the three assessments was determined using the following case definition: a S . stercoralis patient had at least one S . stercoralis positive fecal sample . A S . stercoralis-free participant had negative results in all four available diagnostic results ( one Baermann and one KAP on each of two samples ) . The same approach was used for the other helminths , i . e . all positives ( determined by at least one positive diagnosis result by any method ) and only negatives with two Kato-Katz results were included . Since there was only one FECT result per participant , the case definition does not apply to protozoa . Clinical and laboratory data were managed and analyzed in STATA version 13 . 0 ( StataCorp LP; College Station , United States of America ) . Anthropometric measures and summaries were calculated using the WHO code ( http://www . who . int/growthref/tools/en/ ) in R [25] . In the cross-sectional study , age of participants was categorized into four groups , as follows: ( i ) < 6 years , ( ii ) 6–18 years , ( iii ) 19–59 years , and ( iv ) ≥ 60 years . The age of children included in the growth retardation analysis was categorized into three groups , corresponding to school level ( i . e . primary , secondary , high school ) , as follows: ( i ) 5–9 years , ( ii ) 10–13 years and ( iii ) 14–19 years . First , the association between each symptom , S . stercoralis infection status , and parasite load was investigated using multivariate logistic regression models ( with the symptom as outcome ) , adjusting for any other helminthic or protozoan infection , sex , age , and treatment uptake in the past year . Second , the symptom resolution achieved by ivermectin treatment was assessed with McNemar’s exact test among patients harboring S . stercoralis mono-infections , by comparing the proportions of participants with a particular symptom before and after treatment . An additional analysis including co-infections with other parasites used conditional logistic regression models with each symptom as outcome , and survey ( before vs . after treatment ) and presence of any other infection as explanatory variables . Third , for assessing growth retardation and thinness in children , the z-scores for height-for-age ( HAZ ) and body mass index ( BMI ) -for-age ( BAZ ) were calculated for children between 5 and 19 years , participating in the cross-sectional survey . The WHO Growth Reference Standard was used to calculate anthropometric indicators for school-aged children and adolescents [26 , 27] . Children with HAZ and BAZ values lower than -2 were classified as “stunted” or “thin” , respectively , as opposed to “normal” for values larger than -2 . The association between stunting , thinness and S . stercoralis infection status and parasite load was assessed using logistic regression with the nutritional variable as outcome . The final multivariate model was built based on Akaike Information Criterion ( AIC ) . Interactions between sex and S . stercoralis parasite load , age , and socioeconomic status , as well as between age and S . stercoralis parasite load and socioeconomic status were checked using the Likelihood Ratio Test ( LRT ) .
Among the 3 , 837 participants enrolled in the study , diagnostics could not be performed for S . stercoralis or for protozoan infections for 134 and 320 individuals , respectively , due to absent samples ( participants did not return any sample ) or an insufficient amount of stool to perform the FECT . The remaining 3 , 377 participants all completed the interview and the clinical assessment and were included in the study . Females and children under six years old were significantly more prone to missing diagnostics and exclusion from the sample . All regression models were adjusted for sex and age , which controlled for this potential bias . The association between S . stercoralis infection and symptoms was assessed for 2 , 744 participants with confirmed infection status for all investigated parasites ( Fig 1 ) . This final sample size results from excluding all participants with uncertain negative status for helminth infection , i . e . those with fewer than four negative results ( two methods applied to two samples ) for S . stercoralis and fewer than two negative results ( Kato Katz on two samples ) for other helminths , to maximize diagnosis specificity . Basic characteristics of this sample are presented in Table 1 . The prevalence of S . stercoralis infection was 31 . 1% ( 95%CI: 29 . 4–32 . 9 ) . A quarter ( 699/2 , 744; 25 . 3% ) of participants was infected with hookworm , which was the only other common parasite . Among S . stercoralis infected patients , 325/853 ( 38 . 1% ) were co-infected with hookworm . Other helminths species were rare , with prevalence rates below 3% , while 8% of participants were infected with pathogenic protozoa , namely Giardia lamblia or Entamoeba histolytica/dispar . Species specific prevalences are available in S1 Table . For the before-after treatment assessment , 208 of the 316 S . stercoralis positive participants had confirmed infection status for all parasites at both assessments . About half ( 103/208 ) of the participants in that sub-sample were not infected with any other helminth or pathogenic protozoa either before or after treatment , while the other 105 harbored other intestinal parasites before and/or after treatment . Basic characteristics of the sample analyzed for the before-after assessment are presented in Table 1 . The growth retardation assessment was conducted among 1 , 057 children aged 5–19 years , with confirmed infection status for S . stercoralis and hookworm , and who had S . stercoralis parasite load assessed . Of the 1 , 338 children aged 6–19 years participating in the study , 229 were excluded from the final analysis due to incomplete diagnostics for S . stercoralis , seven were excluded because of missing diagnosis results for hookworm , and 47 were excluded due to the absence of parasite load information . Basic characteristics of the sample are presented in Table 1 . Among this group , the prevalence of S . stercoralis was 24 . 0% ( 95%CI: 21 . 5–26 . 7 ) . Most cases ( 160/254; 63 . 0% ) had low parasite loads , whereas 20 . 1% and 16 . 9% had moderate and heavy parasite loads , respectively . Overall , only five ( 1 . 1% ) participants with S . stercoralis mono-infection and 18 ( 1 . 4% ) infection-free participants were reportedly without symptoms . The most frequently reported symptoms by S . stercoralis-infected patients were abdominal pain , cough , epigastric pain , diarrhea , and urticaria , ranging from 83 . 1% to 55 . 3% . However , in most cases they were not significantly associated with S . stercoralis infection . The symptoms with the strongest association were urticaria ( 55 . 3% vs . 46 . 5% , OR: 1 . 35 , 95% CI: 1 . 13–1 . 60 ) and itching ( 52 . 4% vs . 47 . 7% , OR: 1 . 19 , 95% CI: 1 . 00–1 . 41 ) . The frequency of symptoms and the strength of their association with S . stercoralis infection status are presented in Table 2 . When accounting for S . stercoralis parasite load instead of infection status , urticaria was also found to be significantly associated , but not itching ( S2 Table ) . Of note , age was associated with most of the reported symptoms ( S3 Table ) . Table 3 presents the extent of symptom resolution following single-dose ivermectin treatment among patients with S . stercoralis mono-infection ( i . e . excluding co-infection with any other helminth or pathogenic protozoan parasite ) . All participants reported at least one symptom before treatment . The symptoms that declined most after treatment were urticaria ( 66 . 0% vs . 10 . 7% , OR: 0 . 03 , 95% CI: 0 . 00–0 . 13 ) , abdominal pain ( 80 . 6% vs . 27 . 2% , OR: 0 . 07 , 95% CI: 0 . 02–0 . 18 ) , and vomiting ( 23 . 3% vs . 1 . 0% , OR: <0 . 1 , 95% CI: <0 . 01–0 . 17 ) . Other symptoms that declined significantly after treatment were nausea , diarrhea , tiredness , and cough . With the addition of 105 patients co-infected with other helminth or pathogenic protozoan parasites , the same symptoms were found to significantly recede post-treatment . Loss of appetite was also less frequently reported in this group ( S4 Table ) . Fig 2 displays the proportions , stratified by parasite load , of the six symptoms that were significantly less frequently reported by the S . stercoralis patients who were free of any other diagnosed parasite , following treatment . The decreases appear to be of a similar magnitude among participants with light vs . moderate or heavy infections . More than half of the children were either moderately ( 40 . 4% ) or severely ( 11 . 3% ) stunted . The proportion of stunting was 46 . 6% in S . stercoralis-infected children and 37 . 8% in non-infected children . Stunting was associated with S . stercoralis infection status , with a higher risk of being stunted when infected with S . stercoralis ( OR: 1 . 35 , 95%CI: 1 . 00–1 . 81 ) or when heavily infected compared to uninfected ( OR: 2 . 49 , 96%CI: 1 . 31–4 . 71 ) . No interactions were found . Results of the association between stunting and S . stercoralis parasite load are presented in Table 4 . Thinness was also common , with one in five children ( 19 . 9% ) being underweight , but thinness was not associated with either S . stercoralis infection status ( χ2 = 0 . 47 , LRT p-value = 0 . 5 ) or parasite load ( χ2 = 1 . 21 , LRT p-value = 0 . 75 ) .
This is the first report , to our knowledge , showing symptoms specifically associated with S . stercoralis , i . e . excluding other helminth infections and pathogenic protozoa , in a multi-parasitic setting , where it is particularly difficult to assess the morbidity associated with specific STH infections . Both the sensitivity and specificity of helminth diagnosis were maximized by combining all available methods ( Baermann , KAP , and Kato-Katz on two stool samples , and FECT on one stool sample ) , and by including only negatives with the highest “certainty” of negative status , i . e . participants with all four ( Baermann and KAP each on two samples ) available results for S . stercoralis and with two ( Kato Katz on two samples ) results for other helminths . Women and children under six years were more likely to have incomplete diagnostic results for helminths , but this aspect did not bias the analysis as all models were adjusted to account for those factors . Among the 103 participants infected with S . stercoralis only , dermatological , gastrointestinal , and respiratory symptoms , such as urticaria , abdominal pain , nausea , vomiting , diarrhea , and cough , were found to be significantly less frequent after treatment . A previous study conducted in the same province also found that abdominal pain , diarrhea , and urticaria were resolved by ivermectin treatment among 21 heavily infected patients [7] . All of those symptoms are consistent with various phases of the infection [5 , 6] . The most prominent symptom of S . stercoralis infection was urticaria , also called “hives” , which was both mostly resolved by ivermectin treatment and associated with infection status . However , the association was weak given the high proportion of S . stercoralis negative patients also reporting those symptoms and results should be interpreted with caution . The lack of association between reported symptoms and infection status could be due to misclassification of S . stercoralis cases , however we used a diagnosis approach with sensitivity exceeding 92% , even for light infections , so the number of false negative S . stercoralis cases would be low [24] . The weak or absent associations between symptoms and S . stercoralis infection status mostly reflect the difficulty of assessing the relationship between nonspecific clinical signs and STH or protozoan infections in poly-parasite endemic settings [6] . The challenges faced in such assessments tend to include diagnostic approaches with imperfect sensitivity , reported symptoms being subject to recall and reporting biases , and the non-specificity of symptoms that could be due to other pathologies including viral infections . Treatment had no effect on itching , indicating that itching could be the result of numerous other conditions and reasons , particularly in tropical settings , including allergies and insect bites , and not necessarily be disease-specific [28] . Urticaria is a well-known symptom of chronic strongyloidiasis and has already been identified as such in Cambodia through studies that did not , however , exclude cases of co-infection with other helminth species or pathogenic protozoa [4 , 7 , 9 , 29 , 30] . Acute urticaria may occur at the penetration site for hookworms and S . stercoralis and is mostly located on the feet [8 , 9] . Urticaria occurring during the chronic phase of infection is accepted as a systemic reaction due to the parasite-induced immunologic inflammatory response , which results in increased eosinophil and IgE levels , similar to an allergic response [31–33] . However , the actual mechanisms explaining the relationship between skin reactions and helminths remain unclear [33 , 34] . Some authors suggest that urticaria is induced by parasites to ease migration under the skin , in the lymphatic ways and in some parenchymatous organs [33] . Therefore , urticaria would relate to the larval stage of infection or to the parasite migration phase , rather than to the mere presence of parasites in the body [33] . This statement is of particular interest to the etiology of urticaria in strongyloidiasis and would be in line with S . stercoralis’ autoinfection ability , whereby the parasite continuously replicates and produces larvae that re-infect the host . Another striking effect of ivermectin treatment was the resolution of abdominal pain among most of the patients . The impact on other clinical signs—with the exception of vomiting , which was reported by only one patient after treatment—was more modest , as indicated by the substantial proportion of participants still reporting symptoms three weeks after treatment . These symptoms also declined significantly among participants co-infected with S . stercoralis and other parasites ( 75% of which were hookworm ) . Morbidity was not associated with hookworm infection in this study , probably because almost all cases ( 97 . 7% ) were light intensity infections , as defined by the WHO thresholds , or because of a high prevalence of Ancylostoma ceylanicum , a hookworm common to dogs and cats that often infects humans in the region [35–37] . Interestingly , self-reported morbidity was higher among individuals infected with S . stercoralis than among those with hookworm in a setting endemic for both parasites [6] . Surprisingly , while STH morbidity is known to increase with worm load , the clinical manifestations associated with S . stercoralis were not associated with parasite loads in this setting . This result could be due to the irregular larval output of S . stercoralis that might have affected the estimated parasite load . It is also possible that the thresholds used inadequately reflected the impact of parasite load on health or that other undiagnosed pathogenic parasites were effectively treated by ivermectin , thereby having a confounding effect [24 , 38–40] . However , confusion between larval output thresholds used for low ( ≤1 LPG ) vs . high ( ≥ 10 LPG ) parasite load appears unlikely [24 , 38 , 39] . Yet , it cannot be excluded that S . stercoralis is pathogenic even in cases of low parasite load , which would largely affect any indicator of parasite burden or treatment effect , including cost-effectiveness assessments . An important finding was the association between growth retardation and both S . stercoralis infection risk and parasite load . The risk of being stunted increased with age , indicating the accumulation of growth retardation through time . This association may also reflect greater exposure to malnutrition in the past , among older children . Stunting may be due to a number of causes , including suffering from heavy STH infections before STH control was implemented . However , in this setting , where about half of the children suffered from growth retardation , the cross-sectional design of this study could not address causality . Further studies accounting for potential confounders of the relationship between malnutrition and S . stercoralis infection , such as medical history , quantitative and qualitative food intake , and social aspects , are needed to determine specific factors as well as the strength and direction of the association . Moderate to heavy infections with any of the three other STH , A . lumbricoides , T . trichiura , and hookworm , are widely recognized as causes of stunting , which make STH one of the most important causes of physical and intellectual growth impairment [41–44] . Yet , current evidence supporting the association between STH and childhood growth is currently of low quality and warrants further research , which should also include S . stercoralis . Confirming that S . stercoralis infection plays a role in growth retardation due to its contribution to chronic malnutrition in childhood would have an important impact on estimating the disease burden . Protein-calorie malnutrition is a known cause of immunodeficiency in resource-poor countries and may be a pivotal aspect of S . stercoralis morbidity [2] . First , chronic strongyloidiasis causes gastro-intestinal symptoms that potentially lead to malnutrition through lower food intake and nutrient loss [44] . Second , there is evidence from animal studies that nematode infections in malnourished hosts induce a decrease in the T-Helper Type 2 ( Th2 ) mediated immune response , including eosinophil counts , which are known to be an important part of the immune response against S . stercoralis [45 , 46] . Finally , immunodeficiency may increase the risk of complicated strongyloidiasis in malnourished populations , and malnutrition unrelated to known causes of immunosuppression might be responsible for severe strongyloidiasis cases in developing countries [2 , 46 , 47] . The risk of developing severe strongyloidiasis could be high in settings with widespread malnutrition such as Cambodia , while an additional issue is the increased availability of over-the-counter drugs containing corticosteroids [2 , 48] . Our study has several limitations . The difference in durations considered for symptom reporting before and after treatment may have overestimated pre-treatment symptom reporting frequencies as well as treatment effects , particularly in the case of vomiting . Additionally , in the absence of a control group , our study did not account for placebo effects , which could have influenced symptom reporting . Finally , some co-infections with pathogenic protozoa may have been missed due to the limited sensitivity of FECT performed on one stool sample , which might explain the moderate impact of ivermectin treatment on diarrhea and nausea . However , this limitation would not apply to helminths , for which the diagnostic approach used in the present study has been assessed several times and has shown high sensitivity and specificity [17 , 23 , 24 , 29] . Nor would it apply to urticaria , which is commonly associated with other helminths including Ascaris lumbricoides , Hymenolepis nana , and Fasciola hepatica , as well as with protozoans including Giardia lamblia and Blastocystis hominis [28 , 30 , 33 , 49 , 50] . While we cannot exclude that some protozoan infections were missed , ivermectin is not effective against G . lamblia or B . hominis , and so the resolution of urticaria , a widely recognized symptom of chronic strongyloidiasis , in 84% of participants in the before-after study would appear to arise from clearance of S . stercoralis . Scabies is another important cause of itching/urticaria in developing countries that would be resolved by ivermectin treatment [51] . However , the basic medical examination that was conducted during data collection included a skin check , which would have led to scabies diagnosis if present . Our combined results demonstrate that the burden of strongyloidiasis , which encompasses all health states from mild symptoms to severe , life-threatening infection , might be much higher in endemic settings than previously thought . Chronic strongyloidiasis appears to cause both acute gastrointestinal symptoms and urticaria , all bothersome symptoms that are experienced whatever the age of the individual and intensity of infection . It also causes subtle long-term health effects through its association with malnutrition . Next steps towards estimating the S . stercoralis burden include assessing the extent of strongyloidiasis morbidity , including growth retardation and malnutrition , with regard to infection intensity—for which standards have yet to be established , and estimating the risk of severe strongyloidiasis and hyperinfection in endemic settings . S . stercoralis is not currently addressed by the WHO control strategy against STH that relies on “preventive chemotherapy” , i . e . regular mebendazole or albendazole treatment of specific at-risk groups or mass-drug administration ( MDA ) [52 , 53] . Single dose benzimidazoles have suboptimal effects on S . stercoralis , for which ivermectin is the drug of choice . This drug is highly efficacious at a single oral dose of 200 μg/kg body weight ( BW ) and is well tolerated . Moreover , ivermectin is also efficacious against Ascaris lumbricoides . In combination with benzimidazoles , it improves therapeutic outcomes against Trichuris trichiura , while in combination with albendazole , it improves therapeutic performances against hookworm [14–16 , 54 , 55] . In the absence of infection intensity figures , the similarity of reinfection rates and morbidity across age groups support arguments for community-wide control [13] . However , the long-term impact of malnutrition on childhood development could justify integrating S . stercoralis control into ongoing school-based STH control programs , which are well established throughout the country and currently target children from infancy to high school . Monitoring the impact of control on infection levels in various transmission settings would help to assess whether , where and how control measures should be extended , while optimizing cost-effectiveness . The cost of ivermectin poses a challenge to expanding its use . In Cambodia , the drug is neither donated nor subsidized and treating one individual with quality tablets produced by a certified manufacturing company costs 20 to 40 USD , depending on the patient’s weight . The high prevalence of the parasite and its significant morbidity clearly advocate for increased donations or the production of generic ivermectin so S . stercoralis control can be implemented without further delay in Cambodia . | Strongyloides stercoralis is an intestinal parasite that infects humans by penetrating intact skin . It thrives particularly in tropical countries with poor sanitation . Because it can replicate within its host , it causes long-lasting infections and is potentially fatal in patients with a disseminated infection . S . stercoralis is largely neglected due to the difficulty in detecting it with standard field diagnostic techniques but has recently been found to be very common in Cambodia , with prevalence rates exceeding 40% . It is difficult to identify symptoms associated with infection in endemic areas because co-infections with other helminths or protozoan parasites , which cause similar health problems , are common . We compared clinical signs in infected vs . non-infected participants living in eight villages in Northern Cambodia , and before and after treatment with ivermectin , the drug of choice against S . stercoralis , among 103 patients infected with S . stercoralis only . We also assessed the association between infection and growth retardation among children and adolescents . Of the participants , 31 . 1% were infected with S . stercoralis . Infected participants were more likely to report itching and urticaria . After treatment , fewer participants reported urticaria , abdominal pain , vomiting and , to a lesser extent , nausea , diarrhea , cough , and tiredness . S . stercoralis infection was associated with growth retardation as expressed by stunting . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"dermatology",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"respiratory",
"infections",
"urticaria",
"helminths",
"parasitic",
"diseases",
"animals",
"pulmonology",
"strongyloides",
"stercoralis",
"strongyloides",
"parasitic",
"intestinal",
"diseases",
"protozoan",
"infections",
"helminth",
"infections",
"eukaryota",
"nematoda",
"biology",
"and",
"life",
"sciences",
"organisms"
] | 2017 | Strongyloides stercoralis is associated with significant morbidity in rural Cambodia, including stunting in children |
All organisms wishing to survive and reproduce must be able to respond adaptively to a complex , changing world . Yet the computational power available is constrained by biology and evolution , favouring mechanisms that are parsimonious yet robust . Here we investigate the information carried in small populations of visually responsive neurons in Drosophila melanogaster . These so-called ‘ring neurons’ , projecting to the ellipsoid body of the central complex , are reported to be necessary for complex visual tasks such as pattern recognition and visual navigation . Recently the receptive fields of these neurons have been mapped , allowing us to investigate how well they can support such behaviours . For instance , in a simulation of classic pattern discrimination experiments , we show that the pattern of output from the ring neurons matches observed fly behaviour . However , performance of the neurons ( as with flies ) is not perfect and can be easily improved with the addition of extra neurons , suggesting the neurons’ receptive fields are not optimised for recognising abstract shapes , a conclusion which casts doubt on cognitive explanations of fly behaviour in pattern recognition assays . Using artificial neural networks , we then assess how easy it is to decode more general information about stimulus shape from the ring neuron population codes . We show that these neurons are well suited for encoding information about size , position and orientation , which are more relevant behavioural parameters for a fly than abstract pattern properties . This leads us to suggest that in order to understand the properties of neural systems , one must consider how perceptual circuits put information at the service of behaviour .
As with many animals , vision plays a key role in a number of behaviours performed by the fruit fly Drosophila melanogaster , including mate-recognition [1] , place homing [2] , visual course control [3] , collision-avoidance [4] , landing [4] and escaping a looming object ( like a rolled newspaper , for example ) [5] . The benefit of studying these visually guided behaviours in Drosophila is the range of neurogenetic techniques which give a realistic chance of understanding the neural circuits that underpin them . With that goal in mind , we focus on work by Seelig and Jayaraman [6] which mapped the receptive fields ( RFs ) of a set of visually responsive neurons: the ring neurons of the ellipsoid body . These neurons are necessary and sufficient for a range of complex behaviours , including short term spatial memory , pattern discrimination and place memory [2 , 7–9] , and yet are surprisingly small in number . To understand their role in these behaviours , we used modelling to bridge the gap between neurogenetic data and behaviour by evaluating ring neuron responses during simulations of fly experiments . In this way we investigate how small populations of visual neurons in Drosophila , which might represent a sensory bottleneck , can still provide behaviourally relevant information . In laboratory assays , flies show interesting spontaneous visual behaviours . For instance , flies orient towards bar stimuli [10 , 11] and in a circular arena with two diametrically opposed bars will walk between them until exhaustion [12] . The attraction to vertical bars decreases as the bar is shortened and flies are strongly repulsed by small spots [13] . In addition , a number of studies have investigated the process of pattern recognition and its neural underpinnings [7 , 14 , 15] . Flies seem to possess a form of pattern memory analogous to the better-studied pattern memory of bees [16–18] . Interestingly , both bees [19] and flies [14] systematically fail to discriminate certain pattern pairs . These visual behaviours require the central complex , a major neuropil which comprises the ellipsoid body , the fan-shaped body , the paired noduli and the protocerebral bridge [20] . The central complex is thought to be involved primarily in spatial representation , action selection and mediation between visual input and motor output [21] . One class of neurons with projections in the ellipsoid body is the ‘ring neurons’ , which are known to be involved in certain visual behaviours ( R1: place homing [2 , 22 , 23]; R2/R4m: pattern recognition [7 , 14 , 15]; R3/R4: bar direction memory [8] ) . Here we investigate how the ring neurons might contribute to behaviour , by simulating the visual input as it would be processed by this small population of visually responsive cells . In particular , we can address why flies are unable to discriminate certain pattern pairs , whether these subpopulations of neurons are optimised for pattern recognition and , if not , what visually guided behaviours these cells are suited to . In order to do this , we leverage research which has described the RF properties of two classes of ring neuron in the Drosophila ellipsoid body [6] . The two subtypes of neuron investigated were the R2 and the R4d ring neurons , of which only 28 and 14 , respectively , were responsive to visual stimuli . The cells were found to possess RFs that were large , centred in the ipsilateral portion of the visual field and with forms similar to those of mammalian simple cells [24] ( for details of how the RFs were estimated , see Materials and methods ) . Like simple cells , many of these neurons showed strong orientation tuning and some were sensitive to the direction of motion of stimuli . The ring neuron RFs , however , are much coarser than those of simple cells , far larger and less evenly distributed across the visual field and respond mainly to orientations near the vertical . This suggests that ring neurons might have a less general function than simple cells [25] . In mammals , the very large population of simple cells means that small , high-contrast boundaries of any orientation are detected at all points in the visual field . Thus the encoding provided by simple cells preserves visual information and acts as a ‘general purpose’ perceptual network that can feed into a large number of behaviours . In contrast , the coarseness of the ring neuron RFs , allied to the tight relationship between specific behaviours and specific subpopulations of ring neurons , suggests instead that these cells are providing economical visual information that is likely tuned for specific behaviours [25] . To investigate such issues , we use a synthetic approach whereby investigations , in simulation , of the information provided by these populations of neurons can be related to behavioural requirements , thus ‘closing the loop’ between brain and behaviour . We show how the population code is well-suited to the spontaneous bar orientation behaviours shown by flies . Similarly , we verify that our population of simulated ring neurons is able to explain the success and failure of the fly to discriminate pairs of patterns . Upon deeper analysis , we demonstrate that certain shape parameters—orientation , size and position—are implicit in the ring neurons’ outputs to a high accuracy , thus providing the information required for a suite of basic fly behaviours . This contrasts with the rather limited ability of ring neuron populations ( and flies ) to discriminate between abstract shapes , casting doubt on cognitive explanations of fly behaviour in pattern discrimination assays .
We first consider experiments in which flies are presented with bar stimuli , as flies are known to spontaneously orient towards black bars [11] , aiming for the centres of narrow bars and the edges of wide bars [27] . We therefore decided to examine the responses of simulated ring neurons to bars of different widths ( Fig 1A and 1B ) . The summed outputs of the ensembles of ring neurons show peaks to the bars of different widths , which broadly matches experimental results ( Fig 1B ) . For instance , R2 neurons respond maximally to the inside edges of large bars , while peak activity in R4d neurons occurs at bar centres and also at roughly ±90° . While we do not know the details of mechanisms downstream of the ring neurons and hence how their activity is transformed into action , the simulation is an existence proof that the information needed to control the observed behaviour is present in the sparse ring neuron code . We further demonstrate this point by closing the loop between sensory systems and behaviour using a simple model of a fly viewing a bar in which the fly’s heading is controlled by the difference between the summed activation of left and right ring neurons ( Fig 1C; see Materials and methods for details ) . The simulated fly approaches the bar from different distances , demonstrating centre-aiming when far from the bar and fixation of the edges when it is nearer and the bar’s apparent size is thus greater ( Fig 1D ) . Through this example , we can see how the information present in this small population of visually responsive ring neurons can control a specific behaviour . We now turn to a more complex behaviour: pattern discrimination . The standard paradigm for testing pattern discrimination involves putting a fly into a closed-loop system where it is tethered inside a drum , on the inside of which are two different visual patterns , alternating every 90° , giving four visual stimuli in total [7 , 14 , 15 , 28] ( see Fig 1E ) . As the fly attempts to rotate in one direction , the drum rotates in the other , giving the fly the illusion that it is moving in a stable world . To elicit conditioned behaviour , if the fly faces one of the four pattern stimuli it is punished by a heat beam . Over time , if the fly is able to differentiate the patterns , it should preferentially face the unpunished pattern . This procedure has been used to demonstrate that flies can differentiate stimulus pairs such as upright and inverted ‘T’ shapes , a small and a large square , and many others [14] . The ability to discriminate patterns in such an assay requires R2 neurons [7 , 14 , 29] . More specifically , synaptic plasticity afforded by rutabaga in these neurons is sufficient and necessary for observed pattern learning [15] . We therefore investigate the responses of ring neurons in simulations of the classic pattern discrimination paradigm . To recreate the visual information perceived by flies in such experiments , we simulated the typical experimental flight arena with a fly tethered in the centre . We then examined the output of the ensembles of ring neurons for a fly rotating in the drum and looked at the difference in the activation code when the agent was facing the different patterns of a pair . Our logic is that if the ensemble codes were identical , it would be impossible for the patterns to be discriminated by interrogating the outputs of ring neurons alone . Similarly , the greater the difference in the ring neuron ensemble activation codes when looking at the pattern pairs , the easier they would be to discriminate ( Fig 1F and 1G; see Materials and methods for details ) . Our discriminability measure is the root mean square ( r . m . s . ) difference between ensemble outputs when the ( virtual ) fly faces different azimuths in the drum . In this way , we can compare the ensemble output when the ‘fly’ is oriented at 0° ( i . e . with the view centred on one pattern ) and the ensemble output at other azimuths ( Fig 1 ) . We henceforth treat this as a measure of ‘discriminability’ of patterns , following the experimental work that we are modelling , though of course in reality an animal’s ability to discriminate stimuli is not an absolute value and varies depending on many factors , including task and training procedure [30] . The r . m . s . difference , as compared to the view at 0° , rises as the fly rotates in the drum , peaking as it faces the space in between the patterns and dropping to a minimum when facing the centre of the next pattern ( Fig 1F and 1G ) . For some pairs of patterns , there is still an appreciable r . m . s . difference between the codes when facing the centres of each pattern , thus enabling their discrimination . However , in the example of Fig 1F and 1G , if we displace the patterns vertically , we see a drop in the r . m . s . difference between activation codes when the fly fixates the patterns . This is despite the fact that , to the human eye , the patterns still appear very different . Interestingly , the pattern pair in Fig 1G is also harder to discriminate for flies . In this way , we can use the difference between ensemble codes when flies face the patterns to re-examine the discriminability of pattern pairs tested with flies . One illustrative example is shown in Fig 2 ( see pattern set ( 9 ) in Fig 3 ) , which contains pairs of ‘triangles’ , one facing up and the other down . Drosophila are able to discriminate these pattern pairs when they are aligned along the top and bottom , but not when aligned about the vertical centres of mass [14] . Looking at the placement and form of the R2 RFs allows us to determine where this difference comes from ( Fig 2 ) . The excitatory regions of the RFs fall roughly across the middle of triangles that are not aligned about their vertical centres of mass and therefore the difference in width at this point will lead to differences in activation . If the triangles are offset ( Fig 2B ) so as to be aligned about their vertical centres of mass , their width will be similar for the regions of peak R2 coverage and the difference in activation will be lower . Thus the failure to discriminate features with an equivalent vertical centre of mass can be explained by the shape of the RFs interacting with the patterns directly . It is not necessary to invoke an additional system that extracts and compares the vertical centres of mass of the patterns . Performance on poorly discriminated patterns can be improved , however , by simply adding more RFs of the same form . Fig 2D and 2E show the increase in performance with number of RFs for two such pattern pairs: triangles and triangularly shaped horizontal bars aligned about the centre of mass ( from pattern set ( 9 ) in Fig 3; see Materials and methods for details ) . This demonstrates that the patterns could be discriminated by flies simply with the addition of more RFs centred on other portions of the visual field . Similarly , pattern set ( 2 ) in Fig 3 gives examples of pattern pairs that are not discriminable by flies and also give only small differences in the outputs of R2 filters . This may seem surprising , given that these patterns appear quite different to human observers and are also very dissimilar if compared retinotopically . Thus we can see that the R2 ring neuron encoding is informationally sparse . Whilst the V1 region of human visual cortex contains neurons representing the full range of orientations across the visual field , R2 neurons have large RFs and poor orientation resolution . Hence , a pattern pair consisting of a diagonal line facing left and a diagonal line facing right , for example , have only a small difference in R2 outputs in our simulation and are also not discriminable by flies . This could , in the light of behavioural experiments alone , be interpreted as evidence that flies do not discriminate patterns on the basis of orientation . A more parsimonious explanation , however , is that the flies are failing because the form of the RFs means that the output code is similar for these particular orientations . To emphasise the independence of apparent similarity of patterns and the visual encoding from R2 cells , we designed shape pairs that appear similar to humans , but are easily discriminable by the R2 population ( white bars in Fig 2F ) , as well as shape pairs that are considered similar by the R2 population but not by human observers or in terms of retinal overlap ( black bars in Fig 2F; see Materials and methods for details ) . Despite the similarity between the pairs of patterns , the first is readily discriminable , especially from the outputs of glomeruli 1 , 3 , 5 and 11 , while the second pair—which we easily see as having a different orientation—has very low overall differences across the glomeruli . This shows that the irregular RF shapes can lead to counterintuitive results . The small population of visually responsive R2 neurons can be thought of as a sensory bottleneck . If the information that passes through this bottleneck is all that a fly has available for pattern discrimination , then we should see a close relationship between the r . m . s . difference in simulated R2 output for a pattern pair and the flies’ ability to learn to discriminate that pair . We thus examined the difference in the outputs of the R2 filters between patterns from pairs drawn from work by Ernst and Heisenberg [14] ( Fig 3 ) . In general , the pattern pairs for which flies show a significant learned discrimination have a greater r . m . s . difference in R2 population activity [14] . All of the pattern pairs where flies show significant learning ( n = 8 ) have R2 r . m . s . differences above the overall mean ( Fig 3A and 3B ) , whereas 13 out of 18 patterns that flies found more difficult to learn had below-average r . m . s . differences ( there were nine pattern pairs for which a significance level was not given that were excluded . ) Across all pattern pairs , we find a significant correlation between the strength of the learning index reported for flies in [14] and the r . m . s . difference in R2 activation ( Spearman’s rank , n = 30 , ρ = . 420 , p < . 05 ) . Of course , these differences could simply result from the apparent similarity of the patterns . Therefore , as a control comparison , we quantified the similarity of pattern pairs based on the degree to which the patterns overlap in a pixel-by-pixel manner ( see Materials and methods ) . There was no significant correlation with the flies’ learning index over the pattern pairs ( Spearman’s rank , n = 32 , ρ = − . 068 , p = n . s . ) . We additionally looked at the relationship between the two visual similarity metrics ( R2 population code and pixelwise retinal overlap ) and the degree to which flies show a spontaneous preference ( i . e . without any conditioning ) for one of the patterns within a pair ( Fig 3D and 3E ) . There was no correlation for R2 population codes ( Spearman’s rank , n = 29 , ρ = . 289 , p = n . s . ) , but for retinal overlap there was a weakly significant correlation ( Spearman’s rank , n = 29 , ρ = − . 371 , p < . 05 ) . This is consistent with research showing that R2 neurons alone are critical for learned pattern differences [14] , but not spontaneous preferences which , by contrast , seem to result from activity across all subsets of ring neurons [31] . There are , however , some discrepancies where the learning performance of flies for a particular pattern pair does not match the r . m . s . difference of our R2 population code . In some cases flies are better at discriminating pairs of patterns that differ along the vertical rather than horizontal axis ( set ( 3 ) vs set ( 4 ) , and the pairs in set ( 12 ) , marked with red Xs in Fig 3 ) . In contrast , the r . m . s . difference in the R2 population code discriminates horizontal and vertical patterns equally . This is because while our R2 filters are presented with static stimulus pairs to simulate a fly facing the centre of a pattern , for real flies the patterns were moving horizontally but fixed in the vertical axis making it harder for flies to resolve horizontal information [14] . Overall , we have shown that the behavioural performance of flies on a pattern discrimination task is approximated by a simple discriminability metric applied to the population activity of a small number of simulated R2 cells . There were , however , a number of seemingly ‘easy’ pattern pairs which neither flies nor the simulated population of R2 cells , perhaps surprisingly , could discriminate . On further investigation , we found that performance for poorly discriminated pattern pairs could be improved with the addition of extra R2-type RFs . Thus , it seems likely that the pattern discrimination capability of a set of R2-like neurons could easily have been improved over evolutionary time with the simple addition of more cells and we therefore suggest that there must have been little selection pressure specifically for a specialised pattern recognition module in fruit flies . Information from 3000 ommatidia is funnelled to just 28 R2 and 14 R4d ring neurons , yet these cells are able to support a number of complex behaviours . We have shown how the R2 population code provides sufficient information to discriminate some pattern pairs , and also that , as performance could be improved with the addition of more ring neurons , general-purpose pattern recognition seems unlikely to be the purpose of the ring neuron system . So what information is this system tuned to extract ? Examining the pattern pairs which flies and the R2 population were able to discriminate , we see that certain pattern parameters are implicitly coded for in the R2 population . Pattern sets ( 6 ) and ( 9 ) ( Fig 3 ) suggest that , for instance , stimulus size and vertical centre of mass are parameters that can be recovered from the R2 population code after this sensory bottleneck . We now address in more general terms the question of what shape information is implicitly conveyed in the ring neuron population code . To do this , we generated large sets of ellipse-like ‘blob’ stimuli varying in size ( specified by major-axis length ) , position ( azimuth and elevation ) and orientation . The blob generation procedure was stochastic and so the precise shape of each blob was random and unique ( see Materials and methods ) . We then trained an artificial neural network ( ANN ) to recover this shape information from either a raw image of the shape ( the control condition ) or from the output of the R2/R4d populations on presentation of the blobs . We are using ANNs here as statistical engines interrogating the output of the ring neuron population code to determine if shape information is implicit to the code and has therefore passed through the sensory bottleneck . We first examined whether ANNs could be trained to extract positional information ( the elevation and azimuth ) of randomly generated blobs . Note that as the blobs are initially aligned about their centres of mass , elevation is equivalent to vertical centre of mass , except where the blobs are partially outside the visual field . The blobs varied along four parameters: elevation ( ≥ −60° and ≤ 60° ) , azimuth ( ≥ −135° and ≤ 0° ) , orientation ( ≥ 0° and ≤ 90° ) and major-axis length ( ≥ 12 . 79° and ≤ 60° ) . Each parameter had 22 possible values , giving a total of 234 , 256 ( = 224 ) stimuli . Of these , approximately 40% ( n = 93 , 702 ) were used for training and the remainder ( n = 140 , 554 ) for testing . Results for the test set ( Fig 4A–4D ) show that ANNs are indeed able to extract information about elevation and azimuth from any of the input types ( ‘raw view’ , ‘R2’ , ‘R4d’ or ‘R2 + R4d’ ) . Performance was better with parameter values near the middle: at the extremes , portions of the stimuli lay outside the visual field of the simulated fly , meaning stimuli begin to disappear ‘off the edge’ of the visual field ( Fig 4A and 4B ) , making the task harder ( i . e . , is this a large object projecting outside the visual field , or a smaller object at the edge ? ) . While performance was best with raw views as inputs ( Fig 4C and 4D ) , positional information could still be reliably extracted from ring neuron outputs . The R2 code performs better than the R4d and the addition of R4d RFs to the R2 code ( ‘R2 + R4d’ ) , while adding dimensionality , does not improve performance , suggesting that either an R2-like encoding is sufficient to extract positional information , or that the information in the two codes is redundant . Thus small populations of ring neurons retain positional information . We next trained ANNs to decode information about stimulus orientation and size . The stimuli were random blobs , as before , with the same possible values for elevation , orientation and size . This time , however , azimuthal position was fixed at −90° . The reason for this was that the neural network struggled to encode information about orientation when azimuth also varied , presumably because the centres of the receptive fields—and thus the position on the visual field where they can best extract information—are clustered at around −90° . For this experiment there were therefore 10 , 648 ( = 223 ) stimuli , of which approximately 40% ( n = 4259 ) were used for training and the remainder ( n = 6389 ) for testing . The ANNs were able to extract this shape information from both raw images and the ring neuron outputs ( Fig 4E–4H ) . Orientation was the parameter with the highest error score , possibly because its calculation requires a second-order statistic ( the covariance of the shape ) . Nonetheless , both parameters could be simultaneously estimated by an ANN neural network fed with ring neuron outputs . In summary , we have shown that information about a number of shape properties passes through the bottleneck created by the small number of ring neurons . This indicates that such information is available downstream of the ring neurons for the guidance of behaviour .
One striking feature of the ring neuron receptive fields is that they are in general tuned to vertically oriented objects . We know that fruit flies are strongly attracted to vertical bars , a finding that has been leveraged across a range of behavioural paradigms ( e . g . bar fixation: [8] ) . In one , individual flies are placed into a virtual-reality arena with two vertical stripes 180° apart: flies will typically head back and forth between the two bars repeatedly . Occasionally , when a fly crosses the arena’s midline , the bars disappear and a new bar is presented at 90° to the originals , to which the flies reorient . The new target then also disappears , and the flies resume their initial heading , even though the original bar is no longer visible . This indicates that directional information is stored in short-term memory and updated . Work by Neuser and colleagues [8] has shown that R4 ( and R3 ) ring neurons are involved in this spatial orientation memory . We found that both R2 and R4d neurons were responsive to vertical bars of varying widths , mimicking flies’ preference for the edges of larger bars and the centres of narrower ones [27] . We also showed that the cells provide sufficient information to guide homing towards a large vertical object and , separately , that the azimuth of bar stimuli makes it through the sensory bottleneck . Taken together , these findings demonstrate a viable role for the small R4d population in the behaviours described above . The more general role of R4d cells within the central complex is still unknown . There is evidence that R4d neurons are able to act as a ring attractor , maintaining a stable encoding of the fly’s orientation with respect to a landmark [9 , 32] . Therefore , R4d neurons could be conceived variously as functioning like mammalian head-direction cells [33] , playing a part in a path integration system [8] or in conditioning of visual orientation [34] . These possibilities are not mutually exclusive , of course , and their true function ( or functions ) will become apparent only with a better understanding of the behaviours in which they are involved . Drosophila can discriminate patterns differing in size , orientation and elevation and other complex shape parameters , an ability for which R2 cells are critical [7 , 14 , 15] . We have shown that the discriminability of a given pattern pair is predicted by the outputs of the small population of R2 cells , which have coarse receptive fields and therefore do not encode higher-order visual properties explicitly . Does this limited ability of the R2 population ( and , of course , the fly ) to discriminate patterns suggest that flies might be a good model for the study of a universal perceptual process of pattern recognition , or might limited pattern recognition be an artefact of a perceptual system tuned to other tasks ? Any selection pressure on flies’ ability to discriminate patterns ( as bees need to do , for instance ) would surely have led to a larger R2 population or , possibly , visual input to the mushroom body [35 , 36] , and we can therefore be confident that ring neurons have not been tuned for arbitrary , general-purpose pattern recognition . Accordingly , we must suggest caution if research on flies is used with the aim of understanding the neural basis of pattern recognition or even visual cognition more generally [37] . So what behaviours are served by the information that makes it through this sensory bottleneck ? It is interesting to consider to what extent Drosophila’s ecological needs are served by general learning mechanisms—such as a capacity to learn arbitrary visual stimuli—and to what extent by domain-specific abilities . For example , bees have a well-attested ability to learn many varied patterns , which presumably derives from a need to learn about flowers [38]; it is not apparent , however , that there has been a comparable selection pressure on Drosophila for such general-purpose learning . Across the animal kingdom there are many cases where a task-specific heuristic can provide an elegant solution . For example , male fiddler crabs ( Uca pugilator ) treat salient objects above the horizon as predators and everything below as conspecifics [39] . Similarly , Drosophila have a mechanism to approach bars and to avoid small objects [13]; presumably to approach vegetation ( for oviposition , etc . ) and avoid predators , respectively . In order to fully understand these circuits we need to examine further how flies depend on a balance of innate visual responses versus learned visual information . So , if the R2s are not truly ‘pattern-recognition cells’ , the question remains: what are they for ? Though we have not attempted to answer this question here , we have shown that there is implicit information about higher-order properties , such as stimulus position , orientation and size , in the RFs’ code , which could drive any number of natural behaviours . For example , elsewhere we have shown that the information content of ring neuron RFs is suitable for place learning and homing [26] , and although this behaviour in flies involves a subset of ring neurons other than those examined here ( R1 ) , it gives an indication of how small populations of coarse , wide-field cells can be used to drive behaviour . The goal of this work was to investigate the information encoded in a population of visually responsive ring neurons , in simulations of classic pattern discrimination assays . Our aim was to examine the behavioural uses to which the information encoded in this population of cells could be put by a fruit fly . Of course , a full understanding of these neurons requires detailed knowledge of how they interact with other neural circuits for behaving flies in natural environments . Hence future work needs to address the interaction between brain , behaviour and environment [40] . For the brain , a sensible starting point is to ask how ring neurons and the information they carry are integrated in the central complex circuitry . Recent work has shown the presence of a ring attractor network [9 , 32 , 41 , 42] in the ellipsoid body of the central complex which integrates both visual and proprioceptive information . This circuit is able to retain a heading in short-term memory [8] and thus the cells we have modelled could be useful in contributing information about the position of behaviourally relevant objects . Of course , there are many details to be determined , such as the dynamics of neural coding in this circuit and the sensory pathways that lead to the observed receptive fields [43] . In the current study we have not considered neural dynamics and have assumed that the information would be extracted as rate codes . While this is a common assumption for models of visual perception ( e . g . [24] ) we note that information could be extracted via a timing code , perhaps even more efficiently—especially if the fly is actively perceiving its environment . Though it is possible to convert from an analogue neural network to a spiking neural network [44] , more work would be needed to establish this . Finally , the story is complicated further by the sheer variety of behaviours in which these cells have been implicated: for example , different subsets of R2 neurons have also been implicated in an olfactory decision task [45] and in sleep drive [46] . The sensory ecology of fruit flies is still largely a mystery ( see [47] ) , despite the immense promise and productivity of Drosophila neuroscience research . We thus know relatively little about ‘natural’ Drosophila visual behaviours—in contrast to bees , for which we know much about behaviour but comparatively little about the nervous system . One pertinent example of this is visual pattern recognition , where for bees we have a good understanding of the real-world challenge facing a forager . This has enabled models of pattern recognition to be developed for bees [48] . Without a detailed understanding of sensory ecology and the natural behaviour of flies , it is hard to understand what type of pattern vision flies might need . However , some fly behaviours are easier to relate to the natural environment . Flies show an innate attraction to long bar-like objects , on which they might perch [13] and in Fig 5 we show one example of how behaviourally relevant information is maintained in the output of R4d cells even for complex scenes . Taking this further , we could consider how an active fly in a complex environment might be able to shape its own visual input to give us a better understanding of the true potential of the fly as a pattern discriminator . More generally , it is only with a deeper knowledge of fruit fly ecology that we will be able to close the loop between brain , body and environment and thus obtain a full understanding of the whole system . One of the advantages of studying insects is the potential for describing their neural processes with modelling . In this way , simulations can help bridge the gap between biology and behaviour [49] . We have shown that the sensory bottleneck produced by small populations of cells is not a barrier to the specific information that is required for particular behaviours . However , this modelling work , and the neuroscience that invited it , do suggest caution when proposing that flies possess general-purpose visual cognition . We thus hope that future experiments , grounded in both the ecological needs of the animal and the information given by neural circuits , will be able to better inform the next generation of models , and vice versa .
The goal of Seelig and Jayaraman’s work [6] was to examine responses of lateral triangle microglomeruli ( which house the cell bodies of the ring neurons ) to visual stimuli . For this , they employed two-photon calcium imaging to examine the activity of genetically targeted subsets of microglomeruli , the R2 and R3/R4d neurons . Fluorescence was recorded for head-fixed flies held in an arena with a curved display composed of an LED array . In order to map the RFs , the flies were presented with a series of flashing dots at random locations on the visual display; the fine structure of the RFs was then revealed by using white-noise stimuli [50] . The accuracy of the estimated receptive fields was then verified by correlating predicted with actual responses to novel bar stimuli ( and a high degree of correspondence was found ) . The predicted responses were calculated by using the RFs as linear filters through convolution and so we follow a similar procedure here . To create the visual filters which represent the RFs , we first extract the image representations of the RFs from Seelig and Jayaraman ( Extended Data Figure 8 in [6] ) . This gives us images of 112 × 252 pixels for R2 neurons and 88 × 198 pixels for R4d . Given the visual field is taken as 120° × 270° , this corresponds to a resolution of 1 . 07° and 1 . 36° per pixel , respectively . As data is given for multiple flies , we averaged the RFs for the different glomeruli across flies ( 2 ≤ N ( R2 ) ≤ 6 , 4 ≤ N ( R4 ) ≤ 7 ) . This process is summarised in Fig 6 . Each pixel on the extracted image is initially assigned a value ranging from –1 for maximum inhibition to 1 for maximum excitation , based on the values given by the colour scale bars in [6] . These images are then thresholded to give a kernel g ( i , j ) : g ( i , j ) = { 1 for R i , j ≥ T ; - 1 for R i , j ≤ - T ; 0 otherwise . where g ( i , j ) is the ( i , j ) th pixel of the kernel , Ri , j is the ( i , j ) th value of the processed RF image and T is the threshold value , here 0 . 25 ( Fig 6A ) . We take the centroid of the largest excitatory region as the ‘centre’ of each of the kernels . The excitatory region is then extracted using MATLAB’s bwlabeln function ( with eight-connectivity ) and its centroid , ( x , y ) , with the regionprops function . The mean centroid , ( x ¯ , y ¯ ) , across flies is then calculated and the kernels are all recentred on this point: g ^ ( i , j ) = { g ( i + y - y ¯ , j + x - x ¯ ) for 1 ≤ i + y - y ¯ ≤ m and 1 ≤ j + x - x ¯ ≤ n ; 0 otherwise . where g ^ ( i , j ) is a recentred kernel ( Fig 6C ) . In order to calculate the activation for a given RF on presentation of an image the RF must first be resized to have the same number of pixels as the image . This is accomplished by resizing the average RF , g ¯ ( i , j ) , using Matlab’s imresize function with bilinear interpolation and then scaled to [−1 , 1] . Finally , the filter is thresholded and the excitatory and inhibitory regions are assigned different normalised values: K i , j = { g ¯ ( i , j ) ÷ S exc , for g ¯ ( i , j ) > 0 ; - g ¯ ( i , j ) ÷ S inh , for g ¯ ( i , j ) < 0 ; 0 , otherwise . where Sexc and Sinh indicate the sums of excitatory and inhibitory pixels , respectively . This method of normalising values has the result that the activation ( see below ) for an all-white or -black image will be zero . Other normalisation schemes are possible , but the choice is somewhat arbitrary , as we are only interested in the differences in output values . Furthermore , RFs are sensitive to contrast differences , so a zero-sum filter , as seen in edge detectors , is appropriate . Additionally , assigning biologically relevant values is not possible because of a lack of data . The activation of an average kernel , K , to the presentation of a greyscale image , I , at rotation θ , is then: A ( I , K , θ ) = ∑ i = 1 m ∑ j = 1 n I i , j ( θ ) K i , j , where 0 ≤ I i , j ( θ ) ≤ 1 ( 1 ) where Ii , j ( θ ) and Ki , j are the ( i , j ) th pixels of the image and kernel , respectively . The equation for describing the bar fixation mechanism shown in Fig 1C is as follows: ϕ turn = G · π 4 ( ∑ K ∈ G left max ( 0 , A ( I , K , 0° ) ) - ∑ K ∈ G right max ( 0 , A ( I , K , 0° ) ) ) where I is the view of the bar from the agent’s current location and Gleft and Gright are the sets of left- and right-side filters . ‘G’ is a parameter to control the gain of the system , and here was set to 2 . For the pattern recognition tasks ( see Fig 3 ) , the difference in activation is calculated as follows: D ( I ) = ∑ K ∈ G ( A ( I , K , 0° ) - A ( I , K , 90° ) ) 2 | G | where G is the set of R2 filters , I is the current pattern pair and A ( ⋅ , ⋅ , ⋅ ) is the activation of the kernel to the pattern , as described in Eq 1 . The choice of r . m . s . difference as a difference function is somewhat arbitrary , but r . m . s . difference is commonly used ( e . g . [51] ) ; alternatively one could use mean absolute difference or mutual information etc . , although the choice is not critical as we are looking at relative differences and the RF output code is normalised . The retinal overlap for two binary patterns , A and B , is calculated in two steps . Firstly we measure the number of overlapping pixels between A and B; this value is referred to as Q . Next the proportion of pixels which this overlap represents for A and B is calculated and , finally , we calculate the retinal overlap as the average of these two values: O ( A , B ) = Q 2 ( 1 ∑ i = 1 m ∑ j = 1 n A i , j + 1 ∑ i = 1 m ∑ j = 1 n B i , j ) where Ai , j and Bi , j represent the ith , jth pixel of patterns A and B , respectively . We also carried out simulations of pattern discrimination where the number of R2 RFs was varied between 28 , 56 and 112 ( Fig 2D and 2E ) . The ‘28-kernels’ condition simply used the original kernels in their original positions ( see Fig 6D ) . For the 56- and 112-kernel conditions ( double and quadruple the number of kernels , respectively ) , the original kernel types and positions were used for the first 28 kernels for every trial and the remainder were placed in random locations , with the only constraint being that the ‘centre’ of the new kernel could not be within 10° of the centre of any already-placed kernel . Equal numbers of each kernel type ( corresponding to each one of the original R2 filters ) were used in each condition; that is , each kernel type was repeated twice for the 56-kernel condition ( once in the original position and once in a random position ) and four times for the 112-kernel condition ( once in the original position and three times in a random position ) . The kernels were also shrunk for the latter two conditions by a factor of 1 2 and 1 2 , respectively , to keep the sum of the retinal area covered by all kernels constant across conditions , although the shapes of the kernels were otherwise kept the same . As the kernels were in fixed positions for the first condition , only one test was performed; for the other two conditions , 1000 trials were carried out . The ANNs were implemented using the Netlab toolbox for MATLAB . All networks were two-layer feedforward networks , with 10 hidden units and a linear activation function for the output units . There were 100 training cycles and optimisation was performed with the scaled conjugate gradient method . | A general problem in neuroscience is understanding how sensory systems organise information to be at the service of behaviour . Computational approaches can be useful for such studies as they allow one to simulate the sensory experience of a behaving animal whilst considering how sensory information should be encoded . In flies , small subpopulations of identifiable neurons are known to be necessary for particular visual tasks , and the response properties of these populations have now been described in detail . Surprisingly , these populations are small , with only 14 or 28 neurons each , which suggests something of a sensory bottleneck . In this paper , we consider how the population code from these neurons relates to the information required to control specific behaviours . We conclude that , despite previous claims , flies are unlikely to possess a general-purpose pattern-learning ability . However , implicit information about the shape and size of objects , which is necessary for many ecologically important visually guided behaviours , does pass through the sensory bottleneck . These findings show that nervous systems can be particularly economical when specific populations of cells are paired with specific visual behaviours . This is a general-interest finding for computer vision and biomimetics , as well as sensory neuroscience . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"pattern",
"recognition",
"receptors",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"immunology",
"social",
"sciences",
"neuroscience",
"animals",
"artificial",
"neural",
"networks",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"cognition",
"artificial",
"intelligence",
"experimental",
"organism",
"systems",
"computational",
"neuroscience",
"vision",
"drosophila",
"neuronal",
"tuning",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"proteins",
"behavior",
"insects",
"arthropoda",
"biochemistry",
"signal",
"transduction",
"cellular",
"neuroscience",
"psychology",
"eukaryota",
"cell",
"biology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"sensory",
"perception",
"immune",
"receptors",
"cognitive",
"science",
"computational",
"biology",
"organisms"
] | 2017 | Neural coding in the visual system of Drosophila melanogaster: How do small neural populations support visually guided behaviours? |
Host-range shifts in influenza virus are a major risk factor for pandemics . A key question in the study of emerging zoonoses is how the evolution of transmission efficiency interacts with heterogeneity in contact patterns in the new host species , as this interplay influences disease dynamics and prospects for control . Here we use a synergistic mixture of models and data to tease apart the evolutionary and demographic processes controlling a host-range shift in equine H3N8-derived canine influenza virus ( CIV ) . CIV has experienced 15 years of continuous transfer among dogs in the United States , but maintains a patchy distribution , characterized by sporadic short-lived outbreaks coupled with endemic hotspots in large animal shelters . We show that CIV has a high reproductive potential in these facilities ( mean R0 = 3 . 9 ) and that these hotspots act as refugia from the sparsely connected majority of the dog population . Intriguingly , CIV has evolved a transmission efficiency that closely matches the minimum required to persist in these refugia , leaving it poised on the extinction/invasion threshold of the host contact network . Corresponding phylogenetic analyses show strong geographic clustering in three US regions , and that the effective reproductive number of the virus ( Re ) in the general dog population is close to 1 . 0 . Our results highlight the critical role of host contact structure in CIV dynamics , and show how host contact networks could shape the evolution of pathogen transmission efficiency . Importantly , efficient control measures could eradicate the virus , in turn minimizing the risk of future sustained transmission among companion dogs that could represent a potential new axis to the human-animal interface for influenza .
Respiratory pathogens that emerge as the result of host-range shifts can cause serious epidemics in humans , livestock , and wild animals [1]–[4] . Two recent pandemics in humans – Severe Acute Respiratory Syndrome ( SARS ) in 2003 and H1N1 influenza in 2009 – involved host-range shifts in respiratory zoonotic viruses [5] , [6] , while the recently documented Middle East respiratory syndrome coronavirus ( MERS-CoV ) has similarly emerged from an animal reservoir to pose a growing risk to the human population [7] . Importantly , however , cross-species transmission events do not always lead to pandemics . Rather , zoonoses emerging in new host species tend to have patchy and dynamic prevalence patterns in space and time . As a result , the probability that an emerging zoonosis will take hold in a new host population has been difficult to assess a priori , which limits our capacity to use targeted interventions to avert pandemics before they happen [8] . A variety of host-pathogen interactions may follow a species jump , and revealing their determinants is essential to understanding the process of zoonotic emergence . First , the emerging pathogen may be poorly adapted for replication and onward transmission in the new host population . This leads to inefficient transmission , where many potentially infectious contacts between susceptible and infected individuals fail to spread the disease , due for example to a low pathogen load in the infected individual . In this case , the disease will have a lower basic reproductive number ( R0 – the number of secondary infections caused by a typical infected individual in an entirely susceptible population ) in the recipient host than its recent ancestor in the donor host . Inefficient transmission following a spillover event may lead to “stuttering chains” of infection marked by patchy patterns of disease prevalence interspersed with stochastic fadeouts . Even if a pathogen has R0 above 1 ( a necessary but not sufficient condition for self-sustaining spread ) , values of R0 that are only marginally above 1 are associated with a higher probability of stochastic extinction . The probability that a pathogen will establish itself ( in a large homogeneously mixed population of susceptible hosts ) following the introduction of n infected individuals is given by 1− ( 1/R0 ) n [9] . The heterogeneity in prevalence of emerging pathogens may also reflect the demographic variability inherent to host populations . In smaller host populations random variation in the timing and frequency of births , deaths , immigration , emigration , and contacts between infected and non-infected individuals , as well as in the timing of individual infections , can have profound effects on epidemic dynamics [10]–[12] . Emerging pathogens that result from spillover into new hosts are by definition initially confined to a small population , in the sense that the first infected individual ( s ) will have limited numbers of potential contacts to whom they can spread the disease . This , in turn , makes the epidemic dynamics of emerging pathogens inherently stochastic [13] , [14] . Finally , evolutionary change in emerging pathogens can affect both their basic reproductive number , and their response to demographic variability . Pathogen evolution can result in R0 increasing toward or above 1 . 0 after repeated spillover events from the reservoir population , or during a chain of transmission in the new host , either of which could result in the emergence and selection of host-adaptive mutations . The occurrence of multiple outbreaks over time may also increase the likelihood that the pathogen evolves toward a point when it can be self-sustaining in the new host [12] . Recent analytical frameworks that unite the ecological and evolutionary dynamics of host-pathogen interactions can help identify the processes that drive epidemiological and phylogenetic patterns during and after host-range shifts [11] , [12] , [15] . Here we study the population dynamics and evolution of equine-H3N8 derived canine influenza virus ( CIV ) in the US , and use the results to propose control strategies . CIV emerged following the transfer of a single H3N8 equine influenza ( EIV ) to dogs from horses around 1999 . Direct descendants of that virus have been circulating continuously in dogs since that time [16]–[18] . CIV was first recognized as the cause of disease in greyhounds in a training facility in Florida in 2004 and was transferred to various states in the US with the racing greyhounds , eventually spreading to other breeds [16] . The hemagglutinin ( HA ) sequence of CIV was genetically distinct from EIV by 2004 , forming a separate monophyletic group from EIV in phylogenetic trees [16] . Notably , there is no evidence of CIV transfer back to horses , onward to humans , nor of reassortment involving CIV and other influenza viruses [19] . Furthermore , although some other H3N8 EIV spillovers from horses into dogs have been reported , these only comprised single infections or small outbreaks that rapidly faded out [20] . Although CIV can readily transmit among dogs its prevalence remains patchy: it is enzootic in some regions of the US , but has thus far failed to establish outside of these enzootic regions [21]–[24] . The overall seroprevalence of CIV in the companion ( pet ) dog population appears to be low ( ∼3% or less depending on the region ) , with visits to canine daycare a risk factor [22] , [24] . CIV enzootic regions are typically associated with large animal shelters [25] , and the movement of the virus to different parts of the US is most likely associated with the transport of infected shelter dogs to facilities in other regions where they may be more readily adopted [26] . In contrast to CIV , its recent ancestor , H3N8 EIV , has been circulating widely in horses since before 1963 when it was first reported in Florida , having most likely been introduced with horses from South America [27] . The virus appears to spread continuously within and between many parts of North and South America , Europe , and Asia [28]–[31] . EIV has been introduced into countries that were previously free of the virus , including Australia and South Africa , causing significant outbreaks that extended over large distances , although these were controlled and the virus eradicated [28] , [32] . Data from an outbreak in an unvaccinated population of racehorses places R0 for EIV at 10 . 18 ( 95% confidence interval: 9 . 57–10 . 89 ) in that context . In contrast , the reproductive number of EIV in vaccinated populations of racehorses has been estimated to be between 1 . 4 and 2 . 3 [33] . EIV has experienced marked evolution in all gene segments since it emerged , with evidence of antigenic variation in the HA gene , including phylogeographic patterns in HA variation , with distinct clades in Europe versus the US , and among US states [28] , [34] , [35] . Although CIV and EIV are closely related , their epidemiology and evolutionary dynamics differ , with EIV seemingly more successful , and less heterogeneously distributed . Moreover , EIV continues to spread despite active control measures ( particularly vaccination ) whereas CIV retains a patchy distribution in the absence of significant control measures . An analysis of the phylogenetic history and ecology of CIV since its recent emergence from EIV may therefore reveal how host demography , disease dynamics , and pathogen evolution can combine to determine the prevalence patterns and risk posed by emerging zoonotic pathogens . Here we combine individual-level data on the intake , output , and transfer rates of dogs among US animal shelters of different sizes , with CIV gene sequence data and available seroprevalence estimates , to identify the processes controlling disease dynamics in emerging zoonoses at the human-animal interface . We hypothesize that CIV persists through the presence of transmission hotspots , which rescue chains of transmission that fade out in other populations . The putative hotspots are large animal shelters in major metropolitan areas . After estimating R0 from all available data we ask: are the population sizes of small shelters small enough to make fadeout significantly more likely than in large shelters ? And do large shelters have good prospects of maintaining CIV in an enzootic state ? We then use the parameters from our analyses to determine what control measures would result in eradication of the virus . CIV is a prime target for eradication because it has both the potential to cause significant disease burden , and it is currently confined to a small subset of its host population . Effectively controlling CIV would improve conditions in metropolitan animal shelters , as well as minimizing the risk of zoonotic human infection posed by CIV , before it has the opportunity to evolve higher transmissibility in the companion dog population . A possible future scenario of sustained CIV transmission amongst companion animals would represent the evolution of a potentially significant new axis to the human-animal interface for influenza .
To put the CIV sampled from animal shelters in a wider geographical context , and to reveal movement of the virus on a continental scale , we determined the HA1 , M and NP gene sequences of recent CIV isolates , and conducted a phylogenetic analysis of these sequences combined with homologous sequences available on GenBank as well as the Influenza Research Database . Viruses or sequences were sampled from the US states of Colorado , New York , Pennsylvania , Florida , California , Kentucky , Wyoming , Philadelphia , South Carolina , Virginia , Vermont , Connecticut , Texas , and Iowa . The most striking result of this analysis is that CIV exhibits a marked geographical clustering by US state with distinct clades being observed in New York ( and nearby states ) , in Pennsylvania , and in Colorado , which also represent our largest sampling sets ( Figure 1 ) . This geographical clustering was confirmed in Association Index ( AI ) and Parsimony Score ( PS ) phylogeny-trait association statistics [36] , with significantly more clustering by US state of origin than expected by chance alone across the data set as a whole ( p<0 . 001 ) . Similarly , the Maximum Clade ( MC ) statistic reveals significant ( p<0 . 001 ) clustering in the individual states of Colorado , New York , Pennsylvania , Vermont and Wyoming . In addition , many of the viruses from the northeastern states of Vermont , Connecticut , New Hampshire clustered with the viruses that are circulating continuously in New York ( also in the northeast ) , suggesting that those viruses were derived from the New York enzootic hotspots . Next , we investigated the demographic and epidemiological dynamics of CIV at the local scale in animal shelters . Here we used individual-level records of dog arrival and departure from 13 animal shelters of varying size across the US , comprising a total of 124 , 519 dogs , as well as published seroprevalence estimates from a large shelter [25] , coupled with a stochastic epidemic model . The epidemic model was an SIR-type model incorporating empirical rates of arrival and departure from the shelter as well as CIV infection and removal dynamics , and implemented at the level of individual dogs using the Gillespie algorithm [37] . The majority of animal shelters in the US house relatively small populations of dogs—the median dog population size in our sample of shelters is 43—but a few shelters are much larger , housing hundreds of dogs . In precise terms , the distribution of dog population sizes in our data is close to a negative binomial distribution with mean 71 . 23 and standard deviation 82 . 24 ( Figure 2A ) , which indicates significant overdispersion in population sizes relative to a homogeneous Poisson model . This overdispersion in host population size is a potentially important characteristic for the epidemiology of CIV because it indicates the presence of a few extraordinarily large shelters where a pathogen might persist more easily than in a host population of average size . Large well-connected populations are more favorable environments for emerging pathogens because variance in vital rates ( e . g . the rate of arrival of new susceptible individuals ) decreases predictably with population size . All else being equal , this makes large host populations more stable for sustained pathogen transmission . Shelters with larger populations are fueled primarily by higher intake rates ( Figure 2B ) , as the median residence time of dogs does not vary significantly among shelters of different sizes ( Figure 2C ) . The residence time of dogs in a shelter is roughly exponentially distributed with a mean of 9 . 88 days and a standard deviation of 8 . 22 days ( Figure 2D ) . Transfer rates among shelters appear relatively low—among the eight shelters in our demographic data for which there was transfer information the median proportion of dogs whose stay at a shelter ended with a transfer is 0 . 067 and the mean is 0 . 1 . Transfer probability is not correlated with dog population size in our data ( data not shown ) . Most dogs arriving to shelters are susceptible to CIV [22] , [25] . The arrival rate of susceptible dogs places an upper limit on CIV prevalence by continual dilution with uninfected individuals , which leads to a saturating relationship for prevalence as a function of R0 ( Figure 3A ) . We estimated a posterior distribution for R0 given seroprevalence data and demographic data by using a Markov Chain Monte Carlo ( MCMC ) method based on a stochastic SIR model parameterized with the demographic data ( see Methods ) . Point estimates of seroprevalence are normally-distributed about the long-term equilibrium value given by the mean-field model in our simulations ( Figure 3B ) , and a seroprevalence estimate of 0 . 42 [25] from a large shelter where CIV is enzootic , combined with the demographic data on dog intake and outcome rates , yield a mean estimate for R0 of 3 . 9 for CIV in large animal shelters . The posterior distribution of R0 has a median of 3 . 3 , and a highest probability density ( HPD - the central 95% of the posterior distribution ) interval extending from 2 . 0 to 8 . 9 ( Figure 3C ) . Moving from estimating the reproductive potential ( R0 ) of CIV in animal shelters to estimating the effective spread rate in the general population from genetic data , we employed phylodynamic birth-death models to analyze HA1 sequence data collected across the US to determine the rate of spread of the virus in the wider dog population . Our estimates for the effective reproductive number ( Re; the average number of secondary infections produced by a typical infected individual at a given time , in a population where not all potential hosts are necessarily susceptible ) show considerable temporal variation ( Figure 3D , E ) . At the time when CIV was first recognized in 2004 the posterior distribution of Re roughly matches that of R0 , consistent with the initial , exponential phase of the epidemic . During the period 2004–2008 Re drops to a value of approximately 1 . 0 . A similar pattern was observed in the New York data set . Across the USA as a whole the mean estimate of Re is currently 1 . 02 ( 95% HPD = 0 . 79 , 1 . 26 ) , with a similar number found in New York ( Re = 1 . 06 , 95% HPD = 0 . 72 , 1 . 47 ) ( Figure 3 ) . The low Re observed toward the present suggests that CIV spread has now reached an equilibrium , where stochastic fadeouts often associated with outbreaks are balanced with new infections , which usually occur in the large animal shelters where it is enzootic . Using the shelter demography data and the stochastic epidemic model , we simulated CIV outbreaks in shelters of a realistic distribution of sizes , intake rates and output rates , and for varying levels of R0 . From these simulations we estimated the probability that a shelter ( of a given population size , N ) infected with a CIV virus ( of a given R0 ) could maintain the virus for 100 days . The response surface for this experiment yielded a cut-off curve in the N-R0 plane , below which fadeout was almost certain and above which persistence was almost certain ( Figure 4A , shaded surface ) . Interestingly , the posterior distribution for R0 ( estimated by MCMC as described above ) , and the empirical distribution of shelter population sizes from the demographic data , suggest that CIV straddles the border between persistence and stochastic fadeout . That is , the demographic and seroprevalence data indicate that CIV cannot persist in the majority of shelters , because the median of the posterior distribution of R0 is below the cutoff for persistence in a shelter of median size . But CIV can persist in existing shelters that are larger than the median size , because the posterior distribution for R0 , combined with the distribution of shelter sizes from the demographic data , has significant density at R0-N combinations that would allow persistence ( Figure 4A , points ) . This suggests that CIV persists on the brink of extinction—its current transmission efficiency is only sufficient to persist in large , high-throughput populations , but not yet to invade more widely . More generally , for CIV in animal shelters the stochastic epidemic simulations parameterized with demographic data reveal that the impact of demographic stochasticity is considerable; the majority of shelters are too small to maintain the virus in the long term at its present rate of transmission . Thus variation in contact rates among host subpopulations , rather than inherent limitations on the evolved transmission efficiency of the pathogen , is sufficient to explain observed heterogeneity in prevalence in CIV , since the virus has been shown to transmit efficiently in large shelters and under laboratory settings [23] . We then used the parameterized epidemic model and demographic data to simulate the impact of control strategies for CIV . The purpose of this analysis is to test how the observed hotspot dynamics are predicted to interact with potential eradication strategies , and to estimate , given available demographic and epidemiological data , the degree of control efficacy required to eliminate CIV from animal shelters where it is persisting , allowing for the complete eradication the virus from the canine population . Figure 4B shows the effect of a generic control measure applied to a single shelter , that reduces the inflow of susceptible dogs to a proportion 1/<R0> = 0 . 26 of current levels , where <R0> = 3 . 9 is the mean of the posterior distribution for R0 , as above . This generic control measure has the effect of reducing the effective reproductive number in a single shelter to unity . Consistent with theory [38] this is a reduction in the susceptible portion sufficient to eradicate the disease from a large isolated shelter . We then simulated a control program across multiple shelters connected by the transfer of dogs . The simulated control measure applied to the metapopulation is represented in the model as a vaccination program , but the results apply to any control that achieves the same reduction in the force of infection . A possible strategy ( but to our knowledge untested; see below ) that might be used is a live attenuated influenza vaccine ( LAIV ) that is able to generate an interfering response ( possibly through generation of interferon ) which prevents infection by wildtype CIV . We find that such a control program could eradicate CIV within 1–2 months if it is applied to dogs immediately upon arrival to the shelter , and removes them from the chain of transmission within 24 hours with 85% probability ( Figure 5A ) . A control strategy that has an efficacy of 75% might also efficiently eradicate CIV from isolated shelters , but transfers of dogs between shelters at the observed mean rate will allow CIV to persist through connected chains of outbreaks ( Figure 5B ) . Control strategies with efficacies of 65% or less would reduce the prevalence of the infection , but are not predicted to lead to CIV eradication in every case ( Figure 5C ) . In addition , without appealing to models it is clear from the demographic data that turnover rates in most shelters are too high for an inactivated vaccine to be effective , because those vaccines take more than a week to generate protective immunity . Since the expected residence time of a dog in an animal shelter is around 10 days , most dogs would have been part of the chain of transmission by the time an inactivated vaccine given at intake took effect , and then they would leave the shelter . We therefore suggest one possible approach for control is the development of an LAIV administered at intake of dogs into shelters , which may provide suppression of the wild type virus infection by direct competition for the target tissues , as well as through stimulation of innate immune responses . This potential approach requires further research . In addition we note that the LAIV performance required to eradicate CIV is higher than what has been typically observed in other live influenza vaccines , and there is still much uncertainty about the rate at which protection would be acquired during the time that dogs spend in shelters ( typically 1–2 weeks ) [39] . However , other control measures ( e . g . quarantine , decrease in population size , changes in population structure , or anti-viral drugs ) or combinations that removed dogs from the chain of transmission with similar efficiency would also have qualitatively equivalent effects in a control strategy . The key result is that a net control efficacy of ∼85% is required for eradication of CIV from its reservoir in a network larger animal shelters , even though a net efficacy of ∼75% would probably be sufficient for eradication from a single shelter . Thus another interesting feature of the control simulations in a metapopulation framework is the sensitivity to migration parameters ( Figure 5 ) . In particular , we show that there are efficacious control strategies that would cause the virus to go locally extinct but that would fail to achieve global eradication . These include scenarios where the expected global prevalence approaches 0 , but where the virus would persists through connected chains of stochastic outbreaks ( Figure 5B ) . We also used our epidemic model to explore the passage of CIV from an infection in one large shelter to other shelters through the transfer of dogs ( Figure 6A ) . The hotspot dynamics predicted by our model show regularities in the way CIV spreads outward from a single shelter . Large shelters are predicted to receive the infection earlier , as well as maintaining it for longer , creating a wave in the population size—time-of infection plane ( Figure 6B ) . The probability that a single infection introduced to a susceptible shelter would start an epidemic that persisted for at least 100 days increases with population size ( Figure 6C ) . For the median population size of 43 dogs the probability was approximately 0 . 5 .
Since its emergence more than a decade ago , equine H3N8-derived CIV has maintained a patchy distribution , occurring most often in sporadic and short-lived outbreaks in US animal shelters [21] . In contrast , strains of CIV's recent ancestor ( EIV H3N8 ) have been commonly found in horses around much of the world since at least 1963 . EIV thus transmits efficiently among horses , sometimes despite vaccination programs [33] , [40] . Our study investigates heterogeneity in CIV transmission and prevalence to better understand the processes that determine how zoonotic pathogens spread following a host range shift , and how these processes affect the performance of control and eradication strategies . Stuttering chains of infection in recently emerged zoonoses are caused by two distinct mechanisms that may operate in tandem . The first is poor adaptation of the pathogen to its new host species , causing lower within-host replication rates that may reduce pathogen shedding and lead to inefficient transmission [11] . Transmission is inefficient because many potentially infectious contacts between infected and susceptible individuals fail to spread the infection . The second mechanism that generates stuttering chains is variation in contact rates among different subsets of the new host population , which can increase the probability of stochastic fadeouts [13] . In this case the fadeouts are not caused by inefficient transmission but by exhausting the local supply of susceptible hosts . Strikingly , host contact heterogeneity has received less attention as a potential driver of stuttering chains despite its fundamental role in disease dynamics . We found that contact heterogeneity plays a critical role in the patchy distribution of CIV . Demographic data indicate that most shelters are too small , and import susceptible individuals too slowly , to protect CIV from local stochastic extinction at its current reproductive rate . At the same time , the influence of contact heterogeneity on the spread of CIV would diminish if transmissibility were higher and the epidemiology of CIV ( and similar zoonoses ) generally represents an interaction between contact heterogeneity and transmission efficiency . More evidence of contact heterogeneity appears at the inter-shelter scale . Observed transfer rates suggest that the majority of intakes and outputs are not associated with other shelters , so that shelter-to-shelter transfer has not created an effectively larger metapopulation of multiple shelters . While there are many millions of susceptible household dogs in the USA ( around 80 million , about 25% the size of the human population ) , it is likely that they do not exhibit infectious contact patterns sufficient to maintain the virus in continuous transmission ( see below ) . Hence , this work shows that apparent stuttering chains of transmission are not always driven by poor adaptation of an emerging pathogen to its new host . Rather , in the case of CIV , demographic variability in contact rates is alone sufficient to explain the fadeout in the disease in most situations . The basic reproductive number for CIV is intriguingly close to the minimum value required to persist in a shelter of average size ( Figure 4A ) . This suggests that transmission efficiency in CIV may have evolved to precisely the point of persistence . Yet , this average shelter size belies the high variance in shelter populations , with many small facilities balanced by a few extraordinarily large ones ( roughly matching the distribution pattern of US city sizes; Figure 2A ) . The process by which CIV currently persists is therefore to thrive in a few large populations with high rates of infected-susceptible contact ( as mediated by a high arrival rate of new susceptibles , rather than by increase density ) , but failing to take hold in the general population ( Figure 4A ) . By analogy with conservation biology , large populations thus function as refugia for the virus , protecting it from extinction and also limiting its distribution . These are also exactly the conditions to facilitate further evolution toward higher reproductive capacities , by facilitating repeated outbreaks outside the refugia , followed by selection for higher transmissibility [10] , [12] , [41] . Our mean estimate of R0 = 3 . 9 in the large animal shelters is lower than that estimated for EIV during outbreaks [42] , but close to the upper bound for estimates of human influenza transmission [43] , [44] . It is also considerably higher than that of pandemic H1N1 influenza in humans in 2009 ( R0 = 1 . 4–1 . 6 ) which spread worldwide within weeks of its first recognition in humans [45] . Variation in R0 among different viral strains and host species can be difficult to interpret because of the many factors that can affect transmission and removal rates in different settings . However , these comparisons do indicate that CIV has the biological capacity to spread relatively efficiently among dogs given the right conditions in the host population . Although most of the parameters in the epidemic model were estimated from a large volume of host demographic data from animal shelters , there are currently few estimates of seroprevalence in shelters where CIV is endemic and as a result our estimate for equilibrium seroprevalence relies on data from a single shelter [25] . This introduces a risk of sampling bias because that individual shelter could exhibit individual characteristics that affect its equilibrium seroprevalence , and/or due to non-random temporal fluctuations in seroprevalence . While the scarcity of seroprevalence estimates adds uncertainty to R0 estimates , the extant data would be difficult to explain with values of R0 lower than our estimates . This is due to the rapid rate at which infected individuals are replaced by new arriving susceptibles in the high throughput shelters where CIV is enzootic . The low residence time of dogs in large , high-throughput shelters thus indicates ( consistent with previous results [25] ) that individuals in shelters where CIV is enzootic must acquire the infection within a few days of arriving . This places a lower bound on probable values for R0 by constraining estimates of the generation time of the infection , at least in the context of large shelters [44] . In other facilities that are smaller or where CIV is not enzootic , long-term average seroprevalence over time may be lower [26] due to stochastic fadeouts of the disease . Phylogenetic analyses independently supports several key predictions of our analysis . First , these analyses confirm that CIV remains confined to endemic hotspots , with transfers to other regions causing outbreaks that are generally short-lived ( and thus failing to establish new lineages outside of endemic locations ) . Moreover , the strong phylogeographic structure , with distinct viral clusters in New York , Pennsylvania and Colorado for each gene analyzed , is exactly what might be expected given our empirically-parameterized epidemic model , which predicts geographic segregation . Each US state has only a few large cities that would have an animal shelter capable of supporting CIV in the long term , and with relatively infrequent transfers among cities . Second , the mean and HPD interval for Re from the phylogenetic analysis in 2004 ( when CIV was first detected and when infected greyhounds were being transported to many US states for racing ) roughly matches the distribution of R0 from the stochastic SIR model and demographic data . The initial phase of an epidemic is usually associated with exponential growth ( corresponding to Re = R0 ) , and Re must always be less than or equal to R0 by definition . This makes the phylogenetic estimate of Re in 2004 a conservative independent assessment of R0 . The third concurrence between the demographic and phylogenetic analysis involves the current estimate of Re∼ = 1 . While R0 in our analysis measures the reproductive potential of the disease where it is enzootic , the phylodynamic estimates of Re reflects the net spread rate of CIV across the US as a whole , including multiple shelters and the companion dog population . As such , Re∼ = 1 indicates that on balance CIV is currently failing to persist where it is not already enzootic , which is consistent with both epidemiological observations [21] and the predictions of the demographic analysis which showed that most shelters are too small to support CIV in an endemic state ( see Figure 4 ) . Targeted strategies for control and eradication depend on understanding the conditions under which an emerging enzootic pathogen can maintain itself in a new host population . Our simulations indicate that eradication may be possible , but will require relatively efficient control , and that success may depend on the rates at which dogs are transferred between animal shelters . For control measures with ∼75% efficiency , CIV may be eradicated from most shelters but still persist overall through connected chains of outbreaks , with large shelters serving as focal points for staging new infections elsewhere . This suggests that control programs for CIV will be most successful if implemented across multiple shelters and that participating shelters should maintain control measures even after the virus has been eliminated locally . Uncertainties in our analysis would be reduced by more information on CIV prevalence and additional CIV sequence data . A key area of uncertainty is transmission rates between companion dogs within and among households whose contact patterns would differ significantly from those of dogs in animal shelters . Working from first principles , if a proportion p of contacts between infectious and susceptible dogs result in the infection being transmitted , then the probability that an infected individual in the companion population will first transmit the infection on their kth contact with a susceptible dog is p ( 1-p ) k−1 , which gives an expected value for k of 1/p . Studies of CIV transmission in comingling trials estimate p = 0 . 75 [23] . This suggests that for a CIV lineage to avoid extinction in companion dogs , the average infected dog must contact k = 1 . 33 susceptible individuals during the time they are infected . Another approach that yields the same result is to recognize that if a proportion p of contacts produce secondary infections , then R0>1 requires k>1/p , again translating to approximately two contacts per week . It is evident from common experience that while some companion dogs are highly sociable , others do not frequently interact with other dogs . Therefore , some companion dogs probably achieve the minimum contact rate required to sustain CIV , but others probably do not . Many dogs with higher contact rates than this minimum would be necessary for CIV to actively spread among companion dogs , and to protect the virus from stochastic extinction in the general dog population , at its current transmission efficiency . Conversely , if CIV had emerged with much higher transmissibility upon entry to the greyhound population , stuttering chains of transmission would not have been observed . Thus , the capacity for contact heterogeneity to control epidemic spread is facilitated by lower transmissibility . Furthermore , while endemic hotspots created by contact heterogeneity can theoretically increase the chances of evolving higher transmission efficiency , it is evident that not all cross-species transmission events will involve small stepwise gains in transmissibility in the new host . Indeed , a variety of adaptive models , involving differing numbers and fitness of mutations , can be put forward to explain the process of emergence [46] . Despite these caveats , and limitations in the available data , our analyses provide a coherent view of the ecological and evolutionary dynamics of CIV . After approximately 15 years of continuous circulation among dogs in the US , CIV can be maintained only in relatively dense host populations with high inputs of susceptible individuals ( essentially viral “chemostats” ) , despite a relatively high reproductive potential in that context . These hotspots are weakly connected by migration , leading to geographic signatures in the CIV phylogenies . Most dog populations are too small or diffuse to independently support CIV at its current level of transmissibility , explaining its current modest reproductive rate ( Re∼ = 1 ) , and consistent with the epidemiology of CIV , which is characterized primarily by sporadic short-lived outbreaks outside of enzootic centers [21] . The demographic gradient between high-throughput populations where CIV is enzootic , and smaller or more diffuse populations where sporadic outbreaks can occur , creates hotspot dynamics that can facilitate pathogen evolution toward higher transmissibility [12] , [41] . Our results therefore demonstrate one way that urbanization can increase the risk posed by emerging infectious pathogens [2] , [47] . Although humans exposed to CIV appear not to be commonly infected ( as shown by serological testing ) , the true risk of future human infection by either EIV or CIV is unknown as we do not understand the host barriers that restrict human infection , or the genotypic changes in the viruses that might overcome those barriers [19] . Our analysis can therefor inform a strategy for preemptive eradication of an influenza A virus that is well adapted to mammals , since if CIV did gain high transmissibility among companion dogs then much of the human population would be directly exposed to the virus .
Our analysis is based on an SIR framework that models changes over time in the number of dogs in a shelter who are susceptible ( S ) , infected ( I ) , or removed ( recovered and thus immune; R ) . In what follows , we describe the model for a single shelter: below we expand the model to incorporate the dynamics of an control program that reduces the force of infection , and to consider control in multiple shelters linked through the transfer of dogs . We assume dogs arrive at a shelter of a given size at a rate of μ dogs per day . Dogs leave at a per-capita rate of δ per dog per day , regardless of their state , so the mean residence time in a shelter is 1/δ days . The number of dogs in a shelter , N = S+I+R , is equal to μ/δ at equilibrium . Arrival and departure rates are estimated empirically using individual-level records from 13 animal shelters of varying size across the US ( see Supporting Information , Data S1 ) . The records comprise a total of 124 , 519 dogs , recording the date each individual arrived and left the shelter . In 8 of the 13 shelters , the data included whether or not the departure of the dog represented a transfer to another shelter . Arrival rate , μ , for a shelter was estimated as the median number of dogs arriving in that shelter per day . Departure rate , δ , for a shelter was estimated as the inverse of the median length of stay of dogs in that shelter . When estimating arrival and departure rates we excluded dogs that were admitted to the shelter in response to a euthanasia request , as these dogs had systematically shorter residence times . We also excluded dogs whose length of stay was greater than 40 days , as these represented rare atypical cases ( see Figure 2C , D ) . We assume that dogs in a shelter have a constant rate of contact per day with other dogs where the contact would be capable of spreading infection if one of the dogs were infected . An alternate hypothesis is that contact rate increases with population size , potentially leading to hotspot dynamics in large shelters in the absence of demographic stochasticity . Our assumption of constant contact rates is thus conservative with respect to the hypothesis that demographic structure drives hotspot dynamics in CIV . Assuming that contact between any pair of dogs in the shelter is equally likely , the force of infection is given by λ = βP , where β is the contact rate and P = I/N is the current prevalence of CIV in the shelter [48] . The rate of new infections is given by λS , and susceptible dogs contract the disease an average of 1/λ days after entering the shelter . In this framework , the basic reproductive number of the disease is R0 = β/ ( γ+δ ) , and the disease only persists in the long run if R0>1 , in which case equilibrium prevalence is given by ( 1 ) which is bounded above by δ/ ( γ+δ ) as R0 becomes large ( see Supporting Information , Text S1 , Section 1 . 6 ) . The infected class in our model represents the number of dogs with non-zero viral loads , rather than those exhibiting clinical symptoms . Thus , we avoid including latent or asymptomatic classes in our model . We set γ = 1/7 because viral shedding continues for approximately seven days after inoculation [16] . Seroconversion for dogs infected with CIV also happens at approximately 7 days [16] . Equilibrium seroprevalence is then given by R/N ( see Supporting Information , Text S1 , Section 1 . 7 ) . Variation among individuals in time of infection , recovery , arrival , and departure causes variations in disease prevalence around the predicted long-term average . These excursions from mean prevalence carry with them the risk of visiting zero prevalence , leading to stochastic extinction of the disease . This demographic stochasticity becomes increasingly pronounced in smaller populations . However , the critical population size below which disease dynamics begin to significantly diverge from the long-term average through stochastic fadeouts depends upon R0 , and upon the turnover rate in the population . We parameterize the stochastic SIR model with the demographic data to test the impact of demographic stochasticity on the spread and maintenance of CIV in animal shelters . We implement the model in continuous time at the level of individual dogs using the Gillespie algorithm [37] . We estimated a posterior distribution for R0 given seroprevalence data and demographic data by using a Markov Chain Monte Carlo ( MCMC ) method , as follows . From the stochastic SIR model we simulated seroprevalence samples by observing the seropositivity of n randomly selected dogs from the population at a given time . Seroprevalence thus observed has the property of being normally-distributed about the long-term equilibrium value given by the mean-field model in our simulations ( Figure 3B ) . We then seek the posterior distribution of an unknown equilibrium seroprevalence at an actual shelter , given a real point seroprevalence estimate there . We estimate this distribution by sampling from the Gaussian distribution of deviations between point seroprevalence estimates and equilibrium seropreovalence , using the Metropolis-Hastings algorithm [49] . We assessed convergence by visually examining within- and among- chain mixing . Convergence was determined to have occurred when the long-term variance in sampler state among chains was the same as the variance within chains . Convergence under this definition was easily achieved using 10 chains run for 105 steps each , with a burnin of 10% , and keeping every 100th step . From the posterior distribution of equilibrium seroprevalence , we then map to a posterior distribution for R0 by inverting Equation 1 . We represent control strategies by a vaccination program that reduces the force of infection ( as described below ) , but the results apply to any control measure that provides a similar reduction in risk of infection and in infectiousness . In what follows we discuss control in the context of a LAIV administered to dogs upon arrival at the shelter . The model with vaccination dynamics includes two more compartments , counting the number of dogs in each shelter who are vaccinated ( V ) , and the number of dogs who are infected despite vaccination ( W ) . Vaccination reduces a dog's susceptibility to infection by decreasing the probability that a virus population initially transferred through infectious contact will enter a phase of exponential growth , prerequisite to significant viral shedding and clinical symptoms [50] . By reducing viral load and viral shedding , vaccination reduces the risk of infection in vaccinated dogs and reduces the infectiousness of a dog who becomes infected despite vaccination . Vaccinated dogs thus experience a reduced force of infection ελ , 0≤ε≤1 , and , if they become infected , contribute to the force of infection at a reduced rate 0≤ω≤1 , leading to an overall force of infection of λ = β ( I+ωW ) /N in population which has W vaccinated individuals who have nonetheless become infected . Dogs transition from S to V at a rate of α per dog per day . Because a live vaccine is assumed to be administered to dogs immediately upon arrival , 1/α measures the average time after entry/vaccination that a dog experiences the vaccine-associated decrease in risk of infection from other dogs , and decreased infectiousness if they do become infected . Vaccination changes mean dynamics by reducing R0 by a factor of 1-K , where K is effective vaccination coverage . K is given by ( 1-κ ) V/N , where κ = εω expresses the failure rate of the vaccine , ranging from 0 for perfect vaccine , to 1 for an entirely ineffective one ( see Supplementary material , section 1 . 4 ) . We use a step function for κ as a function of α , where κ goes from 1 to its post-vaccine value at 1/α days . We also model the effects of a generic control strategy equivalent to inoculating some dogs with a perfect vaccine , or to quarantine that partially or completely stops the flow of susceptible dogs into the shelter . We do this by replacing susceptible dogs with removed ones in the intake stream . Reducing the proportion of susceptible dogs in the intake stream to 0<θ<1 , while 1- θ are already removed , has the same effect as reducing R0 to θR0 . The metapopulation model expands the stochastic SIR model for a single shelter to describe multiple shelters whose dynamics are linked by the transfer of dogs . As above , the model is implemented at the level individual dogs using the Gillespie algorithm . Thus at each point in continuous time , each individual in the model has a disease state ( S , I , R , V , or W ) and a location in a given shelter . The metapopulation is composed of shelters that vary in dog population size , intake rate and output rate by sampling with replacement from the demographic data . Transfer probabilities are also based on the demographic data ( see Supporting Information , Text S1 , Section 2 ) . Although the CIV phylogenies show geographic localization ( see Figure 1 ) , the metapopulation model is spatially implicit , consistent with level of detail in the demographic data we used . However , even without including spatial structure in transfer patterns , the metapopulation model reproduces hotspot dynamics , based on transfer hierarchies driven by differences in shelter size ( see Figure 6 ) . We compiled all available CIV HA1 , NP and M gene segment sequences from GenBank and the Influenza Research Database ( www . fludb . org ) and by sequencing samples provided by the Animal Health Diagnostic Center ( AHDC ) at Cornell University . For the sequencing of the virus samples obtained from AHDC we extracted viral RNA using Qiagen viral RNA mini kit and synthesized cDNA using Avian Myeloblastosis Virus ( AMV ) reverse transcriptase and influenza universal primer Uni12 . Three gene segments , HA1 , NP and M , were then amplified by PCR with gene specific primers ( primer sequences are available upon request ) for all samples . The PCR products were purified using EZNA Cycle-Pure Kit and sequenced by the Sanger method . All sequences derived here have been submitted to GenBank and assigned accession numbers KM359803-KM359864 . All sequences were aligned by MUSCLE v3 . 8 . 31 [51] using default parameters , followed by manual adjustment . Phylogenetic trees of each gene were then estimated using the maximum likelihood ( ML ) available in PhyML 3 . 0 [52] and assuming the general time-reversible reversible ( GTR ) model of nucleotide substitution and a gamma distribution of among-site rate variation with 4 rate categories ( i . e . the GTR+I+Γ4 model of nucleotide substitution ) with SPR branch-swapping . The robustness of the phylogeny was estimated using 1 , 000 bootstrap replicates . Because of their greater availability , the analyses of evolutionary dynamics and phylogeography were only performed on the HA1 gene ( see below ) . To estimate R from the CIV sequence data we used a total of 94 HA1 sequences ( alignment length = 1032 nt ) sampled from various locations ( states ) in the US ( Colorado , New York , Pennsylvania , Florida , California , Kentucky , Wyoming , Philadelphia , South Carolina , Virginia , Vermont , Connecticut , Texas , and Iowa ) between 2003 and 2013 . This data set included 40 sequences sampled from dog shelters in New York between 2005 and 2012 , which were analyzed separately using the same protocols . First , we estimated the mean ( and credible intervals ) of R in both data sets using the epidemiological birth-death method [53] available in BEAST v1 . 7 . 5 [54] . This analysis used the simpler Hasegawa-Kishino-Yano ( HKY ) model of nucleotide substitution and a gamma distribution of among-site rate variation ( HKY+Γ4 ) . To account for temporal rate variation in the data an uncorrelated lognormal relaxed molecular clock model was employed . Using the Bayesian Markov Chain Monte Carlo ( MCMC ) framework available in BEAST , 100 million steps were run , sampling every 10 , 000 and removing 10% as a burn-in . Second , temporal changes in R were estimated using the more complex serial-sampled birth-death ( SSBD ) model [15] , available in BEAST v2 . 0 [55] , again using the HKY+Γ4 but this time ( to ensure statistical convergence ) employing a strict molecular clock with a uniform distributed clock rate of 2×10−3 ( 1×10−3–3×10−3 ) nucleotide substitutions per site , as this was found to be best-fit to the data in epidemiological birth-death method . The MCMC was again run for 100 million steps , sampling in the same way as described above . Two independent runs allowed different Re values to be inferred from up to Re = 25 . To determine whether CIV was more clustered on the phylogenetic tree by US state of sampling than expected by chance alone , we employed the Association Index ( AI ) , Parsimony Score ( PS ) and Maximum Clade size ( MC ) phylogeny-trait association statistics incorporated within the Bayesian Tip-association Significance testing ( BaTS ) program [36] . Traits were defined as the US state of sampling . Phylogenetic uncertainty in the data was incorporated by basing estimates on the posterior distribution of trees obtained from the BEAST analysis ( epidemiological birth-death method ) described above . In all cases , 1000 random permutations of sampling locations were undertaken to create a null distribution for each statistic . | Influenza virus infects a range of vertebrate hosts , including domesticated animals as well as humans . Some of the most serious influenza pandemics in humans have involved host range shifts , when an influenza virus jumps from one host species to another . Importantly , however , host range shifts do not always cause pandemics . Rather , epidemiological patterns tend to be unpredictable in new host species , causing disease patterns that change over space and time . In this paper , we analyze epidemiological and evolutionary dynamics of canine influenza virus ( CIV ) , which jumped to dogs in the late 1990s from an equine strain ( EIV ) prevalent in horses . We show that the epidemiology and evolution of CIV is strongly influenced by heterogeneous patterns of infectious contact among dogs in the US . A few large populations in metropolitan animal shelters serve as reservoirs for CIV , but the virus cannot be maintained for long in smaller facilities or in the companion dog population without input from the larger shelters , which represent disease hotspots . These hotspot dynamics give a clear picture of what can happen in the time between the beginning of a host range shift and the onset of a possible pandemic , allowing more targeted strategies for control and eradication . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"systems",
"biology",
"systems",
"ecology",
"medicine",
"and",
"health",
"sciences",
"veterinary",
"epidemiology",
"ecology",
"evolutionary",
"biology",
"population",
"modeling",
"epidemiology",
"population",
"dynamics",
"biology",
"and",
"life",
"sciences",
"population",
"biology",
"computational",
"biology",
"population",
"ecology",
"veterinary",
"science"
] | 2014 | Contact Heterogeneity, Rather Than Transmission Efficiency, Limits the Emergence and Spread of Canine Influenza Virus |
Influenza A virus ( IAV ) enters host cells upon binding of its hemagglutinin glycoprotein to sialylated host cell receptors . Whereas dynamin-dependent , clathrin-mediated endocytosis ( CME ) is generally considered as the IAV infection pathway , some observations suggest the occurrence of an as yet uncharacterized alternative entry route . By manipulating entry parameters we established experimental conditions that allow the separate analysis of dynamin-dependent and -independent entry of IAV . Whereas entry of IAV in phosphate-buffered saline could be completely inhibited by dynasore , a specific inhibitor of dynamin , a dynasore-insensitive entry pathway became functional in the presence of fetal calf serum . This finding was confirmed with the use of small interfering RNAs targeting dynamin-2 . In the presence of serum , both IAV entry pathways were operational . Under these conditions entry could be fully blocked by combined treatment with dynasore and the amiloride derivative EIPA , the hallmark inhibitor of macropinocytosis , whereas either drug alone had no effect . The sensitivity of the dynamin-independent entry pathway to inhibitors or dominant-negative mutants affecting actomyosin dynamics as well as to a number of specific inhibitors of growth factor receptor tyrosine kinases and downstream effectors thereof all point to the involvement of macropinocytosis in IAV entry . Consistently , IAV particles and soluble FITC-dextran were shown to co-localize in cells in the same vesicles . Thus , in addition to the classical dynamin-dependent , clathrin-mediated endocytosis pathway , IAV enters host cells by a dynamin-independent route that has all the characteristics of macropinocytosis .
Influenza A virus ( IAV ) is an enveloped , segmented negative-strand RNA virus infecting a wide variety of birds and mammals . As its first step in infection IAV attaches to host cells by the binding of its major surface protein , the hemagglutinin ( HA ) , to sialic acids , which are omnipresent on the glycolipids and glycoproteins exposed on the surfaces of cells . Where the structural requirements for this interaction have been studied in great detail , much less is known about whether and how the attachment to specific sialylated receptors ( e . g . to N-linked glycoproteins , O-linked glycoproteins or gangliosides or even to specific receptors within these groups ) affects the subsequent endocytic steps . Obviously , knowledge about the repertoire of endocytic pathways that can successfully be used by IAV will increase our insights into cell and species tropism of IAV . In turn , this will contribute to our understanding of the requirements for the generation of novel viruses with pandemic potential that can arise by exchange of RNA segments between currently circulating human serotypes and an animal virus during occasional co-infection in a human or an animal host . Clathrin mediated endocytosis ( CME ) has for long been identified and studied as the major route of IAV cell entry [1] , [2] and is , by far , the best characterized endocytic pathway . Evidence obtained from live cell imaging has revealed the de novo formation of clathrin-coated pits at the site of virus attachment [3] and the requirement for the adapter protein epsin 1 , but not eps15 , in this process [4] . Still , specific transmembrane receptors linking viral entry to epsin 1 or to other adapters have not been identified although a recent study performed in CHO cells indicated the specific requirement for N-linked glycoproteins in IAV entry [5] . Some recent papers provided indications for the utilization of alternative entry pathways by IAV . Studies in which CME was obstructed by pharmacological or genetic intervention indicated the ability of IAV to enter host cells via alternative endocytic routes [4] , [6] , [7] . Also live cell imaging revealed the simultaneous availability of entry routes involving non-coated as well as clathrin-coated pits [4] . However , this alternative IAV entry route has not been characterized in any detail and requirements for any specificity in receptor usage apart from the need for the proper sialic acid moiety have not been established . During the past decades quite a variety of endocytic pathways have been identified in eukaryotic cells [8] , [9] , [10] . Their occurrence , abundance and mechanistic details appear to vary between cell types , tissues and species and their utilization by viruses as a route of entry makes them an important factor in host and cell-type permissiveness for infection [11] , [12] . Besides by CME , different viruses have been shown to enter cells via caveolae , macropinocytosis or other , less well described , routes [11] , [12] . Most often , the selection of a specific endocytic route is linked to the utilization of a specific receptor that facilitates traveling via that particular route . Nevertheless , many receptors allow flexibility by their capacity to enter through multiple pathways . For IAV , an additional level of complexity to the dissection of potential entry routes is added by the apparent lack of an IAV-specific protein receptor . A full experimental characterization of the IAV entry pathways will benefit from separation of the IAV entry pathways into routes that can be studied independently . Whereas co-localization with clathrin is an established marker for endocytosis via this route , the complete lack of unique markers for macropinosomes or most other endocytic compartments [13] , [14] complicates such studies . Furthermore , crucial to any study concerning endocytic pathways is the abundantly documented fact that such pathways are highly dependent on experimental cell culture conditions [15]–[19] . Pathways that are constitutive in one cell type may be absent or inducible by specific experimental conditions in other cell types . Moreover , the manipulation of specific endocytic pathways may result in up or down regulation of other specific pathways . Here we have established entry assay conditions that allow dissecting cell entry of IAV into a dynamin-dependent ( DYNA-DEP ) and a dynamin-independent ( DYNA-IND ) component . Dynamin is a large GTPase forming multimeric assemblies around the neck of newly formed endocytic vesicles . GTP hydrolysis is required for pinching off of the vesicles [20] . Whereas CME is completely dependent on dynamin , several other endocytic routes do not require dynamin [21] . We performed an extensive characterization of the dynamin-independent IAV entry route using pharmacological inhibitors as well as by expressing dominant-negative mutants and applying siRNA induced gene silencing as tools . Taken together the results identify a pathway that closely resembles macropinocytosis as a novel entry pathway for IAV .
To identify and characterize potential non-CME entry routes taken by IAV , we adapted a luciferase reporter assay [22] to enable the quantitative determination of infection or entry by measuring the activity of secreted Gaussia luciferase . Twentyfour hours prior to infection HeLa cells were transfected with a plasmid ( pHH-Gluc ) allowing constitutive synthesis ( driven by the human PolI promoter ) of a negative strand viral RNA ( vRNA ) encoding a Gaussia luciferase under control of the untranslated regions ( UTRs ) of the NP segment of Influenza A/WSN/33 ( H1N1 ) ( hereafter called IAV-WSN ) NP segment . Upon IAV infection , the combined expression of the viral polymerase subunits and NP will drive transcription of luciferase mRNA from the negative strand vRNA and subsequent synthesis of Gaussia luciferase . A dose-response curve demonstrating the applicability of the assay to inhibitor screening ( Fig . 1A ) was obtained for Bafilomycin A1 ( BafA1 ) , a known inhibitor of IAV entry [23] . BafA1 acts upon the vacuolar-type H ( + ) -ATPase , thus preventing endosomal acidification and thereby trapping IAV in peri-nuclear immature endosomes with a lumenal pH that does not permit viral membrane fusion . Remarkably , dynasore , a small molecule inhibitor of the GTPase dynamin 2 that is crucial for endocytic vesicle formation in clathrin- and caveolin-mediated endocytosis [8] as well as in a poorly described clathrin- and caveolin-independent endocytic pathway [8] , [19] , did not give significant inhibition ( Fig . 1B ) . BafA1 specifically inhibits IAV during the entry phase as demonstrated in Fig . 1C . The continuous presence of 10 nM BafA1 ( added to the cells 1 hr prior to infection ) for 16 hrs completely prevents infection . In contrast the addition of BafA1 at 1 hr or 2 hrs post infection resulted in high levels of luciferase activity ( again measured at 16 hrs p . i . ) that were 63% or 90% respectively of the control to which no BafA1 was added , indicating that entry was essentially completed within 2 hrs . The last bar of Fig . 1C shows that the inhibition by BafA1 is reversible as withdrawal of the inhibitor after 2 hrs resulted in high levels of infection . The specific effect of BafA1 on IAV entry was confirmed by confocal microscopy demonstrating that BafA1 , as expected , traps IAV particles in a peri-nuclear location , presumably in non-acidified endosomes ( Fig . 1D ) . BafA1 was subsequently exploited to establish a specific IAV entry assay ( hereafter further referred to as the Gluc-entry assay ) . HeLa cells transfected with pHH-Gluc were inoculated with IAV at a range of MOIs and incubated for 2 hrs after which the entry medium was replaced by complete growth medium containing 10% FCS and 10 nM BafA1 to prevent any further entry of virus . Entry was indirectly quantified by determination of luciferase activity after further incubation for 14 hrs demonstrating a quantitative correlation between infection dose and luciferase activity across a wide range of MOIs ( Fig . 1E ) . The indirect Gluc-entry assay was next tested for its capacity to examine the effects of inhibitors on IAV entry . Dynasore or BafA1 ( Fig . 1F ) were included in the medium ( DMEM containing 10% FCS ) during entry ( the first 2 h of infection ) and were removed when the inoculum was replaced by growth medium containing BafA1 . Concentrations up to 80 µM dynasore did not inhibit entry which is in agreement with the result shown in Fig . 1B . In contrast , 1 . 25 nM BafA1 already inhibited entry for more than 60% ( Fig . 1F ) . As a control , dynasore was also added at 2 hrs post infection to analyze whether the drug affected IAV replication during the post entry phase . As expected , 80 µM dynasore did not significantly inhibit IAV replication when present from 2 to 16 hrs p . i . ( Fig . 1F ) . Thus , with the Gluc-entry assay we can study the effect of specific inhibitors on IAV entry in a quantitative manner , at least as long as the inhibitors do not irreversibly affect IAV replication during the post entry phase . Furthermore , the lack of inhibition of IAV entry by dynasore demonstrates that under these experimental conditions IAV is able to enter cells via a pathway that is fully redundant to any dynamin-dependent ( DYNA-DEP ) entry route , including the classical CME pathway . Also when IAV travels via this novel dynamin-independent ( DYNA-IND ) route , IAV apparently enters via low pH compartments as entry is fully sensitive to BafA1 . As factors present in serum are known for their potential to induce specific endocytic pathways , we further explored the conditions required for the novel DYNA-IND IAV entry pathway ( using the Gluc-entry assay ) by inoculating cells in PBS in the presence of increasing concentrations of fetal calf serum ( FCS ) . Whereas dynasore completely inhibited entry in PBS , inclusion of 5% and 10% FCS resulted in increasing levels of dynasore resistant entry ( Fig . 2A ) , suggesting the existence of a serum-inducible DYNA-IND IAV entry pathway . This effect was not caused by inactivation of dynasore during the experiment as vesicular stomatitis virus ( VSV ) , which enters cells by CME [24] , [25] , was still sensitive to 80 µM dynasore in the presence of 10% FCS ( Fig . 2B ) . In agreement herewith , the uptake of transferin , known to occur via CME , was inhibited by dynasore regardless of the presence of FCS ( Fig . S2 , panel A ) . As expected , both DYNA-DEP entry in PBS and DYNA-IND entry in the presence of 10% FCS and 80 µM dynasore required sialic acid receptors for efficient entry as pre-treatment of HeLa cells with neuraminidases almost completely abolished entry via either pathway ( Fig . 2C ) . The kinetics of the DYNA-DEP and DYNA-IND entry pathways were compared by performing a time-course experiment in which IAV entry was terminated by the addition of 10 nM BafA1 at different time points ( Fig . 2D ) . In comparison to entry via the DYNA-DEP pathway ( the only pathway available in PBS ) entry in the presence of FCS ( when presumably both the DYNA-DEP and DYNA-IND entry pathways are available ) showed similar kinetics . In contrast , entry via the DYNA-IND pathway ( which is the only pathway that is active in the presence of 10% FCS and 80 µM dynasore ) was slower . The difference was most prominent after 15 min , while after 4 hrs similar levels of entry were reached . To validate and extend these results we visualized the reduction of the number of infected cells by immunoperoxidase staining using an antibody against NP ( Fig . 3 ) . A number of different cells of mammalian and avian origin were infected for 2 hours at an MOI of 1 in PBS with or without serum . After 2 hours the inoculum was replaced by growth medium containing 10% FCS and 10 nM BafA1 and the expression of NP was examined after 14 hours later . After incubation in PBS , staining was completely prohibited by the presence of 80 µM dynasore whereas in the presence of serum dynasore had no effect . A serum-inducible , DYNA-IND route of entry was thus functional in all five cell lines , including the human epithelial airway carcinoma cell line A549 . To confirm our results and to obtain further proof for the utilization of DYNA-DEP and DYNA-IND entry routes by IAV , we additionally used an IAV virus-like particle ( VLP ) direct entry assay [26] . These VLPs contain IAV HA and NA in their envelope and harbor a beta-lactamase reporter protein fused to the influenza matrix protein-1 ( BlaM1 ) , which allows the rapid and direct detection of entry , independent of virus replication . Upon fusion of viral and endosomal membrane , BlaM1 gains access to the cytoplasmically retained fluorigenic substrate CCF-2 that , after cleavage by BlaM1 , shifts to a shorter fluorescent emission wavelength that can be detected by flow cytometry . Entry into HeLa cells was performed in the absence or presence of 10% FCS using VLPs containing HA and NA either from IAV-WSN ( having a strict alpha 2–3 linked sialic acid binding specificity ) or from the pandemic 1918 IAV ( HA from A/NewYork/1/18 , binding to alpha 2–3 and alpha 2–6 linked sialic acids; NA from A/BrevigMission/1/18 ) . Entry of VLPs of both IAV strains was severely inhibited by dynasore when no serum was added to the inoculum ( Fig . 4A , 4D ) , whereas the presence of 10% FCS rendered entry completely dynasore resistant . ( Fig . 4B , 4E ) . Quantification of VLP entry is shown in Fig . 4C and F . Importantly , to confirm the existence of the serum-inducible entry pathway by a method that is independent of dynasore , we used siRNA induced silencing of dynamin 2 . Fig . 4G shows that two different siRNAs had a significant inhibitory effect ( 48 hrs after siRNA transfection ) on entry of the Renilla luciferase-encoding pseudovirus WSN-Ren [27] in HeLa cells in the absence of FCS , whereas the presence of 10% serum no reduction in entry levels was observed , confirming the results obtained with dynasore . Knockdown of dynamin 2 protein levels ( 48 hrs after siRNA transfection ) was analyzed by western blotting ( Fig . 4H ) and quantified in Fig . 4I which also shows the knockdown of dynamin 2 mRNA levels as determined by quantitative RT-PCR . We conclude that a DYNA-IND entry pathway can be induced by serum in different cell types from several species . The evidence was obtained using both replication-dependent ( Gluc-entry assay and immunodetection of infected cells ) and replication-independent assays ( entry of VLPs ) , the latter allowing immediate detection of the fusion-mediated delivery of viral M1 protein into the cytoplasm . The DYNA-IND entry pathway was further characterized by inhibitor profiling using an 80-compound kinase inhibitor library . Serum-induced DYNA-IND entry was examined in 10% FCS using the Gluc-entry assay . 80 µM dynasore was added in order to block CME and any other potential DYNA-DEP entry pathways . This allowed the independent inhibitor profiling of the novel pathway by avoiding the potentially masking effect of the presence of redundant entry pathways . Cells were preincubated with the kinase inhibitors ( 10 µM ) for 1 h at 37°C and then inoculated with virus ( MOI 0 . 5 ) in the presence of 10% FCS and 80 µM dynasore for 2 h at 37°C ( DYNA-IND entry ) . In parallel , inoculations were also done in PBS to compare the effects of the inhibitors on DYNA-DEP entry . After 2 hr the medium and inhibitor were replaced by full growth medium containing 10% FCS and 10 nM BafA1 to allow the subsequent expression of Gluc activity under identical conditions for the DYNA-IND and -dependent entry assay . Six kinase inhibitors appeared to act non-discriminatively , inhibiting both DYNA-DEP and DYNA-IND entry ( Fig . 5A ) : the protein kinase C ( PKC ) inhibitors Ro 31-8220 , rottlerin ( both displaying moderate cytotoxicity , result not shown ) and hypericin , which have all three been previously identified as IAV inhibitors [28] , [29]; the highly cytotoxic pan-specific serine/threonine protease inhibitor staurosporine; the irreversible PI-3 kinase inhibitor wortmannin and the receptor tyrosine kinase inhibitor TYR9 . In order to investigate whether some of these inhibitors affect IAV replication during the post-entry phase , we performed the same experiments but now adding the kinase inhibitors after viral entry . Four of the inhibitors thus appeared to induce significant inhibition of post-entry processes ( Fig . 5A ) . Although unlikely , we cannot formally exclude that post-entry processes specific for only one of the two entry pathways are affected . Interestingly , whereas no specific DYNA-DEP entry inhibitors were identified , 15 inhibitors ( none displaying cytotoxic effects , data not shown ) caused significant ( p<0 . 05 ) inhibition ( >5-fold ) of DYNA-IND entry ( Fig . 5B ) . This included inhibitors of the calmodulin dependent kinases myosin light chain kinase ( MLCK ) and CaMKII and seven inhibitors of different growth factor receptor tyrosine kinases . In contrast to the three non-specific PKC inhibitors mentioned above , the PKC inhibitors BIM-1 and HBDDE appeared to have a specific inhibitory effect on DYNA-IND entry . The specific effect of these drugs on DYNA-IND entry is not only shown by the lack of inhibition of DYNA-DEP entry in PBS , but also by the observation that none of the fifteen compounds induced >2-fold inhibition when added post-entry ( at t = 2 hr post infection ) . The kinase library screen was repeated on A549 human epithelial lung carcinoma cells in order to confirm the results in a potentially more natural host cell line . The inhibition profiles obtained were very similar to those found for HeLa cells with the exception of the strong effect of AG879 ( 99% inhibition ) and moderate effects of AG825 ( 39% inhibition ) and Tyr51 ( 68% inhibition ) on DYNA-DEP entry . ( Fig . 5C ) . MLCK inhibitors ML-7 and ML-9 have been reported to be highly specific for their target kinase [30] . Phosphorylation by MLCK activates non-muscle myosin II light chain , indicating that a functional actomyosin network might be essential for DYNA-IND entry of IAV . This was further examined by testing the effect of Blebbistatin , an inhibitor of myosin II heavy chain activity , and of several inhibitors that affect actin dynamics by disrupting actin microfilaments ( Cytochalasin B and D ) , by enhancing actin polymerization ( Jasplakinolide ) or by inhibiting actin polymerization ( Latrunculin A ) . Actin inhibitors were used at the minimal concentration required to induce clearly visible changes in the actin cytoskeleton as pre-determined by staining with FITC-phalloidin ( results not shown ) . Whereas the inhibitors did not affect DYNA-DEP entry ( Fig . 6A ) using Gluc-entry assay , all inhibitors as well as ML-7 and ML-9 significantly inhibited DYNA-IND entry ( Fig . 6B ) . Next , HeLa cells were transfected with plasmids encoding dominant negative or wildtype Rab5 fused to green fluorescent protein ( Rab5 DN and Rab5 wt in Fig . 6 ) 24 h prior to infection with IAV . Rab5 is a small GTPase found in association with several endosomal compartments and crucial for the function and maturation of early endosomes . It is required for the trafficking of a wide range of endocytic cargo following different routes , including DYNA-DEP as well as DYNA-IND routes [31] . Entry of IAV has been shown to require Rab5 [32] . Consistently , we found that HeLa cells expressing Rab5 DN ( as identified by GFP fluorescence , Fig . 6C ) were much less susceptible to productive IAV infection ( as judged by indirect immunofluorescence using Alexa-488 labeled NP antibodies ) than cells transfected with Rab5 wt , both by DYNA-DEP ( 64% inhibition ) and by DYNA-IND ( 47% inhibition ) routes ( Fig . 6D ) . In contrast , when examining the efficiency of DYNA-DEP and DYNA-IND entry routes in cells transfected with a dominant negative mutant of MLCK ( MLCK DN; examined in comparison to MLCK wt ) , only IAV infection under DYNA-IND conditions was significantly reduced ( Fig . 6E; 53% inhibition ) . Similarly , a construct encoding the N-terminal head domain of myosin II MyoII-head ) also only significantly affected DYNA-IND entry ( Fig . 6F; 92% inhibition ) . The combined results indicate that a dynamic actomyosin network requiring the activation of myosin II by MLCK is necessary for efficient entry of IAV via a DYNA-IND pathway . Several dynamin-independent endocytic pathways have been described [8] , [19] . Of these , macropinocytosis has been demonstrated to be stimulated by growth factors present in serum and to depend on actin dynamics [12]–[14] . Yet , studies on macropinocytosis are hampered by a lack of specific inhibitors , cargo , membrane markers and characteristic morphology . Amiloride and the more potent derivative EIPA are inhibitors of epithelial sodium channels ( ENaC ) as well as of several other Na+/H+ antiporters . EIPA has often been used as a hallmark inhibitor that specifically inhibits endocytosis via the macropinocytic pathway [14] . Whereas DYNA-DEP entry of IAV was not inhibited by EIPA ( Fig . 7A ) , DYNA-IND entry was fully blocked EIPA ( Fig . 7B ) . The existence of redundant entry pathways in the presence of 10% FCS is clearly demonstrated by the marginal inhibition by either EIPA or dynasore whereas the combination of EIPA and dynasore resulted in strong inhibition both in the Gluc-entry assay ( Fig . 7C ) and in the direct VLP entry assay ( Fig . 7D and E ) . Supplementary Fig . S1 shows that other cell lines , including the human lung epithelial cell line A549 , display similar IAV inhibition patterns for EIPA and dynasore . Consistently , virus production displayed a similar inhibitor sensitivity profile ( Fig . 7 F and G ) as virus entry indicating that the entry pathways we characterized lead to a productive infection . Clearly , VLPs and viral particles follow similar redundant entry pathways , distinguishable in a DYNA-DEP and a DYNA-IND pathway , the latter being sensitive to EIPA and dependent on actomyosin function . One characteristic of macropinocytosis is the nonselective uptake of large amounts of extracellular solutes [33] . Therefore , the uptake of soluble FITC labeled dextran ( Fdx ) into relatively large vesicles ( 0 . 3 to 5 µM ) has often been applied as a morphological marker for macropinosomes . Using this marker we found that the addition of 10% FCS to the culture medium slightly increased the uptake of Fdx into HeLa cells ( Fig . 8A ) . Notably , the distribution of Fdx changes in response to serum from a random distribution into a more granular pattern . At high magnification and at color settings adjusted to higher intensity it could be seen that these Fdx granules were free of actin staining ( by phalloidin ) indicating that they were in the lumen of vesicles ( result not shown ) . Interestingly , in the presence of IAV ( MOI of 10 ) the uptake of Fdx into vesicles was clearly enhanced . At a higher magnification viral particles could be found to co-localize in Fdx loaded vesicles as well as outside these vesicles ( Fig . 8B ) . Phalloidin staining of actin was used to demonstrate that many virus particles localized to actin-rich protrusions at the periphery of the cell . The uptake of Fdx was studied in a quantitative manner by flow cytometry ( Fig . 8 C ) . A moderate , but reproducible shift to higher Fdx fluorescence was observed at 37°C when virus was added in presence of 10% FCS whereas such a shift was absent when no serum or virus was added . This result confirms the observations by confocal microscopy ( Fig . 8A ) which showed that the combined presence of FCS and IAV increases the uptake of Fdx as compared to FCS alone . In a control experiment the uptake of Fdx in 10% FCS in presence of IAV was shown to be specifically inhibited by EIPA , but not by dynasore ( Fig . S2 , panel B ) . In contrast , transferrin uptake , which serves as a specific marker for CME , was affected by dynasore , but not by EIPA ( Fig . S2 , panel A ) . In conclusion , serum induces the uptake of Fdx into large vesicles , which can be further enhanced by the addition of IAV particles that , after entry , co-localize in part with these vesicles . These results indicate the utilization of a macropinocytic pathway for entry of IAV , which is consistent with the observed sensitivity of the serum-inducible DYNA-IND entry of IAV and VLPs to EIPA . Macropinocytosis has been implicated in the entry of several viruses [12] , [14] . However , differences in susceptibility to inhibitors suggest that distinct forms of macropinocytosis might be used by different viruses [34] , [35] . By screening specific inhibitors in the Gluc-entry assay using DYNA-IND entry conditions we evaluated the possible involvement of a few signaling cascades that have been implicated in the induction of macropinocytosis . Serum-inducible macropinocytosis has been shown to be activated via a myriad of signaling cascades initiated by growth factors binding to transmembrane tyrosine kinase receptors [14] , [17] , [36] , [37] , consistent with the results shown in Fig . 5 . A prominent downstream effect of these signaling cascades is the activation of p21 associated kinase 1 ( PAK1 ) which in turn can activate a number of different pathways leading to actin network rearrangements that can ultimately lead to the induction of macropinocytosis [38] . Fig . 9A–B shows that 20 µM IPA3 , an inhibitor of PAK1 [39] , specifically inhibits DYNA-IND entry of IAV . Activation of PAK1 in response to growth factor stimulation often involves upstream signal transduction by members of the Rho sub-family of small GTPases like CDC42 and/or Rac1 [34] , [40] , [41] . Alternatively , activated CDC42 and Rac1 can induce actin rearrangements independently of PAK1 [34] , [40]–[43] by direct interaction with WASP or WAVE family proteins , respectively [44] , [45] . However , inhibitors of CDC42 ( Pirl1 [46] ) , Rac1 ( NSC23766 [47] ) or N-WASP ( wiskostatin [48] ) did not display inhibitory effects on DYNA-IND or DYNA-DEP entry of IAV ( Fig . 9C–D ) . Instead , Pirl1 and wiskostatin induced a significant , concentration dependent increase of entry . This stimulatory effect was not observed for the control vaccinia virus strain WR , which enters cells via a Rac1-dependent , macropinocytotic pathway [43] ( Fig . 9E ) , indicating that this effect is specific for IAV . The results suggest a requirement for PAK1 in DYNA-IND entry of IAV that does not require activation by either CDC42 or Rac1 . Growth factor inducible activation of the tyrosine kinase src has also been linked to the induction of macropinocytosis [49]–[51]; consistent with this observation the src inhibitor PP2 [52] specifically inhibited DYNA-IND entry of IAV ( Fig . 9A–B ) . Remarkably , 17-AA-geldanamycin , a specific inhibitor of the chaperone protein HSP90 [53] , also caused specific inhibition of DYNA-IND entry ( Fig . 9A–B ) . HSP90 affects the folding and activity of many proteins but the recent demonstration of direct activation of the catalytic activity of src by HSP90 [54] provides another indication of the involvement of src in DYNA-IND endocytosis of IAV . In conclusion , like for other viruses utilizing a macropinocytic entry pathway , PAK1 seems to play a crucial role in DYNA-IND entry by IAV . However , this pathway is independent of Rac1 or cdc42 but may require src , either upstream and/or downstream of PAK1 .
The data presented in this study demonstrate for the first time that IAV can enter cells via DYNA-IND macropinocytosis in addition to the previously described DYNA-DEP classical CME pathway [1] , [2] . Several lines of evidence indicate that the DYNA-IND entry route of IAV that we identified corresponds with macropinocytosis . First of all , the entry pathway is dependent on the presence of serum , a well-known inducer of macropinocytosis . Second , IAV colocalized in vesicles with soluble FITC-dextran , a marker for macropinocytosis . Third , DYNA-IND IAV entry was sensitive to the amiloride-derivative EIPA , the hallmark inhibitor of macropinocytosis [14] , [55]–[58] . Fourth , this IAV entry pathway is sensitive to inhibitors or dominant-negative mutants affecting actomyosin dynamics . Fifth , the specific inhibition of DYNA-IND IAV entry by a number of inhibitors of growth factor receptor tyrosine kinases as well as downstream effectors thereof also points at the involvement of macropinocytosis . Finally , macropinocytosis is independent of dynamin [12] , [14] , [19] . Despite this extensive list of arguments , viral entry by macropinocytosis needs to be considered with caution . The characteristics of the DYNA-IND route of cell entry by IAV are similar , but not identical to the macropinocytic entry routes taken by other viruses , like two different strains of vaccinia virus and by coxsackie virus B [34] , [35] . As is shown in Table 1 and discussed in more detail below , the macropinocytic pathways used by each of these viruses have a few unique characteristics . This may very well reflect the growing notion that macropinocytosis represents a number of differentially induced and regulated processes , rather than being a single endocytic pathway [13] , [14] . Macropinocytosis has collectively been described as an inducible form of endocytosis by which fluid-phase cargo travels via non-coated , relatively large and heterogeneous organelles that have emanated from extensive protrusions ( e . g lamellar ruffles , circular ruffles or retracting blebs ) of the plasma membrane [13] . In the case of DYNA-IND IAV entry more extensive studies using electron microscopy will be required to study the morphology of membrane protrusions with which IAV may associate . In addition , live cell imaging microscopy will be required to characterize the exact itinerary that is taken by IAV virions traveling via a macropinocytic process . This is especially important as different routes of IAV entry are likely to converge at some point in the endocytic pathway . Although unlikely , co-localization of IAV particles with fluid-phase dextran as shown in Fig . 8B may thus represent a situation occurring after convergence of several different routes . The use of microscopy to study macropinocytosis is however complicated by the lack of specific membrane-associated markers for any early step of this endocytic process . A model ( Fig . 10 ) based on our results explains the key steps involved in the macropinocytic entry pathway of IAV , which are described in more detail below . By manipulating the inoculation conditions we were able to experimentally dissect IAV entry into a DYNA-DEP and DYNA-IND route . The DYNA-IND route required the presence of 10% FCS in the entry assay medium . Previously , a strict dependency on a DYNA-DEP entry route for IAV was concluded from experiments with a cell line expressing an inducible dominant-negative mutant of dynamin 2 [59] . In that study , as well as in other entry studies of IAV , entry was performed in DMEM containing 2% serum or BSA . Also in our hands 2 . 5% serum ( Fig . 2A ) or 0 . 2% BSA ( result not shown ) was not sufficient to allow DYNA-IND entry . We are currently investigating which serum component is responsible for the observed effects on IAV entry . Dialysis of FCS ( MW cut off >10 kDa ) did not affect its capacity to induce DYNA-IND endocytosis ( result not shown ) , indicating that low molecular weight solutes are not responsible for the observed effect . Our evidence for a DYNA-DEP and a serum inducible DYNA-IND entry route is based on the use of pharmacological ( dynasore , a highly specific inhibitor of dynamin ) as well as genetic ( siRNA directed against dynamin 2 ) tools , ruling out the possibility that the inhibitory effect of dynasore was due for instance to absorption of the inhibitor by serum components . Whereas dynasore resulted in near 100% inhibition of DYNA-DEP entry , only 65% inhibition was observed upon siRNA induced silencing of dynamin 2 indicating that the residual levels of dynamin 2 that remain after 48 hrs of silencing still support a low level of DYNA-DEP entry ( Fig . 4H ) . Reversible inhibitors like dynasore [60] offer a major advantage for characterization of IAV entry pathways . They can be applied for a limited period thus preventing the secondary adaptive effects of cells that may occur in response to long-term down regulation of a gene product by genetic methods like siRNA interference . Both entry routes were consistently identified by a viral entry assay quantified by virus induced expression of a luciferase reporter as well as by a VLP entry assay allowing direct analysis of the membrane fusion mediated entry step . The consistent performance of an HA with a strict preference for binding to α2-3 linked sialic acids ( from IAV-WSN; our unpublished data ) and an HA also binding to α2-6 linked sialic acids ( from 1918 IAV [61] ) in the VLP entry assay indicates that both pathways can be utilized by HAs of different specificity and may therefore be relevant to avian as well as human IAV infections . Consistently , serum-inducible DYNA-IND entry was observed both in avian DF1 cells and in a human lung epithelial carcinoma cell line A549 ( Fig . 3 ) . The DYNA-DEP and DYNA-IND IAV entry pathways were found by our quantitative assays to be fully redundant . In the presence of serum , the combination of dynasore ( inhibiting DYNA-DEP entry ) and EIPA ( inhibiting DYNA-IND entry ) completely abolished entry whereas either drug alone had no effect . EIPA , an inhibitor of plasma membrane Na+/H+ exchangers , has been shown to invariably inhibit macropinocytosis [14] , [55]–[58] . As other routes of endocytosis are generally not affected , EIPA is considered as a hallmark inhibitor of macropinocytosis [14] , although results obtained with EIPA should be considered with care as long as a mechanistic explanation for its effect on macropinocytosis is not yet fully clear [62] . Occasionally , a moderate two- to three-fold inhibition by dynasore alone was observed ( result not shown ) indicating that the capacity of the serum-inducible entry pathway is somewhat variable , possibly depending on slight variations in serum quality and factors like cell distribution in the wells that have been reported to influence viral infection [63] . A redundancy in the utilization of CME as well as a clathrin-independent route for entry of IAV has been visualized previously by quantitative live cell imaging [4] . Both routes were operative simultaneously in the same sample and the specific down-regulation of CME did not affect the total number of entry events . In response to specific extra-cellular signals ( e . g . serum induction ) , changes in the actomyosin network occur that give rise to membrane protrusions required for macropinosome formation [13] . Compounds inhibiting actin polymerization ( cytochalasin B and D ) , depolymerization ( jasplakinolide ) or sequestering soluble actin ( latrunculin A ) all specifically inhibited DYNA-IND IAV entry . In addition , the requirement for myosinII activity was established by a specific inhibitor ( Blebbistatin ) of myosin II ATPase activity and by the expression of a dominant negative mutant of myosinIIA heavy chain . Also , the regulation of myosinII activity by phosphorylation of myosin light chain through the action of MLCK is suggested by the inhibitory effect of MLCK inhibitors ML-7 and ML-9 as well as by the similar effect of an expressed MLCK dominant negative mutant . Recently , a function for the actin cytoskeleton in IAV entry was reported to be required for the entry into polarized epithelial cells but not for entry into non-polarized cells [64] . When using the low-serum conditions used in that paper ( 2% FCS ) , we only observed DYNA-DEP entry that was not affected by actin dynamics inhibitors . Perhaps , the polarized cells permit DYNA-IND entry at lower serum concentrations . The changes in actin network dynamics that can lead to the formation of macropinosomes can be triggered by a number of signaling cascades . Actin dynamics are induced by the activation of growth factor receptor tyrosine kinases by their respective growth factor ligands that are normally present in serum [12]–[14] , [17] , [36] , [37] The signal transduction cascades that link activation of growth factor receptor tyrosine kinases to actin remodeling and macropinocytosis are only beginning to be revealed . The specific inhibition of DYNA-IND entry of IAV by IPA3 , an inhibitor of PAK1 , provides proof for the involvement of these cascades . PAK1 is a key serine/threonine kinase regulating actin network dynamics but its crucial function in several pathways of endocytosis as well as numerous other cellular processes does not make it a very specific marker [65] . Even so , macropinocytosis has consistently been demonstrated to require PAK1 activation , both in the induction of the process and/or in further downstream trafficking events of macropinosomes [13] , [14] . Growth factor dependent activation of PAK1 has most often been demonstrated to depend on upstream activation of small GTPases Rac1 or cdc42 [34] , [40] , [41] . Different strains of vaccinia virus were recently shown to induce their uptake by macropinocytosis via activation of either Rac1 or cdc42 [34] . Activation of Rac1 has been linked to the induction of macropinocytosis via actin network-mediated formation of lamellipodia and/or circular ruffles whereas cdc42 has most often been implied in the formation of filopodia [44] . An inhibitory effect of the Rac1 inhibitor NSC23766 or the cdc42 inhibitor pirl1 on IAV entry , however , could not be demonstrated . Remarkably , cdc42 inhibitor pirl1 enhanced IAV entry and a similar effect was observed by wiskostatin , an inhibitor of N-WASP which functions directly downstream of cdc42 as a scaffolding complex required for the activation of actin polymerization leading to filopodia formation . Similarly , the macropinocytosis-like entry pathway taken by Coxsackie B virus was also shown to require PAK1 activity that was independent of Rac1 activation [35] . Direct examination of the magnitude and timing of the activation of PAK1 will be required to obtain more insight in the involvement of this complex pathway . The induction of macropinocytosis by a PAK1-dependent mechanism has been associated with ruffling at the cell membrane [12] , [14] , [15] , [37] . The identification of sub-membranous regions with increased actin staining by phalloidin has been interpreted as evidence for ruffling . This was not unambiguously identified by confocal microscopy in the experiments presented in Fig . 8 and Fig . S2 and needs to be investigated in depth by life cell imaging techniques . In agreement with our observation that the DYNA-IND entry of IAV was inhibited by PP2 , an inhibitor of src family kinases , the non-receptor tyrosine kinase c-src has been shown to function as a key signaling intermediate in the induction of macropinocytosis via a mechanism independent of Rac1 or cdc42 [49]–[51] . Downstream effects of c-src on actin networks proceed , amongst others , via phosphorylation of cortactin by c-src resulting in accelerated macropinosome formation [50] . C-src has been shown to associate with macropinosomes [49] , [51] , both during their formation and their trafficking , while c-src kinase activity is required for macropinocytosis following EGF stimulation of HeLa cells [49] . Interaction of HSP90 with c-src was recently shown to induce c-src kinase activity [54] . Also HSP90 has been demonstrated to associate with macropinosomes , while its specific inhibitor geldanamycin reduced the membrane ruffling that preceded macropinocytosis [66] . Thus , the inhibition of IAV entry via macropinocytosis by AA-geldanamcyin may very well involve the effects of HSP90 on c-src . As detailed above , the DYNA-IND entry pathway of IAV shares many characteristics with the endocytic pathway macropinocytosis . This is corroborated by the observation that IAV particles and dextran colocalize in large vesicles in the presence of FCS . Several viruses have recently been reported to enter cells via macropinocytosis [12] , [14] . Apart from common factors like the requirement for PAK1 activation , actin dynamics and independence of dynamin , virus specific details have been described [34] , [35] ( Table 1 ) . In part these might be contributed to differences in experimental conditions ( e . g . cell types tested ) but diversity in the molecular mechanisms by which macropinocytosis can be induced and executed is likely to exist and to be exploited by viruses . Whereas vaccinia virus is able to trigger its own macropinocytic uptake [34] , [43] , we have described a macropinocytosis pathway that is operational under conditions that are activated by components in serum . Still , this does not exclude signaling induced by virus-host cell interactions , which are for instance suggested by the significant increase of FITC-dextran uptake in the presence of IAV . The possible requirement for co-stimulatory signals from serum components and virus imposes an additional layer of complexity on the analysis of IAV entry via DYNA-IND pathways . Influenza viruses cause respiratory infections by targeting the epithelial cells lining the respiratory tract . These surfaces are covered by a mucous layer composed of a variety of small solutes and glycoproteins derived among others from goblet cells [67] . This semi-fluid layer in turn conditions the underlying cells and determines their physiological state , including the activities of their uptake and secretion pathways . It will be important to determine to what extent the DYNA-DEP and DYNA-IND IAV entry pathways are operational under the conditions prevailing along the respiratory tract . Current knowledge on the protein composition of the fluids covering the respiratory epithelium is rapidly expanding by the application of proteomic methods to determine the protein composition of bronchial alveolar lavage fluids ( BALF ) . These studies have extended the previous notion that BALF is highly similar in composition to serum . For example , just as for the serum proteome more than 85% of the total protein mass of the BALF proteome is accounted for by albumin , immunoglobulins , transferring , α1-antitrypsin and haptoglobin . In addition , many other proteins have been identified both in serum and in BALF including growth factors that can bind to growth factor receptor tyrosine kinases [68]–[70] . Thus , BALF is likely to harbor , just as serum , the protein factors that can activate signaling pathways that are crucial for the induction of DYNA-IND entry of IAV . In agreement herewith , macropinocytosis has been described as a functional entry pathway of Haemophilus influenzae into primary human bronchial epithelial cells [71] although the factors involved in signaling the process have not been identified yet . In addition to infecting the respiratory tract , IAV has been shown to be able to cause systemic infections involving multiple organs . This has mainly been studied in avian infections [72] , [73] or by infection of mice with human-derived H1N1 or H3N2 IAVs [74] but is poorly documented for human infections and may have been underestimated thus far . Obviously , during potential systemic spreading of IAV , the serum-rich conditions that we have demonstrated here to enable the use of alternative entry pathways will be encountered and may contribute to such spreading .
MDCK , A549 , DF-1 and HeLa cells were maintained in complete Dulbecco's Modified Eagle's Medium ( DMEM ) ( Lonza , Biowittaker ) containing 10% ( v/v ) fetal calf serum ( FCS; Bodinco B . V . ) , 100 U/ml Penicillin , and 100 µg/ml Streptomycin . Chinese Hamster-E36 cells were maintained at 37°C in α-Minimal Essential Medium ( Gibco ) supplemented with 10% ( v/v ) FCS , 100 U/ml Penicillin , and 100 µg/ml Streptomycin . Cells were passaged twice weekly . Influenza A/WSN/33 ( H1N1 ) ( IAV-WSN ) was grown in MDCK cells . Briefly , ∼70% confluent MDCK cells were infected with IAV-WSN at a MOI of 0 . 02 . Supernatant was harvested after 48 hr of incubation at 37°C and cell debris was removed by centrifigutation ( 10 min at 2000 rpm ) . Virus was stored at −80°C and virus titers were determined by measuring the TCID50 on HeLa cells . The IAV-WSN luciferase pseudovirus ( WSN-Ren ) system has previously been described [27] . Briefly , WSN-Ren pseudovirus harbors a HA segment in which the HA coding region is replaced by Renilla luciferase . The pseudovirus is produced in a MDCK cell line that stably expresses the HA of IAV-WSN . WR-LUC , a firefly luciferase encoding vaccinia virus ( strain WR ) was previously described [75] . VSV-FL , a firefly luciferase encoding VSV virus was also previously described [76] . Stocks of bafilomycin A1 ( BafA1 ) , dynasore , cytochalasin D , cytochalasin B , Blebbistatin , 17-AA-geldanamycin , ML-7 , ML-9 , PP-2 , 5- ( N-ethyl-N-isopropyl ) amiloride ( EIPA ) , IPA-3 ( all obtained from Sigma-Aldrich ) , Latrunculin A ( Enzo ) , jasplakinolide , wiskostatin , NSC23766 ( all obtained from Calbiochem ) and pirl1 ( Chembridge ) were prepared in dimethylsulfoxide ( DMSO ) . All stocks were stored at −20°C . A kinase inhibitor library composed of 80 kinase inhibitors was obtained from Biomol ( 2832A[V2 . 2] ) . HeLa cells ( 10 , 000 cells/well in 96-well plates ) were treated with 2 mUnits of Vibrio cholerae neuraminidase ( Roche ) in 50 µl phosphate-buffered saline ( PBS ) for 2 hr . After washing with PBS cells were infected with IAV as described . Virus-like particles ( VLPs ) were produced as described [26] . Briefly , 293T cells were transfected using Lipofectamine 2000 ( Invitrogen ) with pCAGGS-BlaM1 ( encoding a beta-lactamase reporter protein fused to the influenza matrix protein-1 ) , pCAGGS-HA ( encoding HA derived from either A/NewYork/1/1918 or IAV-WSN ) and pCAGGS-NA ( encoding IAV neuraminidase [NA] derived from either A/BrevigMission/1/18 or IAV-WSN ) and maintained in OptiMEM . Supernatants were harvested 72 h after transfection and centrifuged to remove debris . VLPs were used for inoculation of cells without further concentration . VLPs were incubated for 30 min at 37°C with trypsin/TPCK for activation of HA . MDCK or HeLa cells grown to near confluency in 24-well plates were inoculated with 250 ul of VLPs after pre-treatment of the cells with inhibitors as indicated . Infection was synchronized by centrifugation at 1500 rpm for 90 min at 4°C and was performed by further incubation at 37°C for 2 h in the absence or presence of 10% FCS and inhibitors as indicated . Detection of beta-lactamase activity was performed as described [25] by loading cells with CCF2-AM substrate ( InVitrogen ) and subsequent analysis by flow cytometry on a LSRII flow cytometer ( Becton Dickinson ) . Typically 10 , 000 events were collected and analyzed using FlowJo 8 . 5 . 2 software . The reporter construct pHH-Gluc was derived from plasmid pHH-Fluc [22] by replacing the firefly luciferase coding region with the Gaussia luciferase coding region of pGluc-basic ( New England Biolabs ) . Unique SpeI and XbaI restriction sites were introduced into pHH-Fluc using the Quikchange XL Site-directed mutagenesis kit ( Stratagene ) and oligonucleotides Spe4262 ( 5-′GCCTTTCTTTATGTTTTTGGCACTAGTCATTTTACCGATGTCACTCAG ) , Spe4263 ( 5′-CTGAGTGACATCGGTAAAATGACTAGTGCCAAAAACATAAAGAAAGGC ) , Xba4260 ( 5′-GTATTTTTCTTTACAATCTAGACTTTCCGCCCTTCTTGG ) and Xba4261 ( CCAAGAAGGGCGGAAAGTCTAGATTGTAAAGAAAAATAC ) . A SpeI site was introduced by site-directed mutagenesis in pGluc-basic directly following the start codon of the Gaussia luciferase coding sequence . The unique SpeI – XbaI fragment of pGluc-basic was subsequently cloned into the SpeI-XbaI site of pHH-Fluc resulting in plasmid pHH-Gluc . Cells were seeded in 96-well plates at a density of 10 , 000 cells/well and transfected the next day with 10 ng pHH-Gluc using Lipofectamine 2000 ( InVitrogen ) according to the manufacturer's protocol . After 24 hrs the transfected cells were treated with inhibitors and infected as indicated . At 16 hr p . i . samples from the supernatant were assayed for luciferase activity using the Renilla Luciferase Assay system ( Promega ) according to the manufacturer's instructions , and the relative light units ( RLU ) were determined with a Berthold Centro LB 960 plate luminometer . WR-LUC and VSV-FL were used to inoculate HeLa cells ( 10 , 000 cells/well ) at an MOI of 2 , in complete Dulbecco's Modified Eagle's Medium ( DMEM ) ( Lonza , Biowittaker ) . After 7 hr the luciferase activity was detected using the SteadyGlo assay kit ( Promega ) . The addition of 10% ( v/v ) FCS did not change infection levels for both viruses . Cells were fixed with 3 . 7% paraformaldehyde ( PFA ) in PBS and subsequently permeabilized with 0 . 1% Triton-X-100 in PBS . After blocking with normal goat serum IAV-infected cells were incubated for 1 h with a monoclonal antibody directed against the nucleoprotein ( NP ) ( HB-65; kindly provided by Dr . Ben Peeters ) . After washing , the cells were incubated with a 1∶400 dilution of Alexa Fluor 488- or 568-labeled goat anti-mouse IgG ( Molecular Probes ) secondary antibody for 1 h . Nuclei were subsequently stained with TOPRO-3 and after three washing steps , the coverslips were mounted in FluorSave ( Calbiochem ) . Actin was stained using phalloidin labeled with Alexa Fluor 633 . The immunofluorescence staining was analyzed using a confocal laser-scanning microscope ( Leica TCS SP2 ) . FITC , GFP or Alexa Fluor 488 were excited at 488 nm , Alexa Fluor 568 at 568 nm , and TOPRO-3 at 633 nm . HeLa cells were grown in 24-well plates on glass coverslips ( 50 , 000 cells/well ) . Prior to FITC-dextran uptake cells were serum-starved for 2 hr in PBS . FITC-dextran ( MW70 , 000 , Sigma-Aldrich ) was incubated with HeLa cells ( final concentration of 0 . 5 mg/ml ) in 500 µl PBS or in PBS containing 10% FCS in the absence or presence of IAV ( strain WSN; MOI 10; concentrated and purified by centrifugation through a 15 to 30% sucrose gradient with a 50% sucrose cushion at the bottom ) at 37°C . After 15 min cells were washed 4 times with PBS at 4°C , fixed with 3 . 7% PFA in PBS and subsequently permeabilized with 0 . 1% Triton-X-100 in PBS . Slides were stained for examination by confocal microscopy as described above . For quantification of FITC-dextran uptake 1 . 5×105 HeLa cells were infected with IAV-WSN ( MOI 10 ) in suspension in a volume of 1 ml in the presence of FITC-Dextran ( 1 mg/ml ) . Infections were performed for 15 min in PBS ( containing 2% BSA to reduce unspecific binding of FITC-Dextran ) or in PBS containing 10% FCS at 37°C or at 4°C ( control for binding of FITC-Dextran to cells in the absence of endocytosis ) . Mock-infected samples were analysed in parallel . Infection was terminated by addition of 3 ml ice-cold PBS followed by three washes with cold PBS and fixation with 3 . 7% PFA . 20 , 000 cells were analyzed by FACS and results were represented as the mean fluorescence which was plotted relative to the uptake in the mock-infection in PBS ( after subtraction of background fluorescence obtained at 4°C ) . HeLa cells ( grown on glass cover slips ) were incubated at 4°C for 1 hr with 50 µg/ml Alexa633-labeled Transferin ( InVitrogen ) in PBS . After 1 hr the medium was replaced by PBS or PBS supplemented with 10% FCS containing IAV ( strain WSN; MOI 10 ) and 0 . 5 mg/ml FITC-Dextran ( Sigma; 70 kDa ) and cells were transferred to 37°C for 15 min . After 15 min cells were fixed and stained as described above and examined by confocal microscopy . Cells were fixed with 3 . 7% PFA in PBS and subsequently permeabilized with 0 . 1% Triton-X-100 in PBS . Peroxidase was visualized using an AEC substrate kit from Vector Laboratories . IAV-positive cells were detected using bright-field light microscopy . Two siRNA duplexes targeting different sites within the coding sequences of dynamin 2 were obtained from Ambion Inc ( 15581 ( Dynamin 2 siRNA 1 ) and 146559 ( dynamin 2 siRNA2 ) ) . A scrambled siRNA ( Ambion Inc . ) was taken along as a control for non-specific effects of the transfection procedure and was used for normalization . One day after seeding in 96-well plates ( 6 , 000 cells/well ) , the HeLa cells were transfected with a final concentration of 10 nM siRNA using oligofectamine ( Invitrogen ) . 48 h after transfection , the cells were inoculated with the WSN-Ren pseudovirus ( MOI 0 . 5 ) in PBS or in PBS containing 10% FCS . After 2 h of infection the entry medium was replaced by complete growth medium containing 10 nM BafA1 to prevent further entry . At 16 h post infection intracellular Renilla luciferase expression was determined as described above . Each siRNA experiment was performed in triplicate . Cell viability was not affected as determined by performing a Wst-1 cell-viability assay ( Roche ) . Functional knockdown of dynamin 2 mRNA levels was performed by quantitative RT-PCR . using a TaqMan Gene Expression Assay for DNM2 ( Hs00191900_m1 , Ambion ) and using 18S RNA ( Hs03928985_g1 , Ambion ) as a control for normalization . The comparative Ct-method was used for quantification of the results [77] . Reduction of dynamin 2 protein levels was determined by western blotting using polyclonal goat-anti-dynamin 2 C18 ( Santa-Cruz SC-6400 ) . A monoclonal against alpha-tubulin ( DM1A , Sigma T9026 ) was used to detect tubulin for normalization . Results were quantified by Densitometric scanning of the dynamin 2 and tubulin signals displayed in Fig . 4H . HeLa cells were grown in 24-well plates on glass coverslips ( 50 , 000 cells/well ) for 24 hrs . Cells were then transfected ( 1 µg of DNA with lipofectamine 2000 as described above ) with plasmids encoding wild-type or dominant-negative ( DN ) human MLCK fused to GFP [78] , wild-type or DN Rab5 fused to GFP [79] , or MyoII-tail or MyoII-head domain fused to GFP [80] . 24 hr after transfection cells were inoculated with IAV-WSN ( MOI 1 ) in PBS or in PBS containing 10% FCS and 80 µM dynasore . 4 hr after infection cells were fixed and stained for examination by confocal microscopy as described above . An unpaired Student's t-test was used for detemination of statistically significant differences . The use of the term significant in text refers to a comparison of values for which p<0 . 05 . | Attachment to and entry into a host cell are the first crucial steps in establishing a successful virus infection and critical factors in determining host cell and species tropism . Influenza A virus ( IAV ) attaches to host cells by binding of its major surface protein , hemagglutinin , to sialic acids that are omnipresent on the glycolipids and glycoproteins exposed on the surfaces of cells . IAV subsequently enters cells of birds and a wide variety of mammals via receptor-mediated endocytosis using clathrin as well as via ( an ) alternative uncharacterized route ( s ) . The elucidation of the endocytic pathways taken by IAV has been hampered by their apparent redundancy in establishing a productive infection . By manipulating the entry conditions we have established experimental settings that allow the separate analysis of dynamin-dependent ( including clathrin-mediated endocytosis ) and independent entry of IAV . Collectively , our results indicate macropinocytosis , the main route for the non-selective uptake of extracellular fluid by cells , as an alternative IAV entry route . As the dynamin-dependent and -independent IAV entry routes are redundant and independent , their separate manipulation was crucial for the identification and characterization of the alternative IAV entry route . A similar strategy might be applicable to the study of endocytic pathways taken by other viruses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] | [
"virology/host",
"invasion",
"and",
"cell",
"entry",
"virology"
] | 2011 | Dissection of the Influenza A Virus Endocytic Routes Reveals Macropinocytosis as an Alternative Entry Pathway |
Vector competence of Aedes aegypti mosquitoes is a quantitative genetic trait that varies among geographic locations and among different flavivirus species and genotypes within species . The subspecies Ae . aegypti formosus , found mostly in sub-Saharan Africa , is considered to be refractory to both dengue ( DENV ) and yellow fever viruses ( YFV ) compared to the more globally distributed Ae . aegypti aegypti . Within Senegal , vector competence varies with collection site and DENV-2 viral isolate , but knowledge about the interaction of West African Ae . aegypti with different flaviviruses is lacking . The current study utilizes low passage isolates of dengue-2 ( DENV-2-75505 sylvatic genotype ) and yellow fever ( YFV BA-55 -West African Genotype I , or YFV DAK 1279-West African Genotype II ) from West Africa and field derived Ae . aegypti collected throughout Senegal to determine whether vector competence is flavivirus or virus genotype dependent . Eight collections of 20–30 mosquitoes from different sites were fed a bloodmeal containing either DENV-2 or either isolate of YFV . Midgut and disseminated infection phenotypes were determined 14 days post infection . Collections varied significantly in the rate and intensity of midgut and disseminated infection among the three viruses . Overall , vector competence was dependent upon both viral and vector strains . Importantly , contrary to previous studies , sylvatic collections of Ae . aegypti showed high levels of disseminated infection for local isolates of both DENV-2 and YFV .
Aedes aegypti is the primary vector of yellow fever virus ( YFV ) and all four serotypes of dengue viruses ( DENV1-4 ) , as well as being a known vector of chikungunya virus . Dengue remains an important public health problem with an expected 390 million cases per year [1] . Although there are fewer dengue cases in Africa compared to other regions , there are 11 African countries endemic for DENV , and the World Health Organization ( WHO ) continually reports DENV outbreaks in previously unreported geographic areas . Furthermore , it was recently estimated that in 2007 alone , 656 million fevers occurred in African children under the age of five [2] . However , in only 78 million of these cases was it likely that the child was infected with Plasmodium falciparum . Despite the enormous number of acute non-malarial febrile illnesses in sub-Saharan Africa , their etiologies are poorly defined [2] . Despite an effective vaccine , YF outbreaks still occur . Cases are occasionally reported in South America ( 11 endemic countries ) , but a majority of the cases are in Africa ( 33 endemic countries ) , and most of the outbreaks are in West Africa [3] . According to the WHO 200 , 000 cases of YFV causes 30 , 000 deaths each year [4] . Both YFV and DENV belong to the family Flaviviridae and have a single-stranded positive sense RNA genome . There are four antigenically distinct serotypes of DENV ( 1-4 ) and all four serotypes are currently found in Africa . In West Africa , DENV-2 is an important serotype because it includes the sylvatic genotype with a high potential for emergence [5]–[7] . Sylvatic genotypes are transmitted between monkeys in forested areas , while cosmopolitan genotypes are transmitted between humans in urban areas . The sylvatic genotype has been isolated from mosquitoes , monkeys , and humans [8]–[10] in West Africa , but is genetically distinct from epidemic isolates [11] , [12] . Different genotypes as well as different lineages within genotypes can result in differences in both vector capacity and the severity of human disease [13] . Genetic differences in YFV isolates are also an important predictor of vector capacity . There are seven genotypes of YFV found worldwide , two of which ( West African Genotypes I and II ) are endemic in West Africa . The genetic differences among YFV isolates are geographically associated with outbreaks in Africa . Specifically , West Africa genotype I is responsible for a majority of outbreaks and is genetically heterogeneous relative to other genotypes [14] . The vector for both viruses , Aedes aegypti ( L ) , exists as two subspecies: Ae . aegypti aegypti and Ae . aegypti formosus [15] , [16] . Characters that distinguish the two subspecies were developed primarily in East Africa but are contradictory and confusing when identifying Ae . aegypti forms collected in West Africa . The current definition of the subspecies is based on the number or degree of white scales on the first abdominal tergite as defined by Mattingly and McClelland [15] , [16] . Aedes aegypti aegypti has scales on the first abdominal tergite , has a light tan cuticle , is globally distributed , tends to be endophilic , and has a feeding preference for humans . In contrast , Ae . aegypti formosus has no white scales on the first abdominal tergite , has a dark or black cuticle , is found mostly in Sub-Saharan Africa in sylvatic environments , tends to be exophilic , and has a feeding preference for wild animals [17]–[19] . However , these distinctions become unclear in West Africa where Ae . aegypti with a dark black cuticle and white scales ( albeit usually few ) are frequently detected . Furthermore , Ae . aegypti without scales frequently breed near human habitats and bite humans . In East Africa , the scaling pattern and behavior are also correlated with discrete genetic differences in allozymes and microsatellites [20] , [21] . But in West Africa , the scaling pattern does not correlate with these genetic markers [22] , [23] or behavioral differences and leads to confusion in subspecies identification . Further , these genetic studies reveal that East African Ae . aegypti aegypti and Ae . aegypti formosus are genetically distinct from the monophyletic West African Ae . aegypti [20] , [21] . Due to this ambiguity the present study will hereafter refer to all Ae . aegypti collections based on their breeding site , habitat , and phytogeographic region of the collection site in Senegal . Previous vector competence studies on Ae . aegypti from West Africa have shown these mosquitoes to be more refractory for both DENV [24] , [25] and YFV[24] , [26] compared to Ae . aegypti collected worldwide . Studies that have specifically examined collections across Senegal showed wide variation in vector competence for both high-passage [22] and low passage field isolates of DENV-2 [27] , [28] . In particular , sylvatic collections from southeastern Senegal were more refractory than other collections from throughout Senegal . Great variation in vector competence is seen within closely related collections of Ae . aegypti [29]–[32] and different isolates of DENV-2 [27] , [28] or YFV [33] . Furthermore , geographically distinct collections of Ae . aegypti from Senegal are genetically diverse [22] , [23] . It has been demonstrated that vector competence of Ae . aegypti for DENV is governed by interactions between mosquito strain and virus genotype in natural collections [34] . This underscores the importance of using viruses and vectors that are geographically proximate and genetically diverse to draw conclusions about vector competence among collections . However , mosquito strain by flavivirus genotype interactions have yet to be examined in West African Ae . aegypti . Therefore , the objective of the current study was to examine these interactions by quantifying the vector competence of West African Ae . aegypti populations to DENV-2 and YFV field isolates .
Aedes aegypti were collected as larvae in 8 locations in Senegal ( Table 1 ) . The sylvatic collections from the southeast ( PK10 [35] , and Kedougou , ) were made in 2011 , all others were made in 2010 . The larvae were transported to a temporary local laboratory in Kedougou or Theis , reared to adults , and given a bloodmeal to generate eggs to start a colony . Eggs from each collection were brought back to Colorado State University and maintained for approximately 10–15 generations before being challenged with an artificial bloodmeal containing virus as described below . It was intended to use collections with fewer generations , but we were not able to get F1 mosquitoes collected from the field to survive , mate , or bloodfeed sufficiently to perform vector competence assays . Adult mosquitoes were kept in incubators maintained at 28°C with 70–80% relative humidity and a 12∶12 hour photoperiod . Eggs were collected on filter papers and stored for up to 5 months in a high humidity chamber . The collection sites consist of domestic sites around huts in urban environments and rural villages , tires in urban environments , and forests ( sylvatic ) where mosquito larvae were collected from treeholes and the discarded fruit husks of Saba senegalensis [35] ( Table 1 ) . Two low passage genotypes of YFV from West Africa were used in these studies . The YFV BA-55 from Nigeria is representative of West African Genotype I and YFV DAK 1279 from Senegal is a West African Genotype II [14] . YFV BA-55 was isolated from a human during an outbreak in Nigeria [36] and has been used in previous vector competence studies [26] , [33] . The DENV-2 isolate was from a sylvatic mosquito in Kedougou , Senegal ( Table 2 ) . YFV BA-55 and DAK 1279 were inoculated into Vero cells and DENV-2-75505 was inoculated into C6/36 cells to generate a virus stock . C6/36 cells were used for DENV-2-75505 to insure virus recovery because this isolate came from a mosquito and there was no data on the titer of the initial stock . The supernatants from these infections were clarified and aliquoted and stored in minimum essential media ( MEM ) with 20% fetal bovine serum ( FBS ) at −80°C for all subsequent use . Growth curves were performed on all viruses to determine the optimal number of days post-infection on which to harvest the virus for mosquito oral infection . For mosquito feeds , C6/36 cells were infected with DENV-2-75505 at a multiplicity of infection ( MOI ) of 0 . 001 or YFV BA-55 and DAK 1279 at an MOI of 0 . 01 and grown to a titer of approximately nine logs of plaque forming units ( PFU ) before being mixed with defibrinated sheep blood for a titer of approximately six logs for the mosquito feeds . The medium was removed from the C6/36 cells infected with DENV-2-75505 six days post-infection and replaced with fresh medium . Final virus was harvested 12 days post-infection and fed directly to mosquitoes . Both isolates of YFV were grown in C6/36 cells for five days before being harvested and fed directly to mosquitoes . Groups of approximately 30 five to seven day old mosquitoes from each colony were exposed to a bloodmeal containing approximately six to seven logs of virus for 30 minutes . The bloodmeal titer was determined from the blood before the feed . Almost all of the mosquitoes fed within 10 minutes , so the pre-feed blood represents the bloodmeal titer ingested . Females that were not completely engorged and males were removed from the study immediately following the bloodmeal . The fully engorged female mosquitoes were held for 14 days at 28°C , 70–80% relative humidity , 12∶12 hour photoperiod , and were fed water and raisins under BSL3 containment . After 14 days , legs , heads/thoraces , and midguts were separated into individual tubes . Each tissue was triturated in 100 µL ( legs and midguts ) or 200 µL ( heads/thoraces ) minimum essential medium ( MEM ) media with 20% fetal bovine serum and 1 . 5 µg/ml Fungizone . Virus titer in each sample was determined by plaque assay . The manipulation of DENV-2-JAM1409 is very similar and was previously described [37] . All plaque assays were performed on Vero cells in 12-well tissue culture plates . When Vero cells reached 95% confluency , each triturated sample was diluted and added directly to the cells and allowed to incubate for 1 hour at 37°C . After 1 hour , the first overlay ( 19 . 6% 10× Earle's Buffered Salt Solution , 63 . 4% water , 6 . 6% Yeast extract-Lactalbumin hydrolysate , 4% fetal bovine serum , 6% sodium bicarbonate , 0 . 3% Gentamycin , and 0 . 3% Fungizone mixed 1∶1 with 2% SeaKem LE agarose ) was applied . Four days post infection with either YFV isolate , or 7 days with DENV-2-75505 , the second overlay ( same as the first overlay plus 2 . 0 ml of Neutral Red per 100 ml overlay ) was applied . Plaques were counted for 3 days following the addition of the second overlay . The plaque assays for DENV-2-JAM1409 were similar and have been previously described [37] . For each collection , disseminated infection ( DI ) , midgut escape barrier ( MEB ) rates , and midgut infection barrier ( MIB ) rates were calculated . The number of mosquitoes with a DI is the number of mosquitoes with virus in the head/thorax or legs divided by the total number of bloodfed mosquitoes . The MEB rate is the number of mosquitoes without virus in the head/thorax or legs divided by the total number of midgut infected mosquitoes , and MIB is the number of mosquitoes without virus in the midgut divided by the total number of bloodfed mosquitoes . Proportions of infected mosquitoes were compared among collection sites ( villages ) and among viral isolates by calculating Bayesian 95% Highest Density Intervals ( 95% HDI ) using WinBUGS 1 . 4 [38] and an analysis of contingency tables script ( Box 6 . 13 in [39] ) . Mean virus titers were compared among isolates at each of four locations using a one way ANOVA script ( Box 6 . 1 in [39] ) run with WinBUGS and comparing 95% HDI among isolates . We used two-way ANOVA in R [model = summary ( aov ( Midgut∼Virus*Village ) ) ] to test for significant virus by village interactions . Correlation analyses were also performed in R [cor . test ( BloodmealTiter , ProportionDisseminated ) ] .
Vector competence varied among collections and viral isolates ( Figure 1 , Table 3 ) . The proportion and distribution of mosquitoes with a DI , MEB , and MIB infected with YFV BA-55 , YFV DAK1279 , or DENV-2-75505 varied across Senegal ( Figure 1 ) . Collection sites in the forested area of southeast Senegal were more refractory to YFV BA-55 than the collection sites in western Senegal with the exception of collections in urban sites close to or within the capital of Dakar . Other collections from western Senegal ( Richard Toll , Fatick , and Bignona ) had appreciable , albeit variable , DI rates due to a low frequency of both MIB and MEB . Mosquitoes from Fatick in particular had a 100% MI . In contrast , when infected with YFV DAK1279 the same populations of mosquitoes were highly refractory . Only five mosquitoes from Richard Toll and one mosquito from Mont Rolland developed a DI . Infection rates with DENV-2-75505 were different than both isolates of YFV ( Figure 1 ) . In contrast to Sylla et al . [22] , the collections from Kedougou and PK10 had a greater number of individuals with a DI when infected with DENV-2-75505 rather than DENV-2-JAM1409 as previously reported . We confirmed similar infection rates with DENV-2 JAM1409 to those reported in Sylla et al . [22] that used younger generation mosquitoes , indicating the high DI rates are not attributed to multiple generations in the lab ( Table S1 ) . The large number of individuals with a DI is especially noteworthy because the Kedougou and PK10 collection sites are sylvatic and 84% and 82% respectively of the females collected had no scales on the first abdominal tergite , thereby classifying them as classical Ae . aegypti formosus based on the McClelland scale [35] . Notably , all collections that were tested with DENV-2-75505 developed a disseminated infection . There was no clear association with geographic location and susceptibility . Note that the Mont Rolland and Rufisque sites in the west and Kedougou and PK10 in the southeast that were refractory to YFV BA-55 and YFV DAK1279 had a greater number of individuals with a DI when infected with DENV-2-75505 . To test for a mosquito strain by virus genotype interaction , the proportion of MIs and DIs were compared among the eight geographically distinct collection sites and three viruses . The proportion of MIs and DIs was dependent on both the virus and collection site ( Figures 2a and 2b ) , indicating a mosquito strain by virus genotype interaction . Significant differences in the proportion of MIs and DIs were seen in all the collection sites , except for Richard Toll . There was no significant difference in MI rate in mosquitoes from Goudiry infected with DENV-2 75505 or YFV BA-55 , but the DI rate was significantly different . Also , MI rates in Bignona were significantly different between YFV BA-55 and YFV DAK1279 , but DI rates were not significantly different ( Figures 2a and 2b ) . Although , the MI rate was variable among collections and viruses , the titer of virus in mosquitoes that developed a MI were similar among the eight collection sites and the three viruses ( Figure 3a ) . The average titer of virus in the midgut was distributed around 3 logs ( Figure 3a ) and ranged from 2 . 5–4 . 5 logs ( Figure 4 ) in individual mosquitoes . In contrast there was a broad distribution of viral titers in DIs ( Figure 3b ) extending from zero up to 5 . 5 logs and ranged from 1 . 5–5 . 5 logs ( Figure 4 ) . The titer of virus was dependent on the virus , the collection site , and the virus by collection site interaction in the midgut ( F-statistic = 3 . 41; P - value = 1 . 79×10−2 ) and in DI ( F-statistic = 28 . 50; P - value = 4 . 91×10−4 ) ( Figure 3 ) . The titer of virus in MIs was not correlated with the titer of virus in DIs regardless of viral isolate ( Figure 4a ) or collection site ( Figure 4b ) . Only three collection sites ( Richard Toll , Bignona , and Fatick ) had individuals that developed a DI with all three viruses ( Figure 4b ) . In general , as reported earlier [25] , [26] the efficiency of viral replication in the midgut does not affect the efficiency of viral replication in other tissues during dissemination . The mean titer of virus in MIs was correlated with the proportion of individuals within a collection site that developed DIs with YFV BA-55 ( r = 0 . 843; P = 0 . 0086 ) , but not with YFV DAK1279 or DENV-2-75505 ( Figure 5 ) . No correlation existed between midgut titer and the proportion of individuals with a DI when analyzed by collection site . This indicates that MEB and DI rates are independent of the efficiency of viral replication in the midgut , but in a virus dependent manner .
We document great variability in vector competence for both DENV-2 and YFV in collections of Ae . aegypti from across Senegal . The northwest-southeast decline in the susceptibility to YFV BA-55 is very similar to that seen with DENV-2-JAM1409 [22] . However , contrary to previous work with DENV-2 JAM-1409 , the same collections from across Senegal , including sylvatic collections , developed a DI with DENV-2-75505 , an isolate of DENV-2 from the same region . Comparison of infection rates and the titer of virus revealed a mosquito strain by virus genotype interaction during MI and DI . Although there was significant variability in MI rates , the titer of virus in the midgut was similar among viruses and mosquito collections . The titer of virus in DIs was much more variable among viruses and mosquito collections than MI titers . The efficiency of viral replication in the midgut was not correlated with the efficiency of viral replication in other tissues during DI . The proportion of individuals in each collection that developed a DI was correlated with the efficiency of YFV BA-55 viral replication in the midgut . In contrast , the proportion of individuals in each collection that developed a DI was not correlated with the efficiency of YFV DAK1279 or DENV-2-75505 viral replication in the midgut The current study builds on previous work by quantifying differences in infection rates and viral titers between two field isolates of YFV ( YFV BA-55 and YFV DAK 1279 ) and a field isolate of DENV-2 ( DENV-2-75505 ) in the same mosquito stains with known genetic diversity [22] , [23] . Previous studies on the vector competence of West African Ae . aegypti for YFV [24] , [26] suggested that West African Ae . aegypti are more refractory to YFV infection than Ae . aegypti from the Americas and Asia . Tabachnick et al . [24] showed that two Ae . aegypti collections from western Senegal when infected with the Asibi isolate of YFV , were more refractory than collections from the Americas or Asia . Although the Asibi isolate is from Ghana , it had been passaged many times and may not have been representative of isolates involved in natural transmission cycles . Similar to their western Senegal collections , collections close to Dakar were more refractory than other collections throughout the country . Miller and Mitchell [26] showed that an Ae . aegypti collection from Nigeria was much more refractory ( 10% DI ) when compared with collections from the Americas ( 90% DI ) to a YFV isolate from Peru and was completely refractory ( 0% DI ) with YFV BA-55 , the same isolate used in the present study . The current study demonstrates that some collections from Senegal also had 0% DI when infected with YFV BA-55 , but that other collections did develop DIs . Although it is difficult to draw an informative conclusion by comparing results obtained with mosquitoes from Nigeria with the results obtained here , the discrepancy in vector competence between the two populations with the same viral isolate further demonstrates specificity of the virus/mosquito interaction for vector competence . A number of studies of Ae . aegypti from Senegal have examined vector competence for DENV-2 [22] , [24] , [25] , [27] , [28] Those results differ from those presented here . When compared with Ae . aegypti from the Americas or Asia , West African Ae . aegypti were less susceptible to DENV-2 [24] , [25] . In studies directly comparing collections within Senegal [22] , [27] , [28] , there was variation in susceptibility , but the sylvatic Ae . aegypti were more refractory than domestic Ae . aegypti . However , the study by Sylla et al . [22] only examined the highly passaged DENV-2-Jam1409 isolate . Measuring vector competence in Ae . aegypti with a viral isolate collected in proximity may be the most informative approach [34] . Diallo et al . [27] , [28] did so by examining the vector competence of Ae . aegypti from Senegal with multiple local isolates of DENV-2 . Diallo et al . [28] reported that sylvatic Ae . aegypti collections have lower infection rates than other sylvatic species of Aedes , but some sylvatic Ae . aegypti mosquitos developed a DI . Diallo et al . [27] reported low levels of midgut infection ( 0 . 0–26 . 3% ) and variable disseminated infection ( 0–100% ) in six collections from Senegal regardless of geographic location . Importantly , both studies demonstrated variability in infection rates based on the isolate of DENV-2 and the collection site . The high rates of infection with DENV-2-75505 in sylvatic collections in the current study are not congruent with the low rates of infection in the previous studies [22] , [27] , [28] . An explanation for our differing results could be a result of multiple generations in the lab . Unlike Diallo et al . [27] , we were unable to get F1 mosquitoes collected from the field to survive , mate , or bloodfeed sufficiently to perform vector competence assays . High DI rates as a result of lab adaptation are unlikely because similar DI rates with DENV-2-JAM1409 confirmed that the older generation mosquito collections used in this study had similar vector competence as our younger generation mosquitoes previously reported [22] . Another explanation could be that the DENV-2-75505 isolate is more infectious in these mosquito populations than isolates used by Diallo et al . [27] , highlighting the importance of the viral isolate in vector competence assays . These populations were not screened for insect only viruses , therefore we cannot rule out the possibility they are present or if co-infection with insect only viruses is contributing to our observed differences in vector competence . The role of insect only viruses in vector competence is currently unresolved in Ae . aegypti , however Cx . pipiens infected with Culex flavivirus had a significantly higher proportion of mosquitoes develop a disseminated infection with WNV 7 days post infection ( DPI ) , but not 14 DPI [41] . Local adaptation between the virus and mosquito vector has been documented before and demonstrates the need to use a viral isolate circulating in the same geographic region when making assumptions about vector competence . Lambrechts et al . [34] demonstrated that differences in vector competence among three Ae . aegypti collections from Thailand infected with three genotypes of DENV-1 was a result of mosquito genotype by virus genotype interactions . The current study builds on this and demonstrates mosquito genotype by virus genotype interactions also occur with sylvatic Ae . aegypti and with YFV . Lambrechts et al . [34] and Bosio et al . [25] showed that the amount of virus in the midgut did not correlate with proportion of disseminated infections or the amount of virus outside the midgut . We found a slight correlation between the efficiency of viral replication in the midgut and the proportion of DIs . But this correlation was dependent on the virus . It is possible that the different viruses interact differently with the mosquito innate immune response , or that different genes in the mosquito are involved in the immune response to YFV or DENV-2 . The use of plaque assays in the current study to quantify virus may result in different results than quantitative RT-PCR used by Lambrechts et al . [34] or TCID50 [25] . An advantage of the current study is that we were able to quantify the amount of virus in the midgut and disseminated infections through the use of plaque assays while other studies have only compared infection rates . Looking at quantitative differences in viral titers creates a more complete picture about how different viruses are interacting with different mosquito collections . The lack of variation in titers in the midgut compared to more variation in other tissues observed here could provide interesting insights into vector competence in different tissues . These results could point to different genes or different mechanisms in the mosquito being involved in viral defense inside the midgut as compared with the tissues outside the midgut . The mechanisms underlying Ae . aegypti/virus interactions remain unclear , but these results suggest virus specific mechanisms . Examining more collections throughout Senegal as well as infecting these collections with a non-sylvatic isolate of DENV-2 from Senegal may provide more insight into the mechanisms . Genetic association studies with different collections , different viruses , and different tissues might provide clues as to whether different genes in Ae . aegypti are important for a generalized or specific response to flaviviruses . | Vector competence is defined as the intrinsic permissiveness of an arthropod vector for infection , dissemination , and transmission of a pathogen . The mosquito Aedes aegypti is the main vector for dengue and yellow fever viruses worldwide and is divided into two subspecies: Ae . aegypti aegypti and Ae . aegypti formosus . Aedes aegypti aegypti is found globally in tropical and subtropical regions , while Ae . aegypti formosus is mainly restricted to sub-Saharan Africa . Aedes aegypti formosus is considered to be a poor vector for both yellow fever and dengue , but some of these original studies with yellow fever were performed with highly passaged viral isolates collected at different locations than the mosquitoes . Viral genetics is an important determinant of vector competence and virus/mosquito genetic specificity exists in Ae . aegypti aegypti . We compared the vector competence of multiple collections of Ae . aegypti from throughout Senegal for both yellow fever and dengue viruses to demonstrate that vector competence in Ae . aegypti formosus is dependent on viral genotype . In contrast to earlier claims , populations of Ae . aegypti in West Africa can be competent vectors of flaviviruses . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"ecology",
"biology",
"and",
"life",
"sciences",
"population",
"biology",
"species",
"interactions",
"microbiology"
] | 2014 | Vector Competence in West African Aedes aegypti Is Flavivirus Species and Genotype Dependent |
In the bacterial world , methylation is most commonly associated with restriction-modification systems that provide a defense mechanism against invading foreign genomes . In addition , it is known that methylation plays functionally important roles , including timing of DNA replication , chromosome partitioning , DNA repair , and regulation of gene expression . However , full DNA methylome analyses are scarce due to a lack of a simple methodology for rapid and sensitive detection of common epigenetic marks ( ie N6-methyladenine ( 6 mA ) and N4-methylcytosine ( 4 mC ) ) , in these organisms . Here , we use Single-Molecule Real-Time ( SMRT ) sequencing to determine the methylomes of two related human pathogen species , Mycoplasma genitalium G-37 and Mycoplasma pneumoniae M129 , with single-base resolution . Our analysis identified two new methylation motifs not previously described in bacteria: a widespread 6 mA methylation motif common to both bacteria ( 5′-CTAT-3′ ) , as well as a more complex Type I m6A sequence motif in M . pneumoniae ( 5′-GAN7TAY-3′/3′-CTN7ATR-5′ ) . We identify the methyltransferase responsible for the common motif and suggest the one involved in M . pneumoniae only . Analysis of the distribution of methylation sites across the genome of M . pneumoniae suggests a potential role for methylation in regulating the cell cycle , as well as in regulation of gene expression . To our knowledge , this is one of the first direct methylome profiling studies with single-base resolution from a bacterial organism .
Among a few documented mechanisms , methylation of specific DNA sequences by DNA methyltransferases provides one way by which epigenetic inheritance can be orchestrated [1] . For instance , in many eukaryotes , methylated cytosine residues at 5′-CG-3′ ( CpG ) sequences are recognized by methyl-CpG binding proteins that usually repress the transcription of local DNA regions [2]–[5] . In the bacterial world , methylation is most commonly associated with restriction-modification ( R-M ) systems that provide a defense mechanism against invading foreign genomes [6] . In addition , it is known that a variety of enzymes capable of methylating DNA at adenine [7] and cytosine [8] , [9] play functionally important roles , including timing of DNA replication , chromosome partitioning , DNA repair , transposition and conjugal transfer of plasmids , and regulation of gene expression [7] , [10]–[16] . Phenomena involving inheritance of DNA methylation patterns are also known in bacteria . These systems use DNA methylation patterns to pass on information regarding the phenotypic expression state of the mother cell to the daughter cells . Methylation can alter the DNA structure and affect the binding of regulatory protein ( s ) to its DNA target site , thereby controlling gene expression [17] , [18] . Notably , most adhesion genes in Escherichia coli are regulated by DNA methylation patterns [19] , [20] . Little is known about how widespread heritable epigenetic control is in the bacterial world or the roles that epigenetic regulatory systems play in bacterial biology , including pathogenesis . For instance , it has been shown that DNA methylation in Streptococcus mutans up-regulates the expression of virulence factors like gbpC and bacteriocins [21] . It has also been shown that in E . coli , the expression of the Type IV secretion gene cluster is regulated by a non-stochastic epigenetic switch that depends on methylation of the Fur binding box [22] . In some gram-positive and gram-negative species that have been studied , adenine methylation plays a critical role in regulating chromosome replication . Adenine is generally methylated by members of the Dam family of methyltransferases , such as Dam in E . coli and DpnII in Streptococcus pneumoniae , that recognize the sequence motif 5′-GATC-3′ [23] . In these bacteria , the protein SeqA binds to hemi-methylated DNA target sites ( 5′-GATC-3′ ) clustered at the origin of replication ( oriC ) and sequesters the origin from replication initiation . SeqA also binds to hemi-methylated 5′-GATC-3′ sites in the dnaA promoter , blocking the synthesis of DnaA protein , which is necessary for replication initiation [24]–[27] . All of these events use the hemi-methylated state of newly replicated DNA as a signal . This hemi-methylated DNA is generated by semi-conservative replication of a fully methylated DNA molecule . Because of the transient nature of the hemi-methylation state , none of these phenomena are heritable . However , this mechanism is not universal , and other bacteria , like Bacillus subtilis , lack the Dam methyltransferase and SeqA proteins that E . coli employs to repress ( sequester ) its oriC during replication [28] . While there are many studies demonstrating the potential roles of methylation in epigenetic control of bacteria , the number of studies is significantly smaller than those for eukaryotes . This dearth of studies on bacterial epigenetics is partly due to a lack of a simple methodology that would allow rapid and sensitive detection of common epigenetic markers , such as N6-methyladenine ( 6 mA ) and N4-methylcytosine ( 4 mC ) , in these organisms . Through bisulfite treatment , 5-methylcytosine ( 5 mC ) was the only base modification detectable with efficiency and sensitivity suitable for genome wide epigenetic studies [29] , [30] . Recently , Single-Molecule , Real-Time ( SMRT ) sequencing was described to provide the capability of directly detecting different base modifications beyond the canonical A , C , G , and T bases , in addition to yielding the sequence information [31] . The technique has been successfully demonstrated to identify methyltransferase specificities on plasmids [32] . Here , we use SMRT sequencing to comprehensively determine the methylomes of two mycoplasma species , Mycoplasma genitalium and Mycoplasma pneumoniae , with single-base and -strand resolution . M . pneumoniae and M . genitalium are closely related human pathogens that cause atypical pneumonia and non-gonococcal urethritis , respectively [33] , [34] . These bacteria are members of the Mollicutes class characterized by the lack of a cell wall and by their reduced genomes with a low GC content . The genome sizes of M . pneumoniae and M . genitalium are 816 kb and 580 kb , respectively [35] , [36] . M . genitalium is widely considered to have the smallest genome of any bacteria that can be grown in a test tube in the absence of host cells [37] . Our analysis identified a widespread 6 mA methylation sequence motif common to both bacteria ( 5′-CTAT-3′ , with m6A in italics ) , as well as a more complex Type I m6A sequence motif in M . pneumoniae ( 5′-GAN7TAY-3′/3′-CTN7ATR-5′ ) . Analysis of the chromosome distribution pattern of the first motif in M . pneumoniae suggests that methylation is involved in regulating cell division . To our knowledge , this work is one of the first comprehensive methylome analysis of bacteria .
We analyzed the genomes of M . pneumoniae and M . genitalium for all the putative methyltransferase genes using comparative sequence analysis and our previous functional assignment [38] . In the M . pneumoniae genome , we identified different putative Type I and Type II restriction modification systems . Type I involves a complex consisting of three polypeptides: R ( restriction ) , M ( modification ) , and S ( specificity ) . The resulting complex can both cleave and methylate DNA . The S subunit determines the specificity of both restriction and methylation [39] . M . pneumoniae Type I system includes a methyltransferase ( mpn342 ) , a DNA specific recognition protein that brings the methyltransferase to the target DNA ( HdsS , mpn343 ) , and a restriction enzyme that cleaves unmethylated DNA ( HdsR , mpn345 ) . The restriction protein HdsR gene contains three frameshift mutations which likely make it inactive ( additional protein fragments could be coded by mpn346 and mpn347 ) . There are also some isolated genes encoding duplicated copies of the specificity determining subunit HdsS ( mpn089 , mpn289 , mpn290 , mpn365 , mpn507 , mpn615 , and mpn638 ) . In the Type II , methyltransferase and endonuclease are typically encoded as two separate proteins and act independently [39] . In M . pneumoniae , Type II systems could consist of the methyltransferase protein ( HsdM , mpn107 , mpn108 or mpn111 ) and the restriction enzyme ( HsdR , mpn109 or mpn110 ) . Additionally , a putative uncharacterized methyltransferase ( mte1; mpn198 ) , annotated as an EcoRI-like methylase in Uniprot and not associated with any R-M system , was identified . EcoRI restriction/modification system ( R/M ) is a Type II system that has been well characterized in vivo and in vitro [40] , [41] . M . genitalium has an orthologous of mpn198 ( mg184 ) and only one of the Type II-specificity determining subunits HdsS , mpn638 ( mg438 ) ( Table 1 ) . We looked at the transcript and protein levels for the putative genes involved in methylation systems by using information of the transcriptome [42] , [43] and proteome [44] of M . pneumoniae ( Table 1 ) . Although we could detect transcripts in the tiling array for all genes , albeit at very low level for many of them , we could identify in multiple MS experiments unique peptides for only six of them: mpn109 , mpn198 , mpn342 , mpn343 , mpn615 , and mpn638 ( Table 1 ) . Of these , mpn198 , mpn342 , mpn615 and mpn638 were found to bind DNA by doing affinity chromatography with a DNA column followed by salt elution and MS analysis ( manuscript in preparation ) . Only mpn198 ( mte1; EcoRI-like ) and mpn342 ( Type I ) are putative DNA adenine methyltransferases . Identification of methylated bases in M . pneumoniae and M . genitalium genomes was performed by SMRT sequencing at exponential ( 6 h ) and stationary phases ( 96 h ) . Figure 1A shows the results of the genome-wide base modification detection analysis for the M . pneumoniae genome in stationary phase . The inner and outer most tracks in the Circos plot are the modification values ( Qmod ) of polymerase kinetics for the reverse and forward strands of the genome relative to an unmodified WGA ( whole genome amplification ) control . Qmod is the −10log ( Pvalue ) from a t-test and described in further details in the Materials and Methods section . The plot shows many significant peaks which correspond to methylated template positions . Figure 1B shows examples of the IPD ( interpulse duration ) ratios of a representative genomic section , highlighting both the base and strand resolutions of the technique . The statistically significant peaks , which were defined as Qmod >100 ( Figure 1C; see Methods ) , were clustered as a function of sequence context to determine the recognition motifs of the methyltransferases responsible for the observed signals . The clustering results for M . pneumoniae identified >99 . 9% of all detected genomic positions as falling into two distinct sequence motifs: 5′-CTAT-3′ and 5′-GAN7TAY-3′/3′-CTN7ATR-5′ ( Y = T or C and R = A or G , with m6A in italics ) . The first motif is found in both bacteria and is methylated on only one of the two DNA strands . In the second motif , the first adenines in the plus and minus strands are methylated ( Figure 1B ) . The stretch of degenerate bases that separates the two recognition elements in the motif is characteristic of Type I methyltransferase signatures ( Figure 1C ) [45] . Despite the fact that the second sequence motif appears 1825 times per strand in M . genitalium ( Table 2 ) , there was no instance where it was detected as methylated . In contrast , this motif appears 1681 times in the genome of M . pneumoniae and 1678 are methylated ( 99 , 8% , Table 2 ) . Approximately 1–2% of the assigned peaks were secondary peaks of the primary detected m6A and treated as redundant information for the tabulation in Table 2 [31] . Analysis of two biological replicates of M . pneumoniae grown for 96 hours showed a reproducibility of 99 . 88% in the assignment of methylated positions . Putative Type II independent methyltransferases ( HsdM ) ( mpn198 , mpn107 , and mpn108 ) without an associated DNA recognition partner ( HsdS ) , considered as possible candidates for the methylation of 5′-CTAT-3′ motif , were cloned into pRSS vector and then transformed into a methyltransferase-free E . coli ER2796 ( DB24 ) [46] ( Table S9b ) following procedures described previously [32] . Mpn111 was discarded because it is a duplication of mpn108 . After cloning , the different plasmids were isolated and analyzed by SMRT sequencing . Of the three putative single proteins with methyltransferase activity , only mpn198 was capable of modifying the 5′-CTAT-3′ sequence . Interestingly , this is the only one of this group of methyltransferases that was found to be expressed by mass spectroscopy ( MS ) analyses ( Table 1 ) . As expected , no methyltransferase was identified by this approach for the Type I 5′-GAN7TAY-3′/3′- CTN7ATR-5′motif , since Type I motifs also require the DNA recognition protein HsdS [45] . These results agree with the finding that both mycoplasma species are methylated at the same motif ( 5′-CTAT-3′ ) and share a common methyltransferase , namely , mpn198 in M . pneumoniae and mg184 in M . genitalium . The fact that our MS analysis in M . pneumoniae detected protein expression only for DNA methylases MPN198 and MPN343 , together with the lack of a mpn343 ortologue and the absence of the 5′-GAN7TAY-3′/3′-CTN7ATR-5′ methylated motif in M . genitalium , suggest that MPN343 could be responsible for the methylation of the 5′-GAN7TAY-3′/3′-CTN7ATR-5′ motif . These results validated the motifs observed for M . genitalium and M . pneumoniae and identified them as the recognition sequences of previously unassigned methyltransferases . The new identified methyltransferases have been submitted in the REBASE and re-named using the standard nomenclature ( mpn198: M . MpnI , mpn342: M . MpnII , mpn343: S . MpnII , mg184: M . MgeI ) . M indicates methyltransferase; S refers to the specificity subunit for Type I system; Mpn indicates M . pneumoniae and Mge indicates M . genitalium . We next focused on M . pneumoniae to study the role of methylation in regulating gene expression and DNA replication , since the transcriptome and proteome data are currently available for it [42] , [43] . To study the putative role of methylation in DNA replication , we analyzed the density distribution of the 5′-CTAT-3′ methylation motifs in a sliding window of 1 kb along the M . pneumoniae genome ( Figure 2A ) . The mean number of 5′-CTAT-3′motifs per 1 kb window is two ( ±1 . 6 standard deviation ) . Regions with more than five 5′-CTAT-3′ motifs ( Pvalue<0 . 01 ) were considered to be “hot spots of methylation” for 5′-CTAT-3′ ( Table S2b ) . A functional enrichment analysis of all the genes in M . pneumoniae present at the 5′-CTAT-3′ hotspots showed two functional categories of clusters of orthologous groups ( COGs ) over-represented: defense mechanisms ( Pvalue = 0 . 025 ) and genes coding for membrane proteins or lipoproteins ( Pvalue = 9×10−4 ) ( Table S4a ) . Of the hot spots , there are three regions that have more than 10 motifs/kb . Interestingly , these regions are symmetrically distributed around the first kb of the genome ( Figure 2B ) . This region of the genome comprises an intergenic region of 687 bp with three non-coding RNAs ( MPNs200 , MPNs201 , and MPNs381 ) that frame eight repetitive 5′-TATTA-3′ sequences ( identified as DnaA boxes based on Chip-seq analysis; Yus et al manuscript in preparation; Figure 2C [47] ) . There are three 5′-CTAT-3′ methylation motifs , two of them in overlapping and opposite strands of the region with the putative DnaA boxes suggesting that DNA methylation , although different from E . coli , could play a role in DNA replication . The other two regions are located at approximately 105 kb to the left and right from the putative origin of replication ( Figure 2B ) . Search of common motifs in these two methylation hot spots revealed a common motif of 14 bp ( 5′-GATAG/ACCAAGG/AAGC-3′ ) ( Figure 2D ) . This motif is found at opposite strands in the two regions , but only the left side region contains the 5′-CTAT-3′ sequence overlapping . We also analyzed the genome-wide distribution of the Type I motif . The average distribution for the 5′-GAN7TAY-3′/3′- CTN7ATR-5′motif in 1 kb is 1 motif/kb ( ±1 . 1 standard deviation ) , and hot spot regions were considered to be those with more than 3 motifs within 1 kb ( Pvalue<0 . 01 ) . Most of the genes that overlap with these hotspots are of unknown function with a Pvalue of 0 . 04 ( Table S4b ) . There are four 1 kb regions in the genome that have more than five instances of 5′-GAN7TAY-3′/3′-CTN7ATR-5′methylation ( Figure 3A , and Table S2a ) . Interestingly , this highly methylated region with the most motifs ( 6 in 582 bp ) , is within mpn140 , the first gene of the cytadherence operon that contains one of the main virulence factors of M . pneumoniae ( Figure 3B ) . These motifs are located just upstream of the transcriptional start site ( TSS ) of an antisense transcript ( MPNs383 ) that could be involved in regulating the expression of mpn140 ( Figure 3C ) . The other three enriched regions correspond to mpn684 ( that encodes a conserved hypothetical protein ) , mpn357 ( DNA ligase ) , and mpn358 ( conserved hypothetical protein ) and , surprisingly , to the region containing mpn342 ( M . MpnII ) and mpn343 ( S . MpnII ) . As mentioned above , M . MpnII is the putative methyltransferase responsible for 5′-GAN7TAY-3′/3′-CTN7ATR-5′ methylation . The genome-wide access to methylation information allows for the interrogation of genomic locations which match the methyltransferases sequence targets , but are kept in an unmethylated state by the bacterium . The results in Table 3 show 5′-CTAT-3′ and 5′-GAN7TAY-3′/3′-CTN7ATR-5′ sites that are always unmethylated , two examples are shown in Figure 4 . Only one unmethylated 5′-CTAT-3′ site was identified ( genome position: 466475 ) . This motif is overlapping with the stop codon of the mpn390 gene that codifies for the dihydrolipoamide dehydrogenase ( PdhD ) . This gene together with mpn391 ( PdhC , dihydrolipoamide acetyltransferase ) constitute an operon involved in pyruvate metabolism . Also , three 5′-GAN7TAY-3′/3′-CTN7ATR-5′ unmethylated sites were detected . One is located in an intergenic region and the other two sites are located inside mpn493 ( UlaD , 3-keto-L-gulonate-6-phosphate decarboxylase ) involved in ascorbate and aldarate metabolism and mpn503 ( cytadherence protein ) ( Table 3 ) . We hypothesize that these unmethylated sites indicate the presence of an interacting protein or a DNA structure that is protecting from methylation along the different phases of growth . Recent identification of TSSs in M . pneumoniae [42] allowed us to study methylation patterns in promoter regions . We analyzed the regions comprising 40 bp upstream from the TSS ( e . g . the promoter region ) for 663 transcripts with TSS assigned and found 197 that were methylated in the promoter region ( Table S5 ) , with a total of 162 5′-CTAT-3′ and 74 5′-GAN7TAY-3′/3′-CTN7ATR-5′ motifs ( located on both strands at the context site ) . Of these 197 transcripts , 103 are for non-coding RNAs ( MPNs ) and 89 correspond to ORFs . Fisher's exact test shows that there is a strong enrichment in methylation of MPNs promoters , with a Pvalue of 8 . 98×10−11 . No functional enrichment is found for genes or MPNs ( considering coding genes that overlap ) methylated at the promoter regions ( Table S6a ) . Figure 5 shows the distribution in promoter regions of the distances from the methylation site ( located upstream ) to the TSS . Both motifs show that the highest frequency of methylation is at positions near the TSS and the Pribnow box ( ∼10–12 bases ) ( Pvalue of 0 . 03 for the 5′-CTAT-3′ motif , and of 0 . 005 for the Type I motif ) . These results could suggest the methylation has a potential role in transcription by affecting interaction of the sigma70 , or of specific transcription factors , with the promoter . We have also investigated the methylation pattern of 5′UTR regions encompassing the DNA sequences between the TSS and the translational start codon longer than 40 bp ( long 5′UTR ) . Ninety two of 154 ORFs that have long 5′UTR regions showed methylation ( Table S7 ) . COG analysis of genes showing methylation in long 5′UTRs ( Table S6b ) revealed that genes involved in defense mechanism were three times more represented , with a Pvalue of 0 . 02 . Interestingly , mpn342 gene ( M . MpnII ) has a 56 bp 5′UTR with two5′-GAN7TAY-3′/3′-CTN7ATR-5′ motifs , with 11 bp distance between the TSS and the motifs . As mentioned above , this gene could be responsible for methylating the 5′-GAN7TAY-3′/3′-CTN7ATR-5′motif , suggesting an autoregulatory gene expression mechanism . Although the majority of the 5′-CTAT-3′ sites were methylated in both exponential ( 6 h ) and stationary ( 96 h ) phases , using the conservative Qmod threshold of 100 , a few sites were identified as having significantly different Qmod values which would suggest a change in methylation fraction at the given sites . Figure 6 illustrates the decrease in the 5′CTAT-3′ Qmod distributions from stationary to exponential growth samples , while the 5′-GAN7TAY-3′/3′-CTN7ATR-5′ Qmod distributions remain unchanged . This drop in the Qmod values points to a potential decrease in the methylation fraction at some 5′-CTAT-3′ sites at exponential growth as compared to stationary phase . To address this question of methylation changes at any given 5′-CTAT-3′ site between the growth phases at 6 h vs 96 h , we performed a direct comparison analysis between M . pneumoniae 6 h and 96 h . From this analysis , there are 35 5′-CTAT-3′ sites that were unmethylated at 6 h but became methylated by 96 h ( Qmod≥60 ) , indicating a change in methylation status between exponential and stationary phases of growth ( Table S3 ) . Twenty-five of the 35 methylation motifs are inside genes coding for membrane proteins , one in a 5′UTR , and the rest in intergenic regions . Analyzing the transcriptome for these 25 genes at 6 h and 96 h showed that their expression levels did not significantly change ( Table S3 ) , suggesting that this change in methylation state inside the genes is not related to the regulation of gene expression at different phases of growth . It was also observed that the fraction of methylation increased from 6 h to 96 h but not vice versa , further suggesting that the methylation in these regions are dependent on the phase of growth . It is noteworthy that M . MpnI reaches its maximal level of expression at exponential growth [39] . No general increase or decrease in gene expression was found associated with methylation . However , some specific cases , such as MPNs111 , displayed an increase in promoter methylation with a significant decrease in transcript levels ( fold change log2 = 2 . 93 ) ( Table S8 ) .
In conclusion , using SMRT DNA sequencing , we were able to directly observe and analyze with single-base and strand resolution the genome-wide methylomes of M . genitalium and M . pneumoniae . The two strains share an analogous methlytransferase that targets the sequence 5′-CTAT-3 . M . pneumoniae additionally has a Type I methyltransferase with a 5′-GAN7TAY-3′/5′-CTN7ATR-3′ specificity . Together , these 2 motifs correspond to more than 99 . 9% of all sites directly detected by SMRT sequencing as modified . While ongoing work involving methyltransferase knock-out and over-expression studies are underway to help establish the relationship , this work demonstrates the unique capability of SMRT sequencing to directly sequence and profile the methylome of a whole microbial genome , allowing for unprecedented progress towards understanding the role of epigenomics in the world of prokaryotes .
Escherichia coli TOP 10 strain ( Invitrogen ) and E . coli ER2796 ( DB24 ) [46] deficient in methyltransferases , also called DB24 ( New England Biolabs ) , were grown at 37°C in LB broth or LB agar plates containing 100 µgml−1 ampicillin . The M . genitalium G-37 WT and M . pneumoniae M129 strains were grown in SP-4 and Hayflick media , respectively [62] at 37°C under 5% CO2 in tissue culture flasks ( TPP ) . Cells were grown for 96 h for the stationary phase of growth . Alternatively , after 96 h of growth , the media was removed and replaced by fresh media , and the cells were scraped and re-grown for 6 h ( exponential phase of growth ) . Genomic DNA of M . genitalium and M . pneumoniae was isolated using the Illustra bacteria genomic Prep Mini Spin Kit ( GE Healthcare ) . Plasmid DNA was obtained using the QIAprep Spin Miniprep Kit ( Qiagen ) . All primers and plasmids used in this work are summarized in Table S9a and S9b . PCR products and digested fragments from agarose gels were purified using the QIAquick PCR purification Kit ( Qiagen ) . Genomic and plasmid samples of M . genitalium and M . pneumoniae were prepared for SMRT sequencing following standard SMRTbell template preparation protocols for base modification detection on the PacBio RS [63] . In brief , each genomic sample was used to construct two SMRTbell template libraries: a ∼500 bp randomly sheared insert library of native genomic DNA , and a whole-genome-amplified ( WGA ) library of the same insert size to remove any existing base modifications in the genomic DNA . The WGA sample served as a control . SMRT sequencing was performed using C2 chemistry . At 2–4 SMRTCells each , all samples achieved ∼500× average sequencing coverage across the genome . The principle of base modification detection using SMRT sequencing by synthesis was detailed in previous publications [31] , [32] . The technique relies on the sensitivity of the polymerase kinetics to the DNA template structure as DNA synthesis is recorded in real time . It was observed that the time between base incorporations , or interpulse duration ( IPD ) , is on average longer when the nucleotide incorporation occurs opposite of a methylated base in the DNA template , as compared to an incorporation opposite of a canonical base . In previous studies , the analysis involved computing the ratio of the mean IPD of the native sample to the mean IPD of the WGA control sample for every reference template position , and setting a threshold to call certain template positions as methylated . The data analysis implemented here uses a t-test with a log-normal distribution model for the IPDs and associated Pvalue at every position for identifying the methylated sites . The null hypothesis in this analysis is that the IPDs from the native and WGA samples are part of the same population , and the alternate hypothesis is that the native set of IPDs stems from a population with larger IPDs , namely from incorporations opposite of a methylated rather than canonical template base . A threshold value of 100 for the log-transformed Pvalue from the t-test ( called Qmod = −10log ( Pvalue ) ) at each reference position was used for assigning the given position as methylated . The value of 100 was chosen based on the Qmod distribution observed in the data , where there was a clear bimodal distribution arising from unmodified background and modified positions . Furthermore , a Qmod≥100 corresponds to better than the Bonferroni corrected Pvalue of 0 . 0001 for the 816 kb genome . To detect relative changes of the methylation status between samples grown for different time periods , the two native samples were directly compared against each other , rather than against a WGA control sample , thus highlighting the methylome difference between those samples . This analysis is performed after whole methylome analysis of the genome of interest . Hence , all sites of the discovered motifs were used as the n independent test sites giving a Bonferroni corrected Pvalue of better than 0 . 01 ( 0 . 0067 ) at Qmod≥60 . Plots were made using Circos [64] . Both modes of analysis were carried out using SMRT Portal ( http://www . smrtcommunity . com/SMRT-Analysis/Software/SMRT-Portal ) , while sequence motif cluster analysis was done using Pacific Biosciences's Motif Finder ( http://www . smrtcommunity . com/CodeShare_Project ? id=a1q70000000GtatAAC ) . Data sets containing kinetic values for each reference position and DNA strand are available at http://www . pacbiodevnet . com/Share/Datasets/Senar-et-al . M . pneumoniae mpn107 gene was obtained by PCR using genomic DNA as template and specific primers ( Table S9a ) . 5′-end oligonucleotides incorporated a PstI site followed by the sequence 5′-TTAAGG-3′ ( to terminate translation of the lac α-peptide reading frame of the pRSS plasmid vector and to reinitiate translation of the cloned methyltransferse ( MTase ) genes , followed by an eight nucleotide spacer sequence 5′-TTAATCAT-3′ and sequences complementary to the 5′-end of the relevant MTase coding sequence . 3′-end oligonucleotides were complementary to the 3′-end of the MTase coding sequences , including translation termination codons and a BamHI restriction site . Since the TGA codon encodes tryptophan in Mycoplasma but an opal stop codon in E . coli , the mpn198 and mpn108 genes having several opal codons were codon-transformed and synthesized by GeneScript . After PCR amplification , the different genes were cloned into a PstI-BamHI digested pRSS vector . The resulting vectors were termed pRSS107 , pRSS198 , and pRSS108 ( Table S9b ) . The vectors described above were used to transform the E . coli deficient in methyltransferases ER2796 strain ( kindly provided by R . Roberts , NEB ) . The plasmid DNA of every transformed strain was analyzed by SMRT sequencing as described previously [32] . Transcriptional start sites of the M . pneumoniae transcriptome have been described recently [42] . This information was used to define the 5′-UTR ( RNA sequences from transcriptional start site to translational start codon ) . Transcription levels of M . pneumoniae genes at 6 h and 96 h were previously determined by tiling and ultrasequencing [43] . These data were used to study the relation between methylation and transcription in M . pneumoniae ( Table S1 ) . | DNA methylation in bacteria plays important roles in cell division , DNA repair , regulation of gene expression , and pathogenesis . Here , we use a novel sequencing technique , Single-Molecule Real-Time ( SMRT ) sequencing , to determine the methylomes of two related human pathogen species , Mycoplasma genitalium G-37 and Mycoplasma pneumoniae M129 . Our analysis identified two novel methylation motifs , one of them present uniquely in M . pneumoniae and the other common to both bacteria . We also identify the methyltransferase responsible for the common methylation motif and suggest the one associated with the M . pneumoniae unique motif . Functional analysis of the data suggests a potential role for methylation in regulating the cell cycle of M . pneumoniae , as well as in regulation of gene expression . To our knowledge , this is one of the first genome-wide approaches to study the biological role of methylation in a bacterial organism . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microbiology",
"dna",
"transcription",
"genome",
"sequencing",
"epigenetics",
"chromosome",
"biology",
"gene",
"expression",
"microbial",
"pathogens",
"comparative",
"genomics",
"biology",
"dna",
"modification",
"systems",
"biology",
"genetics",
"genomics",
"genetics",
"and",
"genomics"
] | 2013 | Comprehensive Methylome Characterization of Mycoplasma genitalium and Mycoplasma pneumoniae at Single-Base Resolution |
Pathogenic mechanisms of Candida glabrata in oral candidiasis , especially because of its inability to form hyphae , are understudied . Since both Candida albicans and C . glabrata are frequently co-isolated in oropharyngeal candidiasis ( OPC ) , we examined their co-adhesion in vitro and observed adhesion of C . glabrata only to C . albicans hyphae microscopically . Mice were infected sublingually with C . albicans or C . glabrata individually , or with both species concurrently , to study their ability to cause OPC . Infection with C . glabrata alone resulted in negligible infection of tongues; however , colonization by C . glabrata was increased by co-infection or a pre-established infection with C . albicans . Furthermore , C . glabrata required C . albicans for colonization of tongues , since decreasing C . albicans burden with fluconazole also reduced C . glabrata . C . albicans hyphal wall adhesins Als1 and Als3 were important for in vitro adhesion of C . glabrata and to establish OPC . C . glabrata cell wall protein coding genes EPA8 , EPA19 , AWP2 , AWP7 , and CAGL0F00181 were implicated in mediating adhesion to C . albicans hyphae and remarkably , their expression was induced by incubation with germinated C . albicans . Thus , we found a near essential requirement for the presence of C . albicans for both initial colonization and establishment of OPC infection by C . glabrata .
Oropharyngeal candidiasis ( OPC ) is an opportunistic mucosal infection caused by Candida species [1 , 2] . Candida albicans and Candida glabrata are the first and second major etiological agents of OPC , respectively [3] . Although other Candida species , including C . parapsilosis , C . tropicalis , and C . krusei , may be isolated as the sole species from oral infection sites , single species infection by C . glabrata alone is rare [4 , 5] . C . glabrata is most frequently co-isolated along with C . albicans in mixed species oral infections [4 , 6 , 7] . Oral infections involving C . glabrata have increased by 17% over the past several years [7] , and are particularly common in cancer patients , denture-wearers , or following prolonged use of broad spectrum antibiotics , steriods or following head and neck radiation therapy [3] . These infections were often associated with multiple Candida species [3 , 4] . Oral infections with mixed C . albicans and C . glabrata were found to be more severe and difficult to treat [5] since many C . glabrata strains are innately resistant to azole antifungal agents used in treating mucosal infections . Prophylactic use of azole antifungal drugs has been implicated as a major cause for the increase in non-C . albicans fungemia [8] . Fungemia caused by C . glabrata has high mortality especially in adult patients in intensive care units [9] , and although fluconazole prophylaxis has reduced the incidence of invasive candidiasis in high-risk neonates and immunosuppressed patients , there has been little effect on the overall incidence of C . glabrata candidiasis . Given the frequency of C . glabrata and C . albicans co-infection , it is imperative to understand the mechanisms deployed by C . glabrata in co- infections with C . albicans . C . albicans is a diploid , polymorphic fungus that exists in yeast , hyphal , and psuedohyphal forms [10] . C . albicans hyphae express numerous proteins that enhance virulence by adhering to host cells or damaging host tissue [11] . C . albicans hyphae are known to penetrate epithelial surfaces , damage endothelial cells , and aid in systemic infection by colonizing different organs such as kidneys , spleen and brain [10 , 12] . Als ( Agglutinin Like Sequence proteins ) , Hwp1p ( Hyphal wall protein ) , and Eap1 ( Enhanced Adherence to Polystyrene ) are well-characterized C . albicans hyphal wall adhesins that mediate C . albicans interaction with host epithelial , endothelial and host tissue proteins [13–15] . C . albicans adhesins contribute not only to its ability to adhere and colonize multiple types of host tissues , but also serve as binding moieties for other microbes such as Streptococcus gordonii , Pseudomonas aeruginosa , and Staphylococcus aureus [16–19] . It is therefore possible that one or more C . albicans hyphal-specific adhesins may play a role in C . glabrata interaction as well . In terms of host tissue invasion , C . albicans has a fitness advantage over C . glabrata in terms of its ability to switch between yeast to hyphal forms . By contrast , C . glabrata virulence must be independent of its morphology , since it lacks the ability to form true hyphae . However , C . glabrata is likely to express specific adhesins in order to establish colonization [20 , 21] . Phylogenetic analysis of the C . glabrata genome showed 66 putative cell wall proteins , of which only a few have been well characterized in terms of host cell adhesion [13] . Cell wall protein families known to be involved in adhesion to endothelial and epithelial cells include the EPA ( Epithelial cell adhesin ) , AED ( Adherence to endothelial cells ) , and PWP ( PA14 domain containing Wall Protein ) proteins [13] . C . glabrata Epa1 , 6 , and 7 adhesins bind to both endothelial and epithelial host cells [22 , 23] , while Pwp7p and Aed1p are known to interact with endothelial cells [13] . Deletion of these Epa1 adhesins attenuated virulence in a murine model of disseminated candidiasis [22 , 23] . The role of C . glabrata adhesins , beyond their ability to mediate adherence to host tissues , is understudied . We hypothesize that one or more of these adhesins may promote interspecies interaction with C . albicans during mixed species OPC . Co-adhesion is the basis for both single and multispecies colonization in the host [24] . Co-adhesion in bacteria is well studied and it has been established that the expression of multiple bacterial adhesins drive interspecies oral bacterial colonization [24 , 25] . Although mixed infections of C . glabrata and C . albicans occur frequently , the mechanism of co-adhesion and interspecies colonization is not well understood [4] . In our study , in spite of C . glabrata encoding several cell wall adhesins known to bind host epithelial and endothelial cells , we documented poor colonization in our murine OPC model . We hypothesized that co- infection or prior infection with C . albicans may facilitate C . glabrata infection . Here we characterize the co-colonization of C . glabrata and C . albicans in a murine model of OPC , and explore the role of cell wall proteins from both species in mediating cell-cell interaction and co-colonization .
We initially performed an in vitro biofilm assays to test whether C . albicans and C . glabrata have any cooperative growth effects . Two strains of C . glabrata ( BG2 wild type , WT ) and a GFP-expressing strain CgVSY55 ( ura3Δ::hph ScPGKp-yEGFP-URA3-CEN-ARS ) derived from a CgDSY562 WT [26] and two strains of C . albicans ( CAI4 WT with URA replaced , URA+ ) or CAF2-yCherry strain [27] were used in biofilm experiments . In a static plate assay , C . albicans CAI4 and CgBG2 each formed single species biofilms with similar robustness . However , when grown together as a dual species biofilm , the total dry weight was significantly ( P<0 . 001 ) higher compared to single species ( Fig 1A ) . Fluorescent quantitation of co-culture of C . albicans CAF2-yCherry with C . glabrata CgVSY55 under static biofilm growth showed enhanced growth of both species occurred compared with single species ( Fig 1B ) . In contrast , under dynamic flow conditions , C . glabrata ( CgVSY55 ) alone was unable to form biofilms within the flow chamber , while C . albicans formed abundant biofilms . However , when both species were co-cultured under dynamic flow conditions , C . glabrata CgVSY55 cells ( green ) were found associated with C . albicans ( red ) nascent biofilm regions , and were concentrated along C . albicans hyphae ( Fig 1C arrows ) . To further examine how these two Candida species might be interacting , we examined their association directly by fluorescence microscopy . C . albicans cells were grown in YNB + 1 . 25% glucose ( for yeast phase cells ) or in YNB + 1 . 25% N-acetyl glucosamine at 37°C ( to induce hyphal cells ) for 3 h . C . albicans cells were then incubated with C . glabrata cells at 1:1 ratio for 60 min . C . glabrata cells did not adhere with C . albicans yeast cells ( Fig 2A , upper left ) ; however they showed strong adhesion along the length of germinated C . albicans hyphae ( Fig 2A left ) . Scanning Electron Microscopy ( SEM ) further illustrated this interaction showing that C . glabrata cells adhered along the entire length of C . albicans hyphae ( Fig 2A right ) . We observed that C . glabrata cells formed rows of adherent cells along the length of hyphae , but did not adhere to other C . glabrata cells . Next , we quantified adhesion as defined by the number of C . glabrata cells adhering to 10 μm length of C . albicans hyphae in seven different strains of C . glabrata . Among the C . glabrata strains examined , CgDSY56 ( the parent strain of CgVSY55 ) had significantly ( P<0 . 0001 ) higher adherence ( 6 . 4 ± 0 . 2 cells / 10 μm hyphae , high adherence strain ) when compared to other strains tested . CgBG2 and Cg960032 ( 4 . 2 ± 0 . 2 cells / 10 μm hyphae ) showed medium adherence; and Cg931010 , Cg932474 , Cg148042 , and Cg90030 showed low adherence ( 3 . 0 ± 0 . 2cells / 10 μm hyphae ) ( Fig 2B ) . Yeast to hyphae transition in C . albicans induces expression of hyphal-specific proteins as well as altering mannans and glucans levels in the hyphal cell wall [28] . To identify whether binding between C . albicans and C . glabrata was mediated by cell wall carbohydrates , we performed blocking experiments with C . albicans using concanavalin A ( which binds cell wall mannans ) and an antibody to β , 1–3 glucan at concentrations that we previously showed provided good cell coverage [29] . C . albicans hyphae were treated with concanavalin A or with β , 1–3 glucan Ab for 30 min , washed , then incubated with C . glabrata; however C . glabrata adhesion to C . albicans hyphae was unchanged , suggesting that C . albicans adhesion is not mediated by binding to C . albicans mannose or β , 1–3 glucan . This is consistent with the fact that we did not detect C . glabrata binding to other C . glabrata cells since the C . glabrata cell wall contains both mannan and β , 1–3 glucan . We hypothesized that C . glabrata might bind C . albicans cell wall proteins directly . To test candidate C . albicans hyphal wall adhesins required for C . glabrata adhesion , we performed co-adhesion assays with ALS1 and ALS3 deficient C . albicans ( Fig 3 ) . We found that C . glabrata had significantly decreased adherence to hyphae of a C . albicans als3Δ/Δ mutant ( 72 . 3% reduction ) , an als1Δ/Δ mutant ( 28 . 8% reduction ) , and an als1/als3Δ/Δ double mutant ( 86% reduction ) . ALS1 and ALS3 complementation strains showed restoration of adherence to levels closer to that of wild type strain ( Fig 3A ) . To further validate the role of C . albicans Als1 and Als3 , we performed a quantitative adherence assay by direct microscopic observation using S . cerevisiae strains expressing C . albicans ALS1 and ALS3 with a GFP-tagged C . glabrata strain . Both Als1 and Als3 expressing S . cerevisiae strains showed significantly higher binding ( Binding Index = 52 . 0 ± 3 . 0 and 58 . 2 ± 1 . 4 , respectively ) compared to S . cerevisiae expressing an empty vector ( Binding Index = 19 . 7 ± 2 . 3 ) ( Fig 3B ) . To determine whether our observed binding between C . albicans and C . glabrata has relevance in vivo , we examined the ability of C . glabrata to establish infection in our murine model of OPC . Since C . glabrata has not been used before in OPC infection models , we began with a single species oral infection of C57BL/6 mice with C . glabrata alone as we have previously done with C . albicans ( Fig 4A ) . In this model , sublingual infection with a C . albicans inoculum of 1 X 106 cells/ml typically produces clinical symptoms and white tongue plaques 4–5 days post infection , and recovery of 1 X 107 CFU / gm tongue tissue at 5 days post infection . Surprisingly , in no infection experiments using C . glabrata did we observe the typical appearance of white tongue plaques indicative of clinical infection . We tried varying immunosuppressive agents ( triampicinolone acetonide , cyclophosphamide ) , mouse strains ( BALB/c , IL17RAk/o ) and used inocula size of C . glabrata ranging from 1 X 107 to 1 X 1010 cells/ml . In all cases , infection with C . glabrata alone resulted in no clinical appearance of disease or weight loss in animals . Consistent with this lack of disease , the recoverable C . glabrata CFUs from the tongue were extremely low ( 4–7 X 102 CFU/g of tongue tissue ) . Since our in vitro biofilm and adhesion assays showed enhanced adhesion and growth of C . glabrata when mixed with C . albicans , we next attempted a mixed infection with C . glabrata ( 1X109 cells/ml ) either as a co-infection with C . albicans ( 5 X 107 cells/ml ) ; or as a delayed infection with C . glabrata 24 or 48 h after infection of C . albicans ( Fig 4 ) . C . glabrata CFU were significantly ( P<0 . 02 ) increased by ten-fold ( 3 X 103 CFU/g of tongue tissue ) when mice were co-infected with C . albicans ( Fig 4A ) . Co-infection with C . glabrata did not alter C . albicans infection levels ( 1 . 2 X 107 CFU/g of tongue tissue ) compared with C . albicans infection alone ( Fig 4A ) . However , delaying C . glabrata infection for 24 or 48 h after establishment of C . albicans infection further increased C . glabrata oral infection by a further 10-fold ( 3 . 5–4 . 5 X1 04 CFU/g of tongue tissue , P<0 . 0001 ) compared to C . glabrata single species infection ( Fig 4B ) . Mean animal weights did not change upon C . glabrata infection only ( Fig 4C ) . However , mice lost weight more rapidly following a mixed infection compared with infection by C . albicans alone , so that mice in the mixed infection group had to be sacrified one day sooner due to total weight loss compared with mice infected with C . albicans only ( Fig 4C ) . Thus our data show that levels of oral infection of C . glabrata were signifcantly increased by an established C . albicans oral infection and the rate of weight loss was increased upon dual species infection . Next , we examined tongues of mixed-infected mice histologically to determine whether C . glabrata alters C . albicans invasive properties and to identify the localization of C . glabrata infection within the mucosal epithelium . For these experiments we infected mice with fluorescent-tagged strains of C . albicans ( CAF2-yCherry ) on day 0 and C . glabrata ( CgVSY55 ) on day 2; and collected tongue tissues on day 5 . Tongues were sectioned and stained with either PAS to visualize fungal-tissue architecture or cryo-sectioned for visualization of yeast cell localization by fluorescence microscopy . Tongues from mice with mixed infection showed robust fungal plaque formation as well as extensive C . albicans hyphal penetration of the superficial epithelium ( Fig 5A , boxed region ) as well as invasion into some regions of the underlying epithelium and lamina propria ( Fig 5A , arrows ) . Closer inspection of these regions showed widespread C . albicans hyphae; and in some areas yeast cells were observed both adherent to hyphae and as unattached cells that were likely to be C . glabrata ( Fig 5B and 5C , arrows ) . Fluorescent imaging of these regions confirmed that the majority of tissue invasion was with C . albicans hyphae ( Fig 5D , boxed region , red ) , however C . glabrata cells ( green ) were also observed within these tissues both associated with C . albicans hyphae as well as being unconnected and separate within the epithelium ( Fig 5E and 5F , arrows ) . In contrast , mono-species C . glabrata infection resulted on only very small superficial plaques that were localized on the surface mucosa without any invasion . Thus , infection of oral epithelium with C . albicans and the presence of its hyphae were permissive for infection and tissue invasion by C . glabrata . To further confirm the requirement of C . albicans for C . glabrata for initial infection , we treated mice with fluconazole ( Flu ) after establishing mixed infection using Flu sensitive ( CaFluS ) or Flu resistant C . albicans ( CaFluR ) strains and Flu resistant C . glabrata ( CgFluR ) ( Fig 6 ) . Mice were treated with Flu for four days after an oral mixed infection was already established for four days . As expected , Flu treatment did not alter infection levels of either species in a mixed infection with CgFluR and CaFluR strains . However , for a mixed infection with C . glabrata CgFluR and C . albicans CaFluS strains , Flu treatment resulted in significant ( by two logs , P<0 . 001 ) reduction of both C . glabrata and C . albicans . Flu treated animals infected with CaFluR strains in a mixed infection lost significantly ( P<0 . 05 ) more weight ( 21 . 2 ± 0 . 2% ) than mice infected with C . albicans CaFluS strains ( 18 . 9 ± 0 . 3% ) . Although we could not determine the co-locallization of C . albicans and C . glabrata histologically due to lack of fluorescent markers in Flu resistant strains , examination of tongues confirmed the reduction in superficial epithelial fungal burden and invasion upon Flu treatment ( Fig 6 ) . Thus , C . glabrata infection levels were proportional to those of C . albicans , showing that C . glabrata requires the presence of C . albicans for early infection in vivo . Since our in vitro data showed that C . albicans Als1 and Als3 adhesins were important for C . glabrata adherence , we next examined their role in mixed C . glabrata-C . albicans oral infection in vivo . A 48 h delayed infection of C . glabrata following infection with C . albicans wild type or Als adhesin deficient strains was performed ( Fig 7 ) . C . albicans als1Δ/Δ and als3Δ/Δ mutants were able to establish infection at the same levels as WT cells . However , C . glabrata tongue CFUs were significantly ( P<0 . 05 ) decreased ( 2 . 8 X 104 CFU/g ) following infection with the C . albicans als1Δ/Δ mutant; and were even further reduced ( 6 . 6 X 103 CFU/g , P< 0 . 001 ) following infection by C . albicans als3Δ/Δ . Infection of C . glabrata with C . albicans Als1 and Als3 complemented strains showed restoration of C . glabrata colonization to levels similar to those observed with the wildtype C . albicans strain ( Fig 7 ) . No differences in animal weights between the groups was found since levels of infection by C . albicans were similar between groups . To identify adhesion partners on C . glabrata , we screened 44 S . cerevisiae strains expressing C . glabrata cell wall proteins and identified five strains expressing CgEpa8 , CgEpa19 , CgAwp2 , CgAwp7 or ORF CAGL0F00181 that were most adhesive ( 2–5 cells/10 μm C . albicans hyphae ) ( Fig 8A ) . Most other tested strains , including the S . cerevisiae parental strain , had no adhesion to C . albicans hyphae . Next , we examined comparative transcription levels of these five candidate genes in C . glabrata strains which have high adherence ( CgDSY562 ) , medium adherence ( CgBG2 ) , and low adherance ( Cg90030 ) in vitro to C . albicans . We used C . glabrata EPA1 and EPA6 genes as a negative control since S . cerevisiae expressing C . glabrata Epa1 and Epa6 did not bind to C . albicans hyphae , although they are highly expressed major adhesins in C . glabrata . To confirm that these strains also had differential binding to C . albicans during infection , we compared infection levels in a mixed infection in OPC , and found that indeed , the low and high adherence strains had a significant ( P<0 . 01 ) difference in infection levels ( Fig 8B ) . Then , transcriptional levels of these candidate genes were measured by qPCR before and after incubation with germinated C . albicans . Although CgEPA8 and CgAWP7 were most highly expressed in the high adhesion strain compared to the lower adhesion strains , we did not find significant differences in basal expression levels among the three other candidate genes among the C . glabrata strains . However , transcriptional levels of four genes ( CgEPA8 , CgEPA19 , CgAWP2 , and ORF CAGL0F00181 ) were increased significantly by 6–7 fold , while CgAWP7 was increased by 2-fold in the high adherence strain ( CgDSY562 ) upon incubation with C . albicans hyphae . This induction was less for CgEPA19 , CgAWP2 , and CgCAGL0F00181 in the intermediate adherent strain , while the low adhesion Cg90030 strain had the least induction by C . albicans for all five genes ( Fig 8C ) . Expression of CgEPA6 and CgEPA1 genes , which serve as controls since they do not mediate adherence to C . albicans , were both modestly down-regulated in the presence of C . albicans . Taken together , these results show that C . glabrata cell wall genes EPA8 , EPA19 , AWP2 , AWP7 and CAGL0F0018 are upregulated by C . albicans and may promote a dual species oral infection .
Although clinical studies have shown that C . albicans and C . glabrata are common partners co-isolated from oral infections , C . glabrata alone rarely causes oral infection . This work identifies for the first time that C . glabrata adherence to C . albicans hyphae is the basis for this partnership and that it is mediated by specific adhesins on both species . Previous in vitro studies found that C . glabrata alone was unable to colonize or invade reconstituted human vaginal epithelium ( RHVE ) [30] or reconstituted human oral epithelium ( RHOE ) [31] . Mixed infections using both C . glabrata and C . albicans increased tissue damage in RHOE [32] and were permissive for infection in RHVE [33] and in vivo in tongues of immunosuppressed mice [34] , although others found no difference in host damage or inflammation in co-infected human oral epithelial [35] . These and our own studies are in agreement that C . glabrata alone is non-invasive in respect to oral-esophageal mucosal epithelium , in contrast to its ability to penetrate gastric epithelium [34] . The basis for this difference in tissue tropism is unknown , although it is possible that differences in the gut environment induces differential expression of C . glabrata adhesins . We found that two major fungal hyphal wall adhesins Als3 and Als1 contribute to binding C . glabrata in vitro and to establish oral infection in vivo . C . albicans Als3 appears to make the major contribution towards binding with C . glabrata , with Als1 having a secondary role . Consistent with this , loss of Als1 on its own does not strongly reduce adherence to C . glabrata . However , in strains deleted for ALS3 , additional loss of ALS1 further reduced adherence by an additional two-fold . Als3 is a well known multifunctional surface protein , however we have identified an additional novel function of this adhesin in binding C . glabrata . Since Als 3 proteins are very abundant on C . albicans hyphae , and we only find 2–6 C . glabrata cells per hyphae , we expect that substantial numbers of Als proteins would still be available on hyphae to carry out other functions in the context of oral infection . It is also possible that Als3 might have a similar role in binding other non-hyphal forming Candida species such as C . krusei that are frequently co-isolated along with C . albicans in OPC . C . albicans Als3 seems to be promiscuous in its binding partners since Als3 proteins have been shown to bind the oral bacteria S . gordonii through its SspB cell surface protein in a mixed species biofilm [36] and to S . aureus during polymicrobial biofilm growth [37] . Hence Als3 may to be an excellent target for disruption of mixed species and inter-kingdom biofilms . The C . glabrata Epa family consists of at least 20 GPI-anchored surface exposed adhesins whose expression of individual members is strain dependent [38] . Epa proteins recognize host glycans , and C . glabrata Epa1 is the best characterized member that is involved in adhesion to mammalian epithelium . Epa1 preferentially recognizes Gal β1–3 glycans , and variations of its adhesion domain conferred promiscuity of ligand binding [39] . Recently , Epa binding domains were functionally classified according to their ligand binding profiles , and interestingly our identified adhesins C . glabrata Epa8 and Epa19 were found to be very closely related and within the functional class III of Epa ligands that have weak binding to epithelial cells [40] . Thus , we speculate that some Class III Epa adhesins may have ligand functions with other cell types including C . albicans . Another similarlity among the C . glabrata adhesins we identified ( EPA8 , EPA19 , AWP2 , AWP7 and CAGL0F00181 ) is that their expression levels were all induced by incubation with C . albicans hyphae ( Fig 8C ) . In contrast , C . glabrata EPA1 and EPA6 ( both Class I ligands with high binding to epithelial and endothelial [41] cells , and highly expressed in log phase cells [23] ) , were not up-regulated following incubation with C . albicans hyphae . In agreement with our findings , no increase in expression levels of C . glabrata EPA1 , EPA6 or EPA7 was found following co-infection with C . albicans in RHVE cells [30] . Based on our results , we propose a role of these C . glabrata CWPs ( EPA8 , EPA19 , AWP2 , AWP7 , and CAGL0F00181 ) in interspecies binding and further suggest that C . glabrata is able to transcriptionally regulate selected genes needed for its colonization and survival in a host . It is known that many C . glabrata EPA genes are transcriptional silenced . Since EPA1 and EPA6 ( both of which are strongly silenced ) are not up-regulated by co-culture with C . albicans ( Fig 8 ) , this suggests that the transcriptional regulation of C . glabrata EPA8 , EPA19 , AWP2 , AWP7 , and CAGL0F00181 is not through general antagonism of sub-telomeric silencing [42] . How C . glabrata regulates these genes in the presence of C . albicans remains to be determined . C . glabrata alone was not competent to cause infection in our OPC model . Our data further suggest that while C . glabrata colonizes oral mucosa poorly ( even in an immunosuppressed host ) , it has evolved to exploit the presence of hyphae-producing C . albicans to establish colonization and invasion of oral epithelium; and its presence enhanced the severity of OPC as measured by rate of weight loss of animals . Furthermore , co-infections treated with Fluconazole reduced levels of C . glabrata concomitantly with C . albicans over four days , showing its dependence upon the presence of C . albicans in early infections . However , our results show that C . glabrata is found both together and apart from C . albicans hyphae in tissues , suggesting that once it gains a foothold in oral epithelium by binding C . albicans hyphae , it can survive alone in mucosal tissues , albeit at low levels . These C . glabrata cells existing independently in oral mucosa may be a colonization reservoir for dissemination if the oral epithelium is breached by trauma , chemotherapy or other factors . In this regard , our preliminary experiments showed that mice with mixed C . glabrata and C . albicans oral infections had significantly higher stomach colonization of both species , suggesting that gut colonization might serve as such a reservoir . Also , these reservoirs may become clinically significant following long-term azole therapy providing an environment in which drug resistant C . glabrata could emerge . Our data suggests a model whereby oral tissues that are inherently resistant to infection by C . glabrata , are colonized by piggybacking with C . albicans to establish a foothold of tissue infection . Of interest , and the subject of ongoing studies in our lab , is the role of oral and gut reservoirs of C . glabrata in subequent colonization of other tissues that have a naturally higher tropism for infection by C . glabrata , as well as their role in subsequent dessimination .
All Candida and S . cerevisiae strains used are listed in Table 1 and Table in S1 Table . C . albicans cells were maintained in yeast extract/peptone/dextrose ( YPD; Difco ) medium with the addition of uridine ( 50 mg/ml; Sigma ) when required and stored as -80°C . S . cerevisiae containing pADH or pADH-ALS3 were maintained on synthetic medium lacking uracil ( CSM-glu ) ( 0 . 077% CSM-ura , 0 . 67% yeast nitrogen base [Difco] , 1 . 25% glucose , and 2 . 5% agar ) . S cerevisiae strains expressing N-terminal domains of C . glabrata Cell Wall Proteins ( CWP ) were made as described [42] , and are in preparation for publication elsewhere ) . The ORFs whose domains mediate adherence to Candida hyphae are shown in S1 Table . C . albicans cells were cultured overnight in YPD broth , diluted to an OD600 = 0 . 3 in pre-warmed YNB medium supplemented with 1 . 25% GlcNAc , and incubated for 3 h at 37°C with gentle shaking to induce germination . C . glabrata or S . cerevisiae strains were grown similarly except in YNB + 1 . 25% of glucose . Cells were collected by centrifugation ( 100 X g ) , washed once in PBS , and then re-suspended in PBS . Germination of C . albicans cells was confirmed by microscopic observation . C . albicans cells were then incubated with C . glabrata cells at a 1:1 ratio for 60 min . Blocking experiments described previously [29] , were carried out using washed C . albicans cells incubated with concanavalin A ( 100 ug/ml; mannan binding lectin , Sigma ) or β , 1–3 glucan Ab ( 10 μg/ml , Biosupplies ) for 30 min ( concentrations that gave high coverage of cells as determined by FACScan ) , then washed in PBS before assay . For adhesion assays of S . cerevisiae strains expressing C . albicans Als1 and Als3 adhesins , S . cerevisiae cells or an S . cerevisiae empty vector ( control ) were incubated with CgVSY55 for 1 h at 37°C ( at cell ratio 1:1 ) , then a Binding Index was calculated as the number of C . glabrata cells bound to S . cerevisiae cells divided by ( number of bound C . glabrata cells plus unbound C . glabrata cells plus unbound S . cerevisiae cells ) X 100 per field . At least 10 separate fields were used to obtain averages . Each Candida strain was grown overnight to OD600~2 . 0 , washed twice in Phosphate Buffered Saline ( PBS ) , re-suspended in YNB without uridine , and 1 ml cells ( 1 X106 cells/ml ) were added to polystyrene wells . For mixed species biofilms , 500 μl of each species ( 5 X 105 cells /ml ) for a total of 1 ml was added to the well . After incubation for 3 h to allow adhesion , non-adherent cells were gently removed by aspiration and 1 ml of fresh media was added . Biofilms were grown for 24 h at 37°C on an orbital shaker and biofilm dry weight was measured as previously described [43] . For fluorescence biofilm assays , single and dual species biofilms were grown on 96 well microtiter plates using a yCherry expressing strain of C . albicans and a GFP expressing C . glabrata strain . Fluorescent counts were recorded at 37°C using a Bio-Tek multifunction plate reader and analyzed using Gen5 software . Alternatively , we examined non-static dual species biofilms grown under flow conditions . For these experiments , YPD media containing the C . albicans WT strain CAF2 cells expressing the fluorescent protein mCherry and the C . glabrata WT strain VSY55 expressing GFP ( both at 1 × 106 cells/ml ) were circulated through a μ-Slide I 0 . 8 Luer family ibiTreat flow chamber ( ibidi , Martinsried , Germany ) for 2 h at 37°C and a shear force at the coverslip surface of 0 . 8 dynes/cm2 . Images were obtained using a Zeiss LSM 510 confocal microscope , and analyzed using ZEN imaging software ( Zeiss , Göttingen , Germany ) . Flow was maintained during image acquisition . Overnight cultures of C . albicans were diluted to an OD600 = 0 . 3 in pre-warmed YNB medium supplemented with 1 . 25% GlcNAc and incubated for 3 h at 37°C to induce germination , or diluted in YNB medium supplemented with 1 . 25% glucose at room temperature for yeast cells . C . glabrata CgDSY562 , CgBG2 , and Cg90030 overnight cultures were grown similarly using YNB + 1 . 25% of glucose . Cells were collected by centrifugation ( 100 X g ) , and re-suspended in PBS . C . glabrata cells were then incubated with germinated or yeast form C . albicans at a 1:1 ratio for 30 min . Total RNA was isolated from C . glabrata , C . glabrata and C . albicans 1 X 107 cells using an RNeasy minikit ( Qiagen ) . Reverse transcription ( RT ) was performed using SuperScript III reverse transcriptase , and oligo ( dT ) 20 primer ( Invitrogen ) . cDNA was purified ( Geneflow PCR purification kit ) and quantified with a NanoDrop spectrophotometer ( NanoDrop Technologies ) . Quantitative RT-PCR ( qRT-PCR ) was performed in triplicate for CgACT1 , CgAWP2 , 7 , CgEPA1 , 6 , 8 , 19 and CgCAGL0F00181 using gene-specific primers ( Table in S2 Table ) . C . albicans cDNA was used as a negative control in all experiments to verify specifity of amplification . Genes were normalized to CgACT1 in each respective strain and condition as decribed previously [44] . Fluorescent microscopy was done using a yCherry expressing strain of C . albicans [45] and the GFP expressing C . glabrata ( VSY55: ura3Δ::hph ScPGKp-yEGFP-URA3-CEN-ARS ) derived from a C . glabrata DSY562 clinical isolate [46] . Scanning electron microscope observations were carried out on C . albicans hyphae and C . glabrata cells . C . albicans cells were grown in YNB for yeast phase cells or in YNB + 1 . 25% GlcNAc at 37°C ( to induce hyphae ) for 3 h . C . albicans cells were then incubated with C . glabrata cells at 1:1 ratio for 30 min . Cells were incubated on a concavalin A ( 100ug/ml; Sigma ) coated glass slide for 1 h at RT . Cells were washed twice with PBS , fixed with 2% glutaraldehyde ( Sigma ) for 30 min at 4°C , then washed twice with distilled water . Samples were dehydrated in 30% , 50% , 70% , 85% , and 95% ethanol for 15 min each and 100% ethanol twice for 15 min each . Samples were exchanged into 100% hexamethyldisilazane ( HMDS ) and allowed to dry in a hood before visualization . SEM observation was done under the following analytical condition: L = SE1 and EHT = 2 . 5 kV to study the binding of C . glabrata on C . albicans cells with Hitachi SU70 FESEM operating at 2 . 0 keV . C . albicans murine OPC model [47 , 48] was used for infection with C . glabrata . Mice ( BALB/c , C57BL/6 , and IL17RAk/o ) were immuno-suppressed with cortisone acetate ( 150–250 mg/kg ) , triampicinolone acetonide ( 100–150 mg/kg ) or cyclophosphamide ( 100–150 mg/kg ) one day before infection with C . glabrata ( 1 X 107 to 1 X 109 cells/ml ) . For mixed infections , mice ( female C57BL/6 , 4–6 weeks old ) were immunosuppressed with cortisone acetate 225 mg/kg ( Sigma ) on day -1 , +1 , and +3 , and then infected with C . albicans ( 5 X 107 cells/ml ) on day 0; or infected with C . glabrata , ( 1 X 109 cells/ml ) on day 2 after pre-establishing C . albicans infection on day 0 . C . albicans and C . glabrata colonies from tongue tissues were differentiated on CHROMagar media . On the fifth or sixth day after infection , mice were euthanized by cervical dislocation under anesthesia ( ketamine/xylazine ) ; tongue tissues were excised and hemi-sectioned along the long axis with a scalpel . One half was weighed and homogenized for quantification of fungi , and the other half was processed for histopathological analysis . Tongue hemi-sections were fixed in 10% buffered-formalin for 24 h , paraffin embedded , and then cut into 5μm sections for Periodic Acid-Schiff ( PAS ) staining as we previously described [49] . For histological co-localization experiments , animals were infected with C . albicans yCherry and the GFP expressing C . glabrata ( VSY55 ) strains as described above . For these experiments , tongue hemi sections were fixed in 4% ( w/v ) paraformaldehyde ( PFA ) for 24 h , incubated in 30% sucrose for 3 days , snap frozen in OCT compound ( Tissue-Tek , Sakura , Torrance , CA ) with liquid nitrogen , and cut into 8μm cryosections . For Fluconazole ( Flu ) treatment studies , ten mice were used for each group ( drug treatment and controls using combinations of Flu resistant and sensitive strains of C . albicans and Flu resistant C . glabrata shown in Table 1 ) . Sensitivities of each strain to Flu was verified using MIC assays . Immunosuppression was induced on days −1 , +1 and +3 post-infection . Mice were infected sublingually with C . albicans ( 5 X 107 cells/ml ) on day 0 , and C . glabrata ( 1 X 109 cells/ml ) on day 2 , and were sacrificed on day +7 . Mice received daily intraperitoneal injections of 100 mg/kg Fluconazole that was initiated 48 h after C . glabrata infection and continued through post-infection day 7 . Statistical analyses were performed using GraphPad Prism software version 5 . 0 ( GraphPad Software , San Diego , CA , USA ) using unpaired Student's t-tests . Differences of P<0 . 05 were considered significant . All experiments were performed at least thrice . This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . This protocol was approved by the University of Buffalo Institutional Animal Care and Use Committee ( Project Number: ORB06042Y ) . | Understanding how Candida glabrata is able to establish oral mucosal infections is particularly important since many C . glabrata strains are innately resistant to azole antifungal drugs used in treating mucosal and disseminated infections . The epidemiology of C . glabrata oral infections shows that C . glabrata is very often present as a co-infection with Candida albicans . Here we suggest a mechanism to explain this clinical finding . We show that C . glabrata is unable to colonize the oral mucosa in a murine oral infection model . However , prior or co-colonization by C . albicans allows C . glabrata to colonize and persist in the oral cavity . Mechanistically , we show that C . glabrata binds specifically to C . albicans hyphae , mediated by hyphally expressed ALS adhesins in C . albicans and cell surface proteins in C . glabrata that are transcriptionally up-regulated in the presence of C . abicans . In this sense , C . glabrata is a piggy-back fungus that relies upon binding to C . albicans hyphae for oral colonization . This finding has implications for treatment of oral candidiasis and may shed light on colonization mechanisms of other non-hyphae producing fungi . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"bacteriology",
"biofilms",
"cell",
"walls",
"medicine",
"and",
"health",
"sciences",
"yeast",
"infections",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"fungi",
"model",
"organisms",
"tongue",
"fungal",
"diseases",
"cellular",
"structures",
"and",
"organelles",
"fungal",
"pathogens",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"adhesins",
"saccharomyces",
"infectious",
"diseases",
"mycology",
"microbial",
"physiology",
"medical",
"microbiology",
"microbial",
"pathogens",
"biological",
"tissue",
"yeast",
"bacterial",
"physiology",
"candida",
"mouth",
"anatomy",
"cell",
"biology",
"virulence",
"factors",
"epithelium",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"saccharomyces",
"cerevisiae",
"organisms",
"candida",
"albicans"
] | 2016 | Candida glabrata Binding to Candida albicans Hyphae Enables Its Development in Oropharyngeal Candidiasis |
We present 10 tips for building effective lessons that are grounded in empirical research on pedagogy and cognitive psychology and that we have found to be practically useful in both classroom and free-range settings
There are many kinds of lessons , both formal and informal , from seconds long to lifelong . Most people have sat ( or suffered ) through hundreds of these but have never been shown how to design ones that are effective . These 10 simple tips for creating lessons are The key insight that underpins all of these tips is that learning is both a cognitive and a social activity . On the cognitive side , incoming information ( the lesson ) passes through a “sensory register” that has physically separate channels for visual and auditory information and is stored in short-term memory , where it is used to construct a “verbal model” ( sometimes also called a “linguistic model” ) and a separate “visual model” [12] . These are then integrated and stored in long-term memory as facts and relationships . If those facts and relationships are strengthened by use , they can later be recalled and applied , and we say that learning has taken place . One key feature of this model is that short-term memory is very limited: [13] famously estimated its size as 7 ± 2 items , and more recent studies place the figure closer to 4 . If too much information is presented too quickly , material spills out of short-term memory before it can be integrated and stored , and learning does not occur . A second key feature is that the brain's processing power is also very limited . Effort spent identifying key facts or reconciling the linguistic and visual input streams reduces the power available for organizing new information and connecting it to what's already present . Learning is also a social activity . Learners who feel motivated will learn more; learners who feel that they may not be judged on their merits or who have experienced unequal treatment in the past will learn less ( see the tip "Motivate and avoid demotivating" ) . In [14] , e . g . , Kenneth Wesson wrote , "If poor inner-city children consistently outscored children from wealthy suburban homes on standardized tests , is anyone naive enough to believe that we would still insist on using these tests as indicators of success ? " Lesson designers must take the social aspects of learning into account if they are to create effective lessons; we discuss this further in the final tip ( "Make lessons inclusive" ) .
The first step in creating a good lesson is figuring out who the audience is . One way to do this is to make up biographies of two or three target learners . This technique is borrowed from user interface designers , who create short profiles of typical users to help them think about their audience . These profiles are called “personas” and have five parts: A learner persona for a weekend introduction to programming aimed at college students might be as follows: Rather than writing new personas for every lesson or course , instructors often create and share a handful that cover everyone they hope to teach , then pick a few from that set to describe who particular material is intended for . Used this way , personas become a convenient shorthand for design issues: when speaking with each other , teachers can say , "Would Jorge understand why we're doing this ? " or , "What installation problems would Jorge face ? " Personas help you remember one of the most important tips of teaching: you are not your learners . The people you teach will almost always have different backgrounds , different capabilities , and different ambitions than you; personas help you keep your lessons focused on what they need rather than on what your younger self might have wanted .
Some learning strategies are provably more effective than others [15 , 16 , 17] , so lessons should be designed to encourage their use . As summarized in [18 , 19] , the six most important are as follows: Different subsystems in our brains handle and store linguistic and visual information , and if complementary information is presented through both channels , then they can reinforce one another . However , learning is more effective when the same information is not presented simultaneously in two different channels [23 , 12] because then the brain has to expend effort to check the channels against each other . This is one of the many reasons that reading slides verbatim is ineffective: not only is the reader not adding value , they are actually adding to the load on learners whose brains are trying to check that the spoken and written inputs are consistent .
“Summative assessment” is something done at the end of a lesson to tell whether the desired learning has taken place: a driving test , performance of a piece of music , a written examination , or something else of that kind . Summative assessments are usually used as gates ( e . g . , "Is it now safe for this person to drive on their own ? " ) , but they are also a good way to clarify the learning objectives for a lesson . "Understand linear regression" is hopelessly vague; a much better way to set the goal for that lesson would be to define an exercise , such as the following: This is better because it gives the lesson author a concrete goal to work toward: nothing goes in the lesson except what is needed to complete the summative assessments . This helps reduce content bloat and also tells the author when the lesson is done . Writing summative assessments early in the lesson design process also helps ensure that outcomes are actually checkable . Since telepathy is not yet widely available , it is impossible for instructors to know what learners do and don't understand . Instead , we must ask them to demonstrate that they're able to do something that they couldn't do without the desired understanding . Finally , creating summative assessments early can help authors stay connected to their learners' goals . Each summative assessment should embody an “authentic task , ” i . e . , something that an actual learner actually wants to do . Early on , authentic tasks should be learners' own goals; as they advance and are able to make sense of generalizations , these tasks may be extensions or generalizations of earlier solutions . Continuing with the statistical example above , calculating a regression coefficient may be an authentic task for someone who already knows enough statistics to understand what such coefficients are good for . If the intended learners are not yet that experienced , this exercise could be extended to have them make some sort of judgment based on the regression coefficients to exercise higher-order thinking .
The counterpoint to summative assessment is “formative assessment , ” which is checks that are used while learning is taking place to form ( or shape ) the teaching . Asking learners for questions is a common , but relatively ineffective , kind of formative assessment . What works better is to give them a short problem—one that can be done in 1–2 minutes so as not to derail the flow of the lesson and that will help them uncover and confront their misconceptions about the topic being taught . Checking in with learners this way every 10–15 minutes accomplishes several things: [24 , 25 , 26] offer inspiration for a wide variety of different kinds of summative and formative assessment exercises .
Research by Mayer and colleagues on the split-attention effect is closely related to cognitive load theory [23] . As described in the introduction , linguistic and visual input are processed by different parts of the human brain , and linguistic and visual memories are stored separately as well . This means that correlating linguistic and visual streams of information takes cognitive effort: when someone reads something while hearing it spoken aloud , their brain can't help but check that it's getting the same information on both channels . Learning is therefore more effective when information is presented simultaneously in two different channels , but when that information is complementary rather than redundant . People generally find it harder , e . g . , to learn from a video that has both narration and on-screen captions than from one that has either the narration or the captions but not both because some of their attention has to be devoted to checking that the narration and the captions agree with each other . Two notable exceptions to this are people who do not yet speak the language well and people with hearing exercises or other special needs , both of whom may find that the extra effort is a net benefit . This is why it's more effective to draw a diagram piece by piece while teaching rather than to present the whole thing at once . If parts of the diagram appear at the same time as things are being said , the two will be correlated in the learner's memory . Pointing at part of the diagram later is then more likely to trigger recall of what was being said when that part was being drawn . The split-attention effect does not mean that learners shouldn't try to reconcile multiple incoming streams of information—after all , this is something they have to do in the real world [27] . Instead , it means that instruction shouldn't require it while people are mastering unit skills; instead , using multiple sources of information simultaneously should be treated as a separate learning task .
No matter how good a teacher is , she can only say one thing at a time . How then can she clear up many different misconceptions in a reasonable time ? The best solution developed so far is peer instruction . Originally created by Eric Mazur at Harvard [28] , it has been studied extensively in a wide variety of contexts ( e . g . , [29 , 30] ) . Peer instruction is essentially a scalable way to provide one-to-one mentorship . It interleaves formative assessment with student discussion as follows: The questions posed to learners don't have to be MCQs: matching terms to definitions can be equally effective , as can Parsons Problems ( in which they are given the jumbled parts of a solution and must put them in the right order [31] ) . Whatever mix is used , the lesson must build toward them , and the question must probe for conceptual understanding and misconceptions ( rather than check simple factual knowledge ) . Group discussion significantly improves students' understanding because it forces them to clarify their thinking , which can be enough to call out gaps in reasoning . Repolling the class then lets the teacher know whether they can move on or whether further explanation is necessary . A final round of additional explanation and discussion after the correct answer is presented gives students one more chance to solidify their understanding . But could this be a false positive ? Are results improving because of increased understanding during discussion or simply from a follow-the-leader effect ? [32] tested this by following the first question with a second one that students answer individually and found that peer discussion actually does enhance understanding , even when none of the students in a discussion group originally knew the correct answer . It is important to have learners vote publicly so that they can't change their minds afterwards and rationalize it by making excuses to themselves like “I just misread the question . ” Some of the value of peer instruction comes from having their answer be wrong and having to think through the reasons why . This is called the “hypercorrection effect” [33] . Most people don't like to be told they're wrong , so it's reasonable to assume that the more confident someone is that the answer they've given in a test is correct , the harder it is to change their mind if they were actually wrong . However , it turns out that the opposite is true: the more confident someone is that they were right , the more likely they are not to repeat the error if they are corrected .
A worked example is a step-by-step demonstration of how to solve a problem or do some task . By giving the steps in order , the instructor reduces the learner's cognitive load , which accelerates learning [27 , 34] . However , worked examples become less effective as learners acquire more expertise [35 , 36] , a phenomenon known as the “expertise reversal effect . ” In brief , as learners build their own mental models of what to do and how to do it , the detailed step-by-step breakdown of a worked example starts to get in the way . This is why tutorials and manual pages both need to exist: what's appropriate for a newcomer is frustrating for an expert , while what jogs an expert's memory may be incomprehensible to a novice . One powerful way to use worked examples is to present a series of “faded examples” [37] . The first example in the series is a complete use of a problem-solving strategy; each subsequent example gives the learner more blanks to fill in . The material that isn't blank is often referred to as scaffolding since it serves the same purpose as the scaffolding set up temporarily at a building site . Faded examples can be used in almost every kind of teaching , from sports and music to contract law . Someone teaching high school algebra might use them by first solving this equation for x: and then asking learners to fill in the blanks in this: The next problem might be this: Learners would finally be asked to solve an equation entirely on their own: ( 2x+7 ) /4=1 At each step , learners have a slightly larger problem to solve , which is less intimidating than a blank screen or a blank sheet of paper . Faded examples also encourage learners ( and instructors ) to think about the similarities and differences between various approaches . Worked examples are themselves an example of “concreteness fading” [38 , 39] , which describes the process of starting lessons with things that are specific or tangible and then explicitly and gradually transitioning to more abstract and general concepts . Concreteness fading One way to remember this strategy is the acronym PETE ( Problem , Explanation , Theory , Example ) , which encourages instructors to
It is almost oxymoronic to say that learners spend a lot of their time trying to figure out what they've done wrong and fixing it: after all , if they knew and they had , they would already have moved on to the next subject . Most lessons devote little time to detecting , diagnosing , and correcting common mistakes , but doing this will accelerate learning—not least by reducing the time that learners spend feeling lost and frustrated . In Carroll and colleagues’ “minimal manual” approach to training materials , every topic is accompanied by descriptions of symptoms learners might see , their causes , and how to correct them [40] . When studying second language acquisition , [41] identified six ways in which instructors can correct learners' mistakes: All of these can be used preemptively during the design of lessons . An introduction to chemical reactions , e . g . , could present an incomplete calculation of enthalpy and ask the learner to fill it in ( elicitation ) or present the complete calculation with errors , then draw attention to those errors and correct them one by one ( recasting ) . All of these strategies provide retrieval practice by requiring learners to use what they have just learned and encourage metacognition by requiring them to reflect on the limits and applicability of that knowledge .
One of the strongest predictors of whether people learn something is their “intrinsic motivation , ” i . e . , their innate desire to master the material . The term is used in contract with “extrinsic motivation , ” which refers to behavior driven by rewards such as money , fame , and grades . As [42] describes , the biggest motivators for adult learners are their sense of agency ( i . e . , the degree to which they feel that they're in control of their lives ) , the utility or usefulness of what they're learning , and whether their peers are learning the same things . Letting people go through lessons at the time of their own choosing , using authentic tasks , and working in small groups speak to each of these factors . Conversely , it is very easy for educators to demotivate their learners by being unpredictable , unfair , or indifferent . If there is no reliable relationship between effort and result , learners stop trying ( a particular case of a broader phenomenon called “learned helplessness” ) . If the learning environment is slanted to advantage some people at the expense of others , everyone will do less well on average [43] , and if the lessons make it clear that the teacher doesn't care if people learn things or not , learners will mirror that indifference . One way to tell if learners are motivated or not is to look at the incidence of cheating . In classrooms , it is usually not a symptom of moral failing but a rational response to poorly designed incentives . As reported in [44] , some things that educators do that unintentionally encourage cheating include Eliminating these from lessons doesn't guarantee that learners won't cheat but does reduce the incidence . ( And , despite what many educators believe , cheating is no more likely online than in person [45] . )
“Inclusivity” is a policy of including people who might otherwise be excluded . In STEM education , it means making a positive effort to be more welcoming to women , under-represented racial or ethnic groups , people with various sexual orientations , the elderly , the physically challenged , the economically disadvantaged , and others . The most important step is to stop thinking in terms of a “deficit model , ” i . e . , to stop thinking that the members of marginalized groups lack something and are therefore responsible for not getting ahead . Believing that puts the burden on people who already have to work harder because of the inequities they face and ( not coincidentally ) gives those who benefit from the current arrangements an excuse not to look at themselves too closely . One axis of inclusive lesson design is physical: provide descriptive text for images and videos to help the visually challenged , closed captions for videos to help those with hearing challenges , and so on . Another axis is social: Committing fully to inclusive teaching may mean fundamentally rethinking content . [46] , e . g . , explored two strategies for making computing education more culturally inclusive , each of which has its own traps for the unwary . The first strategy , community representation , highlights students' social identities , histories , and community networks using afterschool mentors or role models from students' neighborhoods or activities that use community narratives and histories as a foundation for a computing project . The major risk is shallowness , e . g . , using computers to build slideshows rather than do any real computing . The second strategy , computational integration , incorporates ideas from the learner's community , e . g . , by reverse engineering indigenous graphic designs in a visual programming environment . The major risk here is cultural appropriation , e . g . , using practices without acknowledging origins . No matter which strategy is chosen , the first steps should always be to ask your learners and members of their community what they think you ought to do and to give them control over content and direction .
Following the 10 tips laid out above doesn't guarantee that your lessons will be great , but it will help ensure that they aren't bad . When it comes time to put them into practice , we recommend following something like the reverse design process developed independently by [47 , 48 , 49]: We also recommend that lessons be designed for sharing with other instructors . Instructors often scour the web for ideas , and it's common for people to inherit courses from previous instructors . What is far less common is collaborative lesson construction , i . e . , people taking material , improving it , and then offering their changes back to the community . This model has served the open source software community well , and as [9] describes , it works equally well for lessons—provided that materials are designed to make fine-grained collaboration easy . Unfortunately , widely-used systems like Git are designed to handle text files and struggle with structured document formats like Microsoft Word or PowerPoint . In addition , their learning curve is very steep and deters many potential users who have deadlines to meet or would rather think about engaging exercises than try to make sense of obscure error messages . One key enabler of collaborative lesson construction is licensing . We strongly recommend using one of the Creative Commons family of licenses since they have been carefully vetted and are widely understood . | As a species , we know as much about teaching and learning as we do about public health , but most people who teach at the postsecondary level are never introduced to even the basics of evidence-based pedagogy . Knowing just a few key facts will help you build more effective lessons in less time and with less pain and will also make those lessons easier for your peers to find and reuse . This paper presents 10 tips that you can apply immediately and explains why they work . | [
"Abstract",
"Introduction",
"1.",
"Use",
"learner",
"personas",
"to",
"define",
"your",
"audience",
"2.",
"Design",
"for",
"effective",
"learning",
"strategies",
"3.",
"Write",
"summative",
"assessments",
"to",
"set",
"concrete",
"goals",
"4.",
"Write",
"formative",
"assessments",
"for",
"pacing,",
"design,",
"preparation,",
"and",
"reinforcement",
"5.",
"Integrate",
"visual",
"and",
"linguistic",
"information",
"6.",
"Design",
"for",
"peer",
"instruction",
"7.",
"Use",
"worked",
"examples",
"and",
"concreteness",
"fading",
"8.",
"Show",
"how",
"to",
"detect,",
"diagnose,",
"and",
"correct",
"common",
"mistakes",
"9.",
"Motivate",
"and",
"avoid",
"demotivating",
"10.",
"Make",
"lessons",
"inclusive",
"Conclusion"
] | [
"learning",
"neurolinguistics",
"linguistics",
"education",
"social",
"sciences",
"neuroscience",
"learning",
"and",
"memory",
"cognitive",
"psychology",
"cognition",
"memory",
"vision",
"cognitive",
"linguistics",
"instructors",
"human",
"learning",
"memory",
"recall",
"people",
"and",
"places",
"professions",
"psychology",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"sensory",
"perception",
"language",
"acquisition",
"cognitive",
"science"
] | 2019 | Ten quick tips for creating an effective lesson |
The CDKN1B gene encodes the cyclin-dependent kinase inhibitor p27KIP1 , an atypical tumor suppressor playing a key role in cell cycle regulation , cell proliferation , and differentiation . Impaired p27KIP1 expression and/or localization are often observed in tumor cells , further confirming its central role in regulating the cell cycle . Recently , germline mutations in CDKN1B have been associated with the inherited multiple endocrine neoplasia syndrome type 4 , an autosomal dominant syndrome characterized by varying combinations of tumors affecting at least two endocrine organs . In this study we identified a 4-bp deletion in a highly conserved regulatory upstream ORF ( uORF ) in the 5′UTR of the CDKN1B gene in a patient with a pituitary adenoma and a well-differentiated pancreatic neoplasm . This deletion causes the shift of the uORF termination codon with the consequent lengthening of the uORF–encoded peptide and the drastic shortening of the intercistronic space . Our data on the immunohistochemical analysis of the patient's pancreatic lesion , functional studies based on dual-luciferase assays , site-directed mutagenesis , and on polysome profiling show a negative influence of this deletion on the translation reinitiation at the CDKN1B starting site , with a consequent reduction in p27KIP1 expression . Our findings demonstrate that , in addition to the previously described mechanisms leading to reduced p27KIP1 activity , such as degradation via the ubiquitin/proteasome pathway or non-covalent sequestration , p27KIP1 activity can also be modulated by an uORF and mutations affecting uORF could change p27KIP1 expression . This study adds the CDKN1B gene to the short list of genes for which mutations that either create , delete , or severely modify their regulatory uORFs have been associated with human diseases .
CDKN1B encodes the cyclin-dependent kinase ( CDK ) inhibitor , p27KIP1 , which negatively regulates the Cdk2/cyclin E and Cdk2/cyclin A protein complexes , thereby preventing the progression from the G1 to the S phase of the cell cycle [1] . In G0 and early G1 , p27KIP1 expression and stability are maximal . During the G1 phase gradual degradation of p27KIP1 is associated with an increased activity of Cdk2/cyclin E and Cdk2/cyclin A complexes to stimulate cell proliferation [2] , [3] . Several mitogenic ( i . e . MAPK , PI3K/AKT ) and anti-proliferative ( i . e . TGFβ/SMAD ) signal transduction pathways regulate p27KIP1 expression and activity , making it a central integration point for cell-fate decision [4] . These pathways can regulate p27KIP1 at different levels , including transcription , translation , intracellular localization or ubiquitin-mediated proteasomal degradation [5] . p27KIP1 acts as an atypical tumor suppressor as it is rarely mutated in human cancers , but frequently underexpressed or mislocalized in human malignancies [4] . Although an augmented proteolysis was initially suggested as the major cause of p27KIP1 loss in human tumors [6] , recent findings propose that reduced translation and/or transcription of CDKN1B also contributes to p27KIP1 deficiency [7]–[9] . Translation of CDKN1B may involve regulatory elements within its 5′UTR , including an internal ribosome entry site ( IRES ) and an upstream ORF ( uORF ) [10] , [11] . The IRES supports p27KIP1 expression when cap-dependent translation is reduced , such as during quiescence or stress conditions [10] , [12] . Reduced IRES-mediated translation , due to mutations in the pseudouridine synthase that alters the ribosome's ability to efficiently engage the CDKN1B IRES element , may contribute to the increased predisposition to cancer in X-linked congenital dyskeratosis [13] . Germline mutations in the CDKN1B gene have been recently associated with the development of a multiple endocrine neoplasia syndrome both in humans ( MEN4 , MIM 610755 ) and in rats ( MENX ) [14] . Multiple endocrine neoplasias , including type 1 ( MEN1 , MIM 131100 ) and type 2 variants , ( MEN2 , MIM 171400 , MIM 162300 ) , are a group of autosomal dominant syndromes characterized by varying combinations of tumors affecting at least two endocrine organs [15] . To date , seven CDKN1B germline mutations have been identified in MEN4 patients primarily associated with MEN1-related lesions , including parathyroid and pituitary tumors , but the presence of other malignancies such as renal angiomyolipoma , papillary thyroid carcinoma and pancreatic masses has also been reported [8] , [9] , [14] , [16] , [17] . Two further germline mutations have been more recently associated with sporadic hyperparathyroidism [18] . In MEN4 , CDKN1B mutations either affect p27KIP1 cellular localization , protein stability or the binding with functional partners such as Cdk2 or Grb2 [8] , [17] . Reduced transcription/translation efficiency due to mutations in elements regulating translation initiation ( i . e . , in the Kozak sequence , or forming a secondary stem loop structure within the CDKN1B 5′UTR ) , has also been described [8] , [9] . Germline CDKN1B mutations are hence rare events in MEN1-like subjects ( individuals with MEN1-related lesions , without MEN1 inactivating mutations ) , being identified in less than 3% of cases [8] , [16] and a clear genotype-phenotype correlation has not been established to date . In the present paper we analyzed the CDKN1B gene looking for point mutations and large rearrangements in order to determine the possible cause of multiple endocrine tumors in 25 consecutive sporadic and familial patients with typical MEN1-related symptoms . We identified a 4-bp deletion that modifies the regulatory uORF in the 5′UTR of the CDKN1B gene in a patient with tumors in the pituitary gland and the endocrine pancreas . Functional studies based on dual-luciferase assay and site-directed mutagenesis further support the deleterious influence of this deletion on translation reinitiation at the CDKN1B starting site , with a consequent reduction of p27KIP1 expression both in vitro and in vivo .
Among the 25 patients with MEN1-related symptoms , a 4-bp deletion ( c . -456_-453delCCTT , NM_004064 ) within the 5′UTR of CDKN1B in a 62 year old female patient with acromegaly and a well-differentiated non-functioning pancreatic endocrine neoplasm has been identified . This sequence variant was not detected in either 600 chromosomes or in the dbSNP/1000 genomes databases . The 5′UTR of the CDKN1B gene is highly structured , containing several translational regulatory elements . An IRES element sustains p27KIP1 translation under poor growth conditions [10] , [12] , while a G/C-rich hairpin domain contributes to cell-cycle dependent regulation of CDKN1B translation [11] . Downstream the G/C-rich domain and encompassing the c . -456_-453delCCTT , an uORF coding for a 29 amino acid-long peptide has been described that has been suggested to inhibit the in vitro synthesis of p27KIP1 and to enhance its cell cycle-dependent translation [11] . An extensive comparative analysis of DNA and protein sequences from multiple species ( Figure 1 ) confirmed previous data of high evolutionary conservation among vertebrates of the uORF [11] , and support the hypothesis of a functional role of this element [11] . In general , uORFs are small open reading frames located in the 5′UTR of genes that influence translation during ribosome scanning , thus modulating gene expression . A scanning ribosome encountering an uORF has multiple fates: it can i ) translate the uORF; ii ) scan through the sequence ( leaky scanning ) and reinitiate translation further downstream at a proximal or distal ATG; iii ) induce ribosome stalling or premature dissociation at the uORF stop codon , thus reducing downstream-cistron translation [19] or down-regulating gene expression by promoting mRNA decay [20] . In our case the 4-bp deletion shifts the uORF termination codon , thus lengthening the uORF encoded peptide from 29 to 158 amino acids and shortening the intercistronic space from 429 to 38 bp , with a possible negative influence on translation reinitiation from the main ATG ( Figure 2 ) . Long uORFs and short intercistronic regions may indeed prevent the 40S ribosomal subunits from keeping and/or re-acquiring appropriate cofactors for translation resumption/reinitiation at the downstream ATG [21] , [22] . To address the possibility that the 4-bp deletion affects transcription and/or mRNA stability , making a decreased translation rate due to reduced reinitiation efficiency biologically irrelevant , or alters the promoter usage pattern preventing transcription of the uORF-containing isoform [10] , we measured the steady state levels of CDKN1B allelic mRNAs from whole blood by 5′RACE and allele-specific qPCR . As reported in Figure 3a , both wild type and mutated alleles were expressed in blood cells in almost equal amounts , suggesting that the identified deletion does not alter mRNA steady state levels , and therefore probably does not alter either transcription or mRNA stability . An apparently unique 5′UTR of >530 bp has been identified in the c . -456_-453delCCTT carrier and in healthy controls , supporting the concept that the transcription pattern is preserved in the mutated subject ( Figure 3b ) . The pancreatic tumor of the mutated patient was then analyzed by immunohistochemistry for p27KIP1 expression and for the proliferation antigen Ki-67 , and compared with similar tumors from CDKN1B-mutation negative subjects . Parallel differences in expression level and localization were found . We observed weak cytoplasmic staining in tumor cells and very strong nuclear staining in the interspersed normal endothelial cells in the MEN4 patient ( Ki67<1% ) , while in contrast p27KIP1 nuclear staining was found in a high proportion of sporadic well-differentiated pancreatic tumors examined ( Figure 4 ) . The reduction in nuclear p27KIP1 and/or its cytoplasmic mislocalization has been reported in different cancers including breast , colon and prostate [4] . Loss of p27KIP1 may occur through different mechanisms , including augmented proteasome-mediated proteolysis and impaired translation [23] . On the other hand , the cytoplasmic mislocalization may be associated with imbalanced p27KIP1 phosphorylation due to the oncogenic activation of PI3K- and MEK-dependent kinases , mimicking protein loss [4] . Indeed , in the cytoplasm p27KIP1 is unable to exert its inhibitory activity on CDK even in the presence of anti-mitogenic stimuli . On the same lesion loss of heterozygosity ( LOH ) analysis was then performed . No loss of the wild type allele was observed ( Figure 2 ) . Moreover , the biallelic expression of an uORF-containing transcript has been observed ( Figure 3c ) , further confirming that p27KIP1 may act as a haploinsufficient tumor suppressor [24] . To identify possible additional uORF mutations , we extended the CDKN1B 5′UTR analysis to additional 41 patients with typical MEN1-like features previously reported negative for mutations in the CDKN1B coding sequence [17] , [25] . A c . -469C>T substitution resulting in a silent change in the uORF was detected in a single patient but not in healthy controls ( see above ) . To determine whether the two identified substitutions negatively affect CDKN1B translation , the wild type and mutated 5′UTRs were cloned upstream of the firefly luciferase gene ( Figure 5a ) . By transfecting lovastatin G1-synchronized or asynchronous HeLa and GH3 cells , we demonstrated that the c . -456_-453delCCTT , but not the c . -469C>T variant , significantly reduced luciferase activity in a cell cycle phase-independent manner ( Figure 6a , 6b ) . When we analyzed the luciferase mRNA from the transfected cells by quantitative real-time RT-PCR , we demonstrated that the effects of the 4-bp deletion are largely due to reduction in translation rate rather than to changed steady-state mRNA levels ( Figure 6c ) , in agreement with our observation on blood CDKN1B mRNA ( Figure 3a ) and with the trend observed in large-scale datasets [26] . We then evaluated the effect of the 4 bp deletion on p27KIP1 translation by transfecting HEK293 cells with vectors with either the wild type or the mutated 5′UTRs cloned upstream the CDKN1B gene ( Figure 5b ) . We confirmed a significant reduction in p27KIP1 protein levels as a consequence of the 5′UTR c . -456_-453delCCTT mutation ( Figure 6d ) . In a previous study on HeLa cells using an identical wild type construct , the CDKN1B 5′UTR induced luciferase expression only during G1 progression or in lovastatin-arrested cells [11] . Although we cannot exclude the presence of DNA variations on regulatory elements between the two cloned sequences , a possible biological variability between batches of cells from the same cell line seems the more plausible explanation . However , similar cell-cycle independent luciferase activation was observed under our experimental conditions in three additional cell lines , namely GH3 ( Figure 6a ) , SH-SY5Y and HEK293 ( Figure S1 ) . Based on such observation we may therefore suggest the need for further studies for better clarifying the cell-cycle dependent translation of p27KIP1 regulated by the CDKN1B 5′UTR . Site-directed mutagenesis ( c . -428A>T ) was then used to reintroduce a stop codon in the c . -456_-453delCCTT containing vector , thus restoring both uORF length and intercistronic distance ( Figure 5c ) . After transfection , the uORF regulatory properties were almost completely rescued in the double mutant compared to the c . -428A>T construct ( Figure 7a ) , further supporting the hypothesis that the 4 bp deletion affects translation reinitiation of the downstream CDKN1B ORF . In addition , the lack of complete recovery of the uORF modulatory activity , possibly due to differences in the C-terminus of the uORF-encoded peptide ( Figure 7b ) , further confirms that the CDKN1B uORF belongs to the class of sequence-dependent uORFs that exert their inhibitory role by acting in cis to regulate components of the translation apparatus [19] . To evaluate the ability of the uORF to be translated , which represents the central point of our hypothesis on the possible deleterious effects of the c . -456_-453delCCTT change , the wild type or the mutated 5′UTRs were placed upstream of the CDKN1B open reading frame and the c . -74insC mutation was introduced by site-directed mutagenesis . This additional DNA variant leads to the in-frame fusion of the mutated uORF with the main gene ( Figure 5d ) . As expected , the chimeric product was detected only in the c . -74insC+c . -456_-453delCCTT transfected HEK293 cells , and was again associated with a significant reduction of p27KIP1 expression ( Figure 6e ) . To our knowledge , this is the first direct evidence of the translation of the CDKN1B uORF in a cellular system . However , it remains to be clarified if this peptide has additional biological functions other than repressing translation of the CDKN1B ORF as an effect of impaired reinitiation , as suggested for a subset of uORFs [27] . To elucidate the molecular mechanism by which c . -456_-453delCCTT determines a decrease in p27KIP1 translational efficiency , we estimated the relative proportion of the two allelic mRNAs engaged in translation in the immortalized lymphoblastoid cells of the heterozygote patient . To this aim , the cell lysates were subjected to polysome fractionation through sucrose gradient ultracentrifugation [28] and we determined the level of each of the two allelic mRNAs for each fraction . Figure 8a reports the distribution of ribosomal RNA in the different fractions , showing a typical distribution with polysomes reproducibly spanning fractions 7–11 . The distributions of the wild type and c . -456_-453delCCTT transcripts present an almost superimposable pattern , being for both about 90% of the total detectable mRNA localized in polysomes with a peak corresponding to fraction 9 ( compare Figure 8b with Figure 8a ) . However , when the amounts of both alleles were expressed as differences between Cq values for mRNA and for genomic DNA for removing the intrinsic variation between the two qPCR assays , a clear preponderance on polysomes of the wild type CDKN1B mRNA could be observed ( Figure 8c ) , which we can estimate to be of the order of about three times . Since the levels of the two allelic mRNAs in the cells are the same ( Figure 3a ) , this implies that the c . -456_-453delCCTT mRNA suffers decreased average polysomal loading with respect to the wild type mRNA . Therefore , the two different CDKN1B mRNAs are differentially loaded in polysomes despite being present in the cells in the same relative amounts , and despite the fact that they share a distribution profile on polysomes of different molecular weights . The result is compatible with a decreased efficiency of translation reinitiation of the CDKN1B ORF due to the c . -456_-453delCCTT mutation . The data we presented here further confirm the role of CDKN1B germline mutations in predisposing to a MEN4 syndrome . Furthermore , they demonstrate that a reduced translation initiation rate of p27KIP1 due to the ineffective regulatory activity of its uORF may be associated with transformation . Based on our data , mutations in the CDKN1B-regulating uORF seem to be rare . However , previous studies on CDKN1B germline mutations in MEN1-like patients did not consider the uORF region [8] , [9] , [14] , [16]–[18] , [25] , [29] , and therefore the prevalence of this type of mutation remains to be established . Our results emphasize thus the need for the inclusion of the entire 5′UTR region of CDKN1B in molecular testing for MEN4 . Increasing evidence suggests that uORF-mediated translational control may represent an important mechanism in the regulation of gene expression . This is supported by the close relationship of mutations that introduce or disrupt uORFs and the pathophysiology of several human diseases , including cancer [30] . To date , only three well-known hereditary diseases have been associated with uORF-affecting mutations: i ) thrombocythemia due to thrombopoietin mutation [31] , ii ) melanoma due to CDKN2A mutation [32] and iii ) Marie Unna hypotrichosis due to mutations in the hairless gene [33] . Other diseases , such as breast cancer , Alzheimer's diseases , arrhythmogenic right ventricular cardiomyopathy have also been suggested to be associated to genes which have uORF-related control [34]–[36] . However , the pathogenic effects of deregulated uORF-mediated translation in these cases remain to be clarified [37] . Many important genes involved in controlling cell growth ( i . e . receptors , oncogenes , growth factors ) harbor uORF in their 5′UTR [38] . Some of these genes override the uORF-mediated translational repression and accumulate their protein product in cancer cells [39] . Translational derepression elements in the 3′UTR may counteract the inhibitory activity of uORFs on translation [39]; however , mutations inducing loss of uORF function in oncogenes might lead to a similar increase of translation rate and consequently to malignant transformation . Conversely , gain of function mutations in uORFs regulating tumor suppressor genes may reduce translation of protective proteins leading to tumor formation [32] . Similarly to a point mutation introducing a regulative uORF in the leader sequence of the tumor suppressor gene CDKN2A in hereditary melanoma [32] , the 4-bp deletion in CDKN1B gene we describe here led to the reduced production of CDKN1B-encoded protein p27KIP1 , probably due to a decreased translation reinitiation rate , which then results in predisposition to tumor development . In conclusion , the CDKN1B mutation functionally characterized in this study represents a novel example of an uORF-affecting mutation . Our functional studies show the negative influence of this deletion on the translation reinitiation at the CDKN1B starting site thus providing novel insights into the role of uORFs in the pathogenesis of human diseases . In addition to the classical mechanisms of degradation by the ubiquitin/proteasome pathway and by non-covalent cytoplasmic sequestration , our findings demonstrate that p27KIP1 activity can also be modulated by its uORF , and mutations affecting this sequence may lead to reduced expression of p27KIP1 protein .
The cohort of patients screened for mutations in the entire CDKN1B gene consisted of 25 consecutive patients with two or more typical MEN1-related symptoms ( hyperparathyroidism , neuroendocrine tumors , pituitary adenoma ) . Patients were collected and diagnosed at the Division of Endocrinology ( University/Hospital of Padova ) and at the Familial Cancer Clinic and Oncoendocrinology ( Veneto Institute of Oncology ) , Padova , Italy , following the recognized clinical practice guidelines [40] . All patients had negative mutational screening for MEN1 , PRKAR1A and AIP genes . A second group of additional 41 patients with similar phenotype has been analyzed only for the uORF sequence since the rest of the gene has been analyzed and published previously without finding any pathogenic mutations [17] , [25] . The study was conducted in accordance with the Helsinki declaration . Local ethical committees from each referring center approved the study , and all subjects gave written informed consent . The whole coding region , intron–exon boundaries , and 5′- and 3′-UTRs of CDKN1B were amplified and directly sequenced as reported elsewhere [41] . All primer pairs used were designed by PRIMER3 ( http://primer3 . sourceforge . net/ ) and synthesized by IDT ( Leuven , Belgium ) . Primers for point mutation analysis of the entire human CDKN1B gene were P0F , 5′-agcagtacccctccagcagt-3′; P0R , 5′-aaagcccgtccgagtctg-3′; P1F , 5′-ccaatggatctcctcctctg-3′; P1R , 5′-ggagccaaaagacacagacc-3′; P2F , 5′-ccatttgatcagcggagact-3′; P2R , 5′-gccctctaggggtttgtgat-3′; P3F , 5′-gagttaacccgggacttggag-3′; P3R , 5′-atacgccgaaaagcaagcta-3′; P4F , 5′-tgactatggggccaacttct-3′; P4R , 5′-tttgccagcaaccagtaaga-3′; P5F , 5′-ccccatcaagtatttccaagc-3′; P5R , 5′-cctcccttccccaaagttta-3′; P6F , 5′-tgcctctaaaagcgttggat-3′; P6R , 5′-tttttgccccaaactacctg-3′; P7F , 5′-gccctccccagtctctctta-3′; P7R , 5′-ggtttttccatacacaggcaat-3′; P8F , 5′-tctgtccatttatccacaggaa-3′; P8R 5′-tgccaggtcaaataccttgtt-3′ . Previously unreported nucleotide changes were screened in 300 healthy , anonymous , unrelated individuals by Tetra-primer ARMS-PCR [42] and searched in the dbSNP and 1000 genomes databases ( http://www . ncbi . nlm . nih . gov/projects/SNP/; http://www . 1000genomes . org/ ) . The NHLBI Exome Sequencing Project - Exome Variant Server database ( http://evs . gs . washington . edu/EVS ) has been queried for the c . -469C>T . Primers for Tetra-primer ARMS-PCR were: hp27delOUTR , 5′-agccgctctccaaacctt-3′; hp27delOUTF , 5′-caatggatctcctcctctgttt-3′; hp27delINF , 5′-cttcttcgtcagcctcccac-3′; hp27-469INR , 5′-tggcggtggaagggaggctgacgcaa-3′; hp27-469INF , 5′-gactcgccgtgtcaatcattttcgtc-3′ . Gene dosage alteration was assessed by the quantitative multiplex PCR of short fluorescent fragments ( QMPSFs ) and by long-range PCR ( LR-PCR ) as previously described [41] using the following primers: CLIF , 5′-tggtcagagagtggcctttctc-3′; CLIR , 5′-tgccgagtagaggcatttagtca-3′; CLIIF , 5′-tgtctgtgacgccgttgtct-3′; CLIIR , 5′-aagggttttctagcacacataggaa-3′; 1IF , 5′-gccgcaaccaatggatctc-3′; 1IR , 5′-acgagccccctttttttagtg-3′; 1IIF , 5′-ctctgaggacacgcatttggt-3′; 1IIR , 5′-aaatcagaatacgccgaaaagc-3′; 2F , 5′-tttcccctgcgcttagattc-3′; 2R , 5′-ccaccgagctgtttacgtttg-3′; 3IF , 5′-ccccatcaagtatttccaagct-3′; 3IR , 5′-gttattgtgttgttgtttttcagtgctta-3′; 3IIF , 5′-aacttccatagctattcattgagtcaaa-3′; 3IIR , 5′- tgagcgatgtggctcggct -3′ . Sequence based-LOH analysis was performed on the pancreatic lesion by direct analysis of the CDKN1B mutation . The human cervical carcinoma HeLa , the rodent p27KIP1-negative GH-secreting pituitary adenoma GH3 , the human embryonic kidney HEK293 and the human neuroblastoma SH-SY5Y cell lines ( American Type Culture Collection , Manassas , VA ) , were maintained at 37°C in a 5% CO2 in complete 10% FCS DMEM . GH3 ( 1 . 5×105 cells/well ) , HeLa ( 1 . 0×105 cells/well ) , SH-SY5Y ( 1 . 30×105 cells/well ) and HEK293 ( 2 . 5×105 cells/well ) cells were plated 24 hours before transfection into 12-well plates . When necessary , 24 hours after seeding cells have been arrested in G1 phase by a 36-hour treatment with either 10 µM ( GH3 ) or 20 µM ( HeLa ) lovastatin . In all cell lines but HEK293 ( see below ) transient transfection was performed by Superfect ( Qiagen , Milan , Italy ) . 1 . 5 µg plasmid and a ratio µg DNA/µl Superfect of 1∶6 following manufacturer's protocol were used . The pRL-TK plasmid ( Promega ) encoding Renilla luciferase was cotransfected and used for normalization of transfection efficiency . After 3 hours , the medium was changed to DMEM with 2% FCS and incubated for further 24 hours . Cells were then harvested in passive lysis buffer ( Promega ) and the relative luciferase activity was measured using the Dual-Luciferase Assay System and a GloMax 20/20 luminometer ( Promega ) according to the manufacturer's instructions . For expression experiments , HEK293 cells were seeded into 12-well plates , grew to 95% confluence and transfected with Lipofectamine 2000 ( Invitrogen , Milan , Italy ) following the manufacturer's protocol . Cells were harvested 24 hours post-transfection , lysed in RIPA Buffer supplemented with proteases inhibitors ( MgCl2 10 mM , Pepstatin 1 µM , PMSF 1 mM , cOmplete 1X ( Roche , Monza , Italy ) ) . Samples were clarified by centrifugation at 13 , 000 rpm for 5 min at 4°C . Concentrations of the HEK293 extracted proteins were determined using the Bio-Rad DC protein assay kit ( Bio-Rad Italia , Milan , Italy ) following the manufacturer's instructions . For each sample , 20 µg were resuspended in NuPAGE LDS sample buffer and NuPAGE sample reducing agent ( Invitrogen ) , boiled for 10 min at 70°C and resolved by SDS-PAGE on 4–12% NuPAGE gels ( Invitrogen ) and Mes buffer ( Invitrogen ) . Separated proteins were transferred onto nitrocellulose membrane by Trans-Blot Turbo transfer system ( BioRad ) that was blocked for 2 hours with 5% non-fat dry milk ( BioRad ) . The membrane was incubated overnight at 4°C with anti-p27KIP1 monoclonal antibody ( BD Bioscience Heidelberg , Germany ) used at 1∶300 . Expression was corrected for differences in protein loading by probing blots for 1 hour at RT with mouse anti-ß-actin antibody ( clone AC-15 1∶5 , 000 , Sigma-Aldrich , Milan , Italy ) . Blots were developed using Pierce ECL Substrate ( Part No . 32106 , Thermo Scientific , Rockford , IL USA ) and exposed to CL-XPosure Film ( Thermo Scientific ) . For total RNA extraction , HEK293 cells were resuspended in TRIzol ( Invitrogen ) and processed according to the manufacturer's instructions . Plasmid DNA contamination was removed by DNase , treating total RNA twice with Turbo DNA free kit ( Applied Biosystems , Milan , Italy ) . One µg of DNase-treated RNA was reverse-transcribed using M-MuLV Reverse Transcriptase RNase H- ( F-572S , Finnzymes , Espoo , Finland ) . qPCR was done with Platinum SYBR Green qPCR SuperMix-UDG ( Invitrogen ) in an ABI PRISM 7900HT Sequence Detector ( Applied Biosystems ) . A final concentration of 300 nM for both forward and reverse primers was used . Primers for qPCR were qLUCF , 5′-gcctgaagtctctgattaagt-3′; qLUCR , 5′-acacctgcgtcgaaga-3′; qrBActF , 5′-agattactgccctggctcct-3′; qrBActR , 5′-aacgcagctcagtaacagtccg -3′; qhGAPDHF , 5′- ctctctgctcctcctgttcgac-3′; qhGAPDHR , 5′- ctctctgctcctcctgttcgac-3′ . Threshold levels were set at the exponential phase of qPCR using Sequence Detection software , version 2 . 4 ( Applied Biosystems ) . The amount of each target gene relative to the proper housekeeping gene ( HK , rat β-actin or human GAPDH ) was determined using a relative standard curve method and the results were expressed as a ratio of target gene/HK . A 38-cycle threshold was set , beyond which the gene was considered undetectable . Total RNA from whole blood samples was obtained using Paxgene Blood RNA Kit ( Qiagen ) following manufacturer protocol , while RNA from paraffin-embedded pancreatic tumor tissue was extracted using a modified RNAzol method , as previously described [43] . RNA was reverse-transcribed as described above . Allele-specific analysis was evaluated by qPCR as described above using two different SYBR assays . A final concentration of 300 nM for both forward primers and 50 mM for the unique reverse primer was used ( -456_-453del_wtF 5′- cttcttcgtcagcctccctt-3′; -456_-453del_mutF 5′- cttcttcgtcagcctcccac-3′; -456_-453del-R 5′-agccgctctccaaacctt-3′ ) . Given the different efficiency that may characterize the two different assays , the value of each allele was referred to the genomic DNA expressed as ΔCq ( Cq value obtained for mRNA minus Cq value for genomic DNA ) . The lack of a possible deletion/duplication of the corresponding genomic locus was proven by QMPSFs as described above . 5′RACE was performed using 5′RACE System 2 . 0 kit ( Invitrogen ) on whole blood derived total RNA following manufacturer's instructions . Briefly , first strand cDNA was synthesized from 2 µg of mRNA by SuperScriptII RNA polymerase reaction using the specific primers GSP1R ( 5′- gttaactcttcgtggtcc -3′ ) . After adding an oligo-dC tail to the cDNAs 3′-ends , a PCR reaction has been performed with GSP2R primer ( 5′-ttctcccgggtctgcacg-3′ ) , coupled with an Abridged Anchor Primer ( AAP ) . The resulting DNA fragments were eluted from agarose gel and analyzed by direct sequencing , as reported above . Immunohistochemistry was performed on an automated immunostainer ( Ventana Medical Systems , Frankfurt am Main , Germany ) , according to the manufacturer's protocols with minor modifications [44] using the monoclonal anti-p27KIP1 antibody cited above ( 1∶1 , 000 ) . The monoclonal MIB5 antibody ( 1∶500 , Dako , Hamburg , Germany ) was used to detect the proliferation antigen Ki-67 . Positive controls were used to confirm the adequacy of the staining . PCR fragments were obtained by amplification of the mutation carrier with forward and reverse primers containing extra HindIII and NcoI sites , respectively ( clonF , 5′- catcataagcttccaccttaaggccgcgct -3′; clonR , 5′- catcatccatggttctcccgggtctgcacg -3′ ) . The PCR product was digested and inserted upstream the luciferase reporter gene into the pGL3 Control Vector ( Promega ) . For expression studies the wild type and mutated 5′UTRs were subcloned into pcDNA3 . 1/p27HA ( kind gift of Prof . Sylvain Meloche , Institute for Research in Immunology and Cancer , Université de Montréal , Canada ) . The c . -469C>T , c . -428A>T and c . -74insC modifications were introduced by QuikChange II XL kit ( Stratagene , La Jolla , CA USA ) following manufacturer's protocol . EBV-transformed lymphoblastoid cells were generated by infection of peripheral blood mononuclear cells from the c . -456_-453delCCTT mutation carrier with culture supernatant from the EBV-producing marmoset cell line B95 . 8 ( American Type Culture Collection ) and maintained in RPMI 1640 medium ( Euroclone , Milano , Italy ) supplemented with 10% FBS , 1 mM Na Pyruvate , 10 mM Hepes Buffer , 2 mM Ultraglutamine ( Lonza BioWhittaker , Basel , Switzerland ) , 1% Antibiotic/antimycotic ( Gibco , Invitrogen Corporation ) . Cyclosporin A ( CsA , Sandoz Pharmaceuticals AG; Cham , Switzerland ) was initially added to the cultures to inhibit T cell growth ( final concentration , 0 . 7 µg/ml ) . For polysomal RNA extraction lymphoblastoid cells ( 25×106 ) were incubated with 100 µg/ml cycloheximide for 4 minutes , washed once with phosphate buffer saline ( PBS ) , resuspended in lysis buffer [10 mM NaCl , 10 mM MgCl2 , 10 mM Tris–HCl , pH 7 . 5 , 1% Triton X-100 , 1% sodium deoxycholate , 100 µg/ml cycloheximide , 0 . 2 U/µl RNase inhibitor , 1 mM DTT] and transferred to a microcentrifuge tube . After 5 minutes incubation on ice , the extracts were centrifuged for 10 min at 12 , 000 g at 4°C . The supernatant was collected and stored at −80°C . The cytoplasmic lysates were fractionated by ultracentrifugation ( Sorvall rotor , 100 min at 180 , 000 g ) trough 15–50% linear sucrose gradient containing 30 mM Tris–HCl , pH 7 . 5 , 100 mM NaCl , 10 mM MgCl2 . Eleven fractions were collected monitoring the absorbance at 254 nm . The RNA in each fraction was isolated after proteinase K treatment , phenol–chloroform extraction and isopropanol precipitation . RNA was resuspended in 30 µl of water . For each fraction 1 µg RNA was reverse-transcribed and analyzed by qPCR using allele-specific assays as reported above . | Gene expression can be modulated at different steps on the way from DNA to protein including control of transcription , translation , and post-translational modifications . An abnormality in the regulation of mRNA and protein expression is a hallmark of many human diseases , including cancer . In some eukaryotic genes translation can be influenced by small DNA sequences termed upstream open reading frames ( uORFs ) . These elements located upstream to the gene start codon may either negatively influence the ability of the translational machinery to reinitiate translation of the main protein or , much less frequently , stimulate protein translation by enabling the ribosomes to bypass cis-acting inhibitory elements . CDKN1B , which encodes the cell cycle inhibitor p27KIP1 , includes an uORF in its 5′UTR sequence . p27KIP1 expression is often reduced in cancer , and germline mutations have been identified in CDKN1B in patients affected with a syndrome ( MEN4 ) characterized by varying combinations of tumors in endocrine glands . Here we show that a small deletion in the uORF upstream to CDKN1B reduces translation reinitiation efficiency , leading to underexpression of p27KIP1 and coinciding with tumorigenesis . This study describes a novel mechanism by which p27KIP1 could be underexpressed in human tumors . In addition , our data provide a new insight to the unique pathogenic potential of uORFs in human diseases . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"genetic",
"mutation",
"gene",
"regulation",
"mutation",
"types",
"molecular",
"genetics",
"gene",
"expression",
"endocrinology",
"diabetes",
"and",
"endocrinology",
"biology",
"autosomal",
"dominant",
"protein",
"translation",
"genetics",
"human",
"genetics",
"genetics",
"of",
"disease",
"genetics",
"and",
"genomics"
] | 2013 | A Novel Mutation in the Upstream Open Reading Frame of the CDKN1B Gene Causes a MEN4 Phenotype |
The pigment cells of vertebrates serve a variety of functions and generate a stunning variety of patterns . These cells are also implicated in human pathologies including melanoma . Whereas the events of pigment cell development have been studied extensively in the embryo , much less is known about morphogenesis and differentiation of these cells during post-embryonic stages . Previous studies of zebrafish revealed genetically distinct populations of embryonic and adult melanophores , the ectotherm homologue of amniote melanocytes . Here , we use molecular markers , vital labeling , time-lapse imaging , mutational analyses , and transgenesis to identify peripheral nerves as a niche for precursors to adult melanophores that subsequently migrate to the skin to form the adult pigment pattern . We further identify genetic requirements for establishing , maintaining , and recruiting precursors to the adult melanophore lineage and demonstrate novel compensatory behaviors during pattern regulation in mutant backgrounds . Finally , we show that distinct populations of latent precursors having differential regenerative capabilities persist into the adult . These findings provide a foundation for future studies of post-embryonic pigment cell precursors in development , evolution , and neoplasia .
A fundamental challenge for modern developmental biology is to determine how populations of stem and progenitor cells are established , maintained , and recruited to differentiate at particular times and places during post-embryonic development and in the adult organism . The significance of the problem cannot be overstated . Not only are these cells essential for normal development and homeostasis , but understanding their biology has profound translational importance . If we seek to evoke regenerative responses in a clinical content , then post-embryonic stem and progenitor populations may well supply the cells for doing so [1]–[3] . If we hope to delay natural tissue senescence , it is the life cycle of these cells that may need to be manipulated [4]–[7] . And if we aim to control malignancy , these cells or their transformed progeny will often be our targets of choice [8]–[10] . Pigment cells are of great utility for understanding the biology of post-embryonic stem and progenitor cells . Pigment cells are a classic and enduring system for studying morphogenesis and differentiation , and a century of effort has provided a firm understanding of many aspects of pigment cell development in the embryo [11]–[14] . These cells arise from neural crest cells , which migrate from the dorsal neural tube and contribute not only to pigment cells , but also glia and neurons of the peripheral nervous system , bone and cartilage of the craniofacial skeleton , and more . Despite the long-standing interest in these embryonic events , it is now clear that pigment cells of adults derive in large part from post-embryonic stem cells that are themselves of neural crest origin [15]–[18] . We know some of the mechanisms that underlie post-embryonic precursor development yet many outstanding questions remain . Foremost among these concern the genes and cellular behaviors by which pigment stem or progenitor cells are established during early development and subsequently maintained , whether there exist distinct subpopulations of such cells with different genetic requirements and potentials , and how these cells are recruited during normal development and homeostasis . Answers to these questions will provide insights into the basic biology of the adult pigment cell lineage , and can inform our understanding of post-embryonic neural crest derivatives as well as stem and progenitor cells more generally . These answers are also of enormous biomedical significance , as the skin pigment cell of mammals , the melanocyte , is associated with a variety of human pathologies [19] and transformed cells of this lineage give rise to melanoma , one of the most common cancers [20] , [21] and also one of the most deadly [21]–[25] . Poor outcomes reflect the inefficacy of non-surgical treatments and the highly invasive character of melanoma cells [26]–[29] . This invasiveness results in part from neural crest and melanocyte-specific factors that are already expressed by untransformed precursors , as well as lineage-specific factors that are re-expressed upon transformation [30] . Better understanding the genetic requirements and dynamics of melanocyte development and homeostasis can thus provide insights into the behaviors of transformed cells , and may suggest novel strategies for clinical intervention . In recent years , the zebrafish has proven to be a tractable system for elucidating features of pigment cell development in the embryo and during the larval-to-adult transformation , a period of post-embryonic development analogous to later organogenesis , fetal and neonatal development of mammals [31]–[33] . In the embryo , neural crest cells differentiate into embryonic/early larval melanophores , the zebrafish homologue of mammalian melanocytes . Melanophores and melanocytes depend on many of the same genes and pathways [12] , [14] , [34] , and melanomas with characteristics equivalent to those of human melanomas can be induced in zebrafish [18] , [34] , [35] . The development of zebrafish adult pigmentation involves a “pigment pattern metamorphosis” in which an embryonic/early larval pigment pattern is transformed into that of the adult [12] , [33] , [36]–[38] . Whereas the embryonic/early larval pigment cells and pattern develop by 3 . 6 SSL ( standardized standard length [33]; about 4 days post-fertilization ) , new “metamorphic” melanophores begin to differentiate scattered over the flank by 5 . 9 SSL and ultimately migrate to form the adult stripes . Simultaneously , additional metamorphic melanophores appear already at sites of stripe formation , and many embryonic/early larval melanophores are lost . These events culminate in a juvenile pigment pattern by 11 . 0 SSL ( 4–5 weeks post-fertilization ) , consisting of two melanophore stripes bounding a lighter “interstripe” . Melanophores comprising these stripes reside in the “hypodermis” between the epidermis and the myotomes [39] , [40] . Other adult melanophores are found in the epidermis , the dorsal scales , and the fins . Two additional classes of pigment cells also develop: yellow–orange xanthophores , which populate the interstripe and are required for organizing melanophores into stripes [38] , [41] , [42]; and iridescent iridophores , which are initially limited to the interstripe but later occupy melanophore stripes as well [33] . During later adult development , additional stripes and interstripes are added as the fish grows . Mutants with pigment pattern defects limited to post-embryonic stages have suggested a model in which embryonic/early larval melanophores develop directly from the neural crest , whereas metamorphic melanophores develop from latent stem cells of presumptive neural crest origin . For example , picasso and puma mutants have normal embryonic/early larval melanophores , but profound deficiencies in their complements of metamorphic melanophores . picasso encodes the neuregulin receptor erbb3b , which acts both autonomously and non-autonomously to the metamorphic melanophore lineage . Pharmacological inhibition of ErbB signaling further revealed that erbb3b activity is required during neural crest migration for the later development of metamorphic melanophores , suggesting this locus is essential for establishing a pool of precursors that will differentiate only later during the larval-to-adult transformation [43] . By contrast , puma encodes tubulin alpha 8-like 3a ( tuba8l3a ) and acts autonomously to the metamorphic melanophore lineage . The temperature sensitivity of this allele allowed the identification of a critical period during pigment pattern metamorphosis , suggesting a role in maintaining or expanding a population of latent precursors , or recruiting these cells as melanophores [36] , [44] , [45] . To date it has not been known where latent precursors to metamorphic melanophores reside , how erbb3b , tuba8l3a or other loci promote the normal morphogenesis and differentiation of these cells and their progeny , or whether pigment cell precursors have indefinite or more limited re-population potential . Here we investigate these issues using molecular marker analyses , transgenesis , vital labeling , and time-lapse imaging in wild-type and mutant backgrounds . We show that during post-embryonic development , proliferative pigment cell precursors are associated with peripheral nerves and ganglia , and migrate to the hypodermis during pigment pattern metamorphosis where they differentiate as melanophores and iridophores . Nerve-associated pigment cell precursors are missing or reduced in ErbB-deficient and tuba8l3a mutant backgrounds . By contrast , these precursors persist in other mutants having less severe metamorphic melanophore deficiencies , though their subsequent development is marked both by defects , and partial regulation , of morphogenesis and differentiation . Finally , we show that latent precursors persist into the adult but that different precursor pools have different regenerative potentials . These findings provide a critical context for understanding the cellular bases of adult melanophore development , the mechanistic underpinnings of mutant phenotypes , and the roles for latent precursors in adult homeostasis , regeneration , and neoplasia .
To identify tissues that might harbor latent precursors to adult melanophores , we examined post-embryonic zebrafish for transcripts expressed by embryonic neural crest cells , reasoning that some of the cells expressing such markers might comprise a population of undifferentiated melanophore precursors . We examined foxd3 and sox10 , which are expressed by early neural crest cells , subpopulations of neural crest-derived glia , and some other cell types [46]–[49] , as well as crestin , which is known only to be expressed by neural crest cells and their derivatives [50] . Cells expressing these loci were present in the hypodermis where the adult pigment pattern forms , but also in the myotomes , adjacent to the spinal cord , and at the bases of the fins ( Figure 1A–1G ) , raising the possibility of both hypodermal and “extra-hypodermal” precursors for metamorphic melanophores . If extra-hypodermal , post-embryonic cells expressing genes typical of early neural crest cells contribute to metamorphic melanophores , we hypothesized that some of these cells should differentiate if supplied with appropriate trophic support . To test this idea , we used a heat-shock inducible transgenic line , Tg ( hsp70::kitla ) wp . r . t2 , to misexpress the melanogenic factor kit ligand-a ( kitla ) [51] throughout the larva ( Figure 2A , 2B ) . These larvae developed ectopic melanophores deep within the myotomes , which were never found in identically treated siblings lacking the transgene ( Figure 2C–2G ) . Previously we showed that erbb3b is essential for establishing precursors to metamorphic melanophores [43] . Accordingly , if extra-hypodermal kitla-responsive melanogenic cells and metamorphic melanophores arise from a common precursor pool , then ectopic melanophores should fail to develop in erbb3b mutants misexpressing kitla . Indeed , erbb3b; Tg ( hsp70:kitla ) larvae exhibited ∼30-fold fewer ectopic melanophores than wild-type Tg ( hsp70::kitla ) larvae ( Figure 2H ) . A similar outcome was observed when wild-type larvae were treated with the ErbB inhibitor , AG1478 , during the embryonic critical period for erbb3b activity in establishing metamorphic melanophore precursors [43] . Together , these results indicated that some extra-hypodermal cells are competent to differentiate as melanophores in the wild-type , but that most of these cells are missing when erbb3b activity is lost . Our identification of extra-hypodermal cells expressing genes typical of early neural crest and glial cells , as well as kitla-responsive melanogenic cells within the myotomes , led us to ask whether any of these cells embark upon a melanophore differentiation program during normal post-embryonic development . Since precursors to metamorphic melanophores require erbb3b , we further predicted that any candidate extra-hypodermal precursors of these cells should be missing in larvae deficient for erbb3b activity . In the wild-type , we found extra-hypodermal cells expressing mitfa , encoding a master regulator of melanophore fate specification ( Figure 1H–1J ) [52] , [53] . Cells expressing mitfa in extra-hypodermal locations typically did so at lower levels than cells within the hypodermis and were at the limit of detection given the sensitivity of in situ hybridization at post-embryonic stages . However , we also identified extra-hypodermal cells expressing GFP driven by the proximal mitfa promoter in the transgenic line , Tg ( mitfa::GFP ) w47 , which faithfully recapitulates mitfa expression in the melanophore lineage and in bipotent precursors to melanophores and iridophores in the embryo [54] , [55] ( Figure 3A; Videos S1 , S2 ) . In contrast to the wild-type , mitfa::GFP+ cells were largely absent from both erbb3b mutants and wild-type larvae treated with AG1478 during the embryonic erbb3b critical period ( Figure 3B and see below; Video S3 ) . In neither genetic background could we detect extra-hypodermal cells expressing transcript for dct , encoding a melanin synthesis enzyme . The many extra-hypodermal mitfa::GFP+ cells in larvae contrasts with embryogenesis [54] and suggests that extra-hypodermal cells may be specified for the melanophore lineage during the larval-to-adult transformation . If so , we predicted that cells expressing markers typical of early neural crest cells ( or glia ) should occassionally be found to express mitfa::GFP as well . We therefore examined larvae for simultaneous expression of mitfa::GFP , sox10 , and foxd3 , and , to learn where these cells reside , we examined larvae with cell-type specific markers for surrounding tissues . These analyses revealed numerous mitfa::GFP+ cells associated with peripheral nerves and ganglia , including the doral root ganglia , ventral motor root fibers , lateral line nerve , and nerve fibers coursing through the myotomes ( Figure 3E–3H; Figure S1 ) . Ectopic melanophores within the myotomes of wild-type Tg ( hsp70:kitla ) larvae were likewise nerve-associated ( Figure S2 ) . Double label analyses with markers of neural crest , glial , and melanophore lineages revealed that , in wild-type larvae , mitfa::GFP+ cells were often in close proximity to cells expressing sox10 and foxd3 , and also co-expressed sox10 , as expected given the direct regulation of mitfa by sox10 ( Figure 3D , 3G , 3H ) [56] , [57] . We also found that 4–15% of mitfa::GFP+ cells co-expressed foxd3 , with doubly labeled cells occurring most frequently during early metamorphosis , when the rate of melanophore population increase is maximal [36] ( Figure 3F , 3I; Figure 4A , 4B ) . This frequency of double labeling is reminiscent of 15–18 h embryos , in which 9–12% of mitfa::GFP+ cells are foxd3+ [54] . In contrast to the co-labeling of mitfa::GFP and foxd3 , we did not find mitfa::GFP expression by myelinating glia expressing myelin basic protein ( mbp ) ( Figure 3E; Figure S1 ) . As anticipated , however , some cells expressing foxd3 or sox10 co-expressed mbp . All of these cell types were deficient in erbb3b mutants and wild-type larvae treated with AG1478 during the embryonic erbb3b critical period ( Figure 3J–3M; Figure 4A , 4B; Figure S4 ) , though a few residual mitfa::GFP+ and foxd3+ cells occurred in anterior and posterior regions , corresponding to axial levels where a few residual melanophores develop during metamorphosis [43] ( data not shown ) . If some foxd3+ and sox10+ cells are progenitors to post-embryonic melanoblasts , we predicted that a period of population expansion could precede the appearance of mitfa::GFP+ cells , which might themselves be proliferative . Consistent with this idea , we found EdU incorporation by all three cell types but EdU-labeling of post-embryonic foxd3+ and sox10+ cells was most frequent prior to pigment pattern metamorphosis , whereas EdU-labeling of mitfa::GFP+ cells was most frequent during the peak of pigment pattern metamorphosis ( Figure 4C , Figure 5 ) . Given these findings , we further asked if doubly labeled foxd3+; mitfa::GFP+ cells constitute an especially proliferative population . These analyses revealed EdU-incorporation in 55% of foxd3+; mitfa::GFP+ cells , a significantly higher frequency than for cells expressing only foxd3 ( 17% EdU+ ) or only mitfa::GFP ( 24% EdU+; χ2 = 131 . 7 , d . f . = 2 , P<0 . 0001 ) . The frequency of EdU labeling amongst foxd3+; mitfa::GFP+ cells did not vary significantly across stages ( χ2 = 2 . 3 , d . f . = 2 , P = 0 . 3 ) . Finally , in erbb3b mutants sampled at selected stages , we found lower levels of EdU incorporation than in wild-type , though small numbers of cells overall resulted in correspondingly low statistical power ( Figure 4C; Figure S4 ) . The foregoing analyses indicated that , during post-embryonic development , a proliferative population of extra-hypodermal , erbb3b-dependent foxd3+ and sox10+ cells associated with peripheral nerves and ganglia becomes specified as precursors to melanophores ( or as bipotential precursors to melanophores and iridophores; see reference [55] ) . The development of post-embryonic , extra-hypodermal mitfa::GFP+ cells suggested the possibility that such cells migrate to the hypodermis during metamorphosis . To test this idea , we injected DiI into the myotomes , the horizontal myoseptum , or the base of the dorsal fin of wild-type or Tg ( mitfa::GFP ) larvae and we assessed after ≥4 d whether DiI-labeled cells were present within the hypodermis distant from the injection sites . In 10–30% of injected larvae we found hypodermal DiI-labeled melanized cells , mitfa::GFP+ cells , or iridophores ( Figure 6 ) , indicating that some extra-hypodermal cells migrated to the hypodermis during metamorphosis . To further test the hypothesis that extra-hypodermal cells contribute to the hypodermal melanophore population , we examined cell behaviors by time-lapse imaging of trunks derived from Tg ( mitfa::GFP ) larvae . Movies revealed the differentiation of mitfa::GFP+ cells into melanophores as well as their migration ( Figure 7A; Videos S4 , S5 , S6 ) . We therefore assessed the migratory pathways by which mitfa::GFP+ cells had reached the hypodermis . Approximately half of all mitfa::EGFP+ cells arrived within the hypodermis during imaging . Some entered the hypodermis after migrating over the dorsal or ventral margins of the myotomes , whereas others originated from within the myotomes , emerging either from the vicinity of the horizontal myoseptum or along vertical myosepta ( Figure 7B–7D , Figure 8; Videos S7 , S8 , S9 , S10 ) . Movies also revealed the movement of mitfa::GFP+ cells along nerves and their detachment from nerves to migrate more broadly through the fish ( Videos S11 , S12 ) . Together then , DiI labeling and time-lapse imaging indicated that extra-hypodermal cells contribute to hypodermal mitfa::GFP+ cells , melanophores , and iridophores during the larval-to-adult transformation . Our findings suggested that a normal complement of adult melanophores depends on contributions from a pool of extra-hypodermal precursors . If this is the case , we predicted that mutants with severe metamorphic melanophore deficiencies should have correspondingly severe deficiencies of extra-hypodermally derived mitfa:GFP+ cells . To test the contributions of extra-hypodermal cells to hypodermal mitfa::GFP+ cells and melanophores , we crossed the Tg ( mitfa::GFP ) transgene into erbb3b and tuba8l3a mutants , which exhibit severely reduced numbers of metamorphic melanophores [36] , [43] , [44] . In comparison to the wild-type , and as predicted from the foregoing analyses , erbb3b mutants had dramatically fewer extra-hypodermally derived mitfa::GFP+ cells ( Figure 9A; Video S13 ) . erbb3b mutant mifa::GFP+ cells originated from the vicinity of the horizontal myoseptum ( Figure 9B ) , and once in the hypodermis , these cells were more likely to differentiate and to divide ( Figure 9C; in contrast to the somewhat reduced rates of EdU incorporation prior to reaching the hypodermis shown in Figure 4C ) . tuba8l3a mutants also had significantly fewer extra-hypodermally derived mitfa::GFP+ cells . These cells were more likely to differentiate , but divided at only one-third the frequency of the wild-type ( Figure 9; Video S14 ) . tuba8l3a mutants exhibit a post-embryonic demyelination of the peripheral nervous system [44] , and we found that regions exhibiting mbp+ glial deficiencies and peripheral nerve defasciculation had fewer mitfa::GFP+ and foxd3+ cells ( Figure 10 ) . We did not observe cells doubly labeled for foxd3 and mitfa::GFP in tuba8l3a mutants . Amongst the erbb3b- and tuba8l3a-dependent metamorphic melanophore populations , are temporally and genetically distinct subpopulations , comprising early metamorphic melanophores that are initially dispersed but later migrate into stripes , and late metamorphic melanophores that develop already at sites of stripe formation [37] , [43] , [45] ( Figure S5 ) . Early metamorphic melanophores are ablated in kita mutants , but persist in colony stimulating factor-1 receptor ( csf1r ) and endothelin receptor b1 ( ednrb1 ) mutants . By contrast , late metamorphic melanophores persist in kita mutants , but are ablated in csf1r and ednrb1 mutants [37] , [38] , [58] , [59] . To test if these differences reflect differential persistence of distinct precursor pools , or differences in the subsequent morphogenesis and differentiation of cells arising from a common precursor pool , we crossed the Tg ( mitfa::GFP ) transgene into kitab5 , csf1rj4e1 and ednrb1b140 mutants and examined the origins of hypodermal mitfa::GFP+ cells as well as their frequencies of differentiation , death and proliferation . In contrast to erbb3b and tuba8l3a mutants , kita , csf1r , and ednrb1 mutants did not exhibit significantly fewer extra-hypodermally derived mitfa::GFP+ cells than the wild-type , though cells in kita mutants typically failed to differentiate and instead died at high frequency , whereas cells in csf1r and ednrb1 mutants were more likely both to differentiate and to die ( Figure 9; Video S15 ) . We did not observe gross defects in mitfa:GFP+ cell motility in any of the mutant backgrounds . Finally , in contrast to the proliferation of unmelanized mitf::GFP+ cells ( Figure 9C ) , proliferation of differentiated melanophores was rare in the wild-type ( 0 . 1%; N = 3822 melanophores observed ) and in most of the mutants ( 0 . 2–0 . 4%; N = 4358 melanophore observed ) . In kita mutants , however , the few melanophores that differentiated divided frequently ( 14%; N = 35 melanophores observed; variation among genotypes: χ2 = 38 . 1 , d . f . = 5 , P<0 . 0001; Video S16 ) . Together these data show that erbb3b and tuba8l3a each promote the development of extra-hypodermal mitfa::GFP+ cells , whereas all five loci promote the differentiation and morphogenesis of these cells once they reach the hypodermis . Our demonstration that extra-hypodermal precursors contribute to hypodermal melanophores led us to ask whether latent pigment cell precursors persist into the adult . Extra-hypodermal foxd3+ and mitfa::GFP+ cells were distributed in adult fish similarly to metamorphic stages and also were found associated with the scales ( Figure S6 ) . To test the capacity of latent precursors to supply new melanophores , we sought to ablate melanophores with the goal of provoking a regenerative response . Because fish doubly mutant for kitab5 and presumptive null alleles of csf1r lack body melanophores [38] , we reasoned that fish doubly mutant for kitab5 and the temperature-sensitive allele csf1rut . r1e174 ( csf1rTS ) [60] should have fewer melanophores ( equivalent to kitab5 single mutants ) at permissive temperature , but should lack all melanophores at restrictive temperature . Repeated exposure to restrictive and permissive temperatures should thus allow for repeated cycles of ablation and regeneration of these kita-independent , csf1r-dependent melanophores . As predicted , kita; csf1rTS double mutants that were initially indistinguishable from kita single mutants lost body melanophores when shifted to restrictive temperature ( Figure 11A , 11B ) . After returning to permissive temperature , fish initially recovered kita-independent hypodermal melanophores , though progressively fewer of these cells were regenerated in successive ablation–recovery cycles ( Figure 11C , 11D ) . Surprisingly , ablations also resulted in the de novo development and regeneration of scale melanophores , which are normally absent from kita mutants ( Figure 11F , 11G ) [37] . The few later hypodermal melanophores that were recovered in kita; csf1rTS mutants were often located beneath scales populated with melanophores , iridophores and xanthophores ( Figure 11H ) , raising the possibility that some of these regenerative hypodermal melanophores may have been scale-derived . Overall , these findings suggest that precursors to kita-independent , csf1r-dependent hypodermal melanophores persist in the adult yet have a finite regenerative potential , whereas an additional precursor pool associated with adult scales has a greater regenerative capability .
A major finding of our study is that post-embryonic mitfa::GFP+ cells are associated with peripheral nerves coursing through the myotomes as well as more medial nerves and ganglia . We further showed that nerve-associated cells could be induced to differentiate ectopically as melanophores , and that extra-hypodermal mitfa::GFP+ cells migrate to the hypodermis where some differentiate as melanophores during normal development . These observations suggest that peripheral nerves or ganglia serve as niches for post-embryonic precursors to adult melanophores and are broadly consistent with a recent study demonstrating a peripheral nerve origin for adult skin melanocytes of amniotes [62] as well as analyses revealing interconversion of glial and melanocyte fates in vitro [63]–[65] . Our study complements and extends recent lineage tracing studies of flounder larvae , in which adult pigment cell precursors were found to migrate to the hypodermis from dorsal and ventral regions during pigment pattern metamorphosis [66] , [67] . Our analyses also provide insights into the molecular and proliferative phenotypes of metamorphic melanophore precursors . foxd3 often acts as a transcriptional repressor and is associated with the maintenance of pluripotency and pluripotent cells [68]–[71] . In the neural crest lineage , foxd3 is expressed by pluripotent cells and presumptive glia , and can inhibit mitfa transcription , favoring iridophore or glial over melanogenic fates [54] , [55] , [61] , [72] . We found that some nerve-associated foxd3+ cells co-expressed mitfa::GFP just before and continuing through pigment pattern metamorphosis , and that such cells were especially likely to have incorporated EdU . These observations raise the possibility that nerve-associated foxd3+ cells are a pluripotent and proliferative population that can give rise to hypodermal melanophores and iridophores during pigment pattern metamorphosis . Our finding that DiI-labeled , extra-hypodermal tissues give rise to melanophores and iridophores with equal frequency is consistent with this idea . Because the mitfa::GFP transgene we employed is repressible by foxd3 , we speculate that co-expression of mitfa::GFP reflects low levels of perduring foxd3 protein as precursors adopt a pigmentary fate , similar to observations of early neural crest morphogenesis [54] , or that a balance between the anti-melanogenic and pro-melanogenic effects of foxd3 and mitfa , respectively , prevents specified cells from differentiating prematurely . Nevertheless , we note that our analyses revealed more extra-hypodermal mitfa::GFP+ cells than mitfa+ cells , which differs from the one-to-one corrrespondence of such cells in embryos [54] . Additional mitfa::GFP+ cells could reflect low levels of mitfa expression that fall below the threshold for detection by in situ hybridization at post-embryonic stages , yet are sufficient for accumulating detectable levels of relatively stable GFP . Or , mitfa::GFP expression in some cells could reflect a partial disregulation of the transgene , as might occur if regulatory elements for post-embryonic expression are missing . Distinguishing between these possibilities will require the production and analysis of additional transgenic reporter lines , but whichever the outcome , the mitfa::GFP reporter we have used will be a valuable tool for further dissecting the mechanisms of post-embryonic melanophore development . Examination of erbb3b and tuba8l3a mutants provides additional support for the idea that extra-hypodermal , nerve-associated precursors are essential for metamorphic melanophore development . We found that presumptive precursors to glia and pigment cells were missing from erbb3b mutants and wild-type larvae in which ErbB activity had been inhibited during the embryonic critical period for adult melanophore development . An on-going requirement for erbb3b is suggested as well , by reduced complements of adult melanophores following acute inhibition of ErbB signaling in sensitized backgrounds during pigment pattern metamorphosis [43] , and by reduced rates of EdU incorporation in erbb3b mutants during post-embryonic development ( this study ) . Because ErbB signaling can repress melanocyte differentiation [62] , [73] , a further role in preventing the premature differentiation of nerve-associated precursors to hypodermal pigment cells seems likely . Notably , our finding that extra-hypodermal , kitla-responsive melanogenic cells were missing in ErbB-deficient backgrounds at post-embryonic stages contrasts with increased numbers of such cells at embryonic/early larval stages [74] . This difference likely reflects the pleiotropic nature of ErbB signals and a difference between the stages examined: blocking ErbB activity in the embryo presumably results in a de-repression of melanophore differentiation amongst transiently persisting precursor cells , whereas by post-embryonic stages , such precursors presumably have been lost . In contrast to erbb3b mutants , tuba8l3a mutants exhibit a post-embryonic demyelination of peripheral nerves and a corresponding critical period for adult melanophore development [36] , [44] , [45] . In agreement with our model , tuba8l3a mutant larvae were deficient for mbp+ glia , nerve-associated foxd3+ cells and foxd3+; mitfa::GFP+ cells , and also had reduced rates of division amongst hypodermal mitfa::GFP+ cells . Because tuba8l3a acts autonomously to the metamorphic melanophore lineage [45] , these findings suggest a defect in the maintenance or expansion of pluripotent precursors to mbp+ glia and mitfa:GFP+ pigment cell precursors . The post-embryonic onset of these phenotypes further suggests the existence of genetically distinct populations of embryonic and adult glia , analogous to embryonic and metamorphic melanophores . Time-lapse imaging comparisons of wild-type and mutant backgrounds further defined roles for genes previously known to function in pigment pattern development: kita , csf1r , and ednrb1 all promoted the survival of mitfa::GFP+ cells whereas kita also promoted the differentiation of these cells as melanophores . Yet , these analyses revealed compensatory responses of pigment cells and their precursors in mutant backgrounds as well . For example , residual mitfa::GFP+ cells in mutants with extra-hypodermal precursor deficiencies ( erbb3b , tuba8l3a ) or hypodermal defects in cell survival ( csf1r , ednrb1 ) exhibited concomitantly greater rates of differentiation , and , in one instance ( erbb3b ) , an increased rate of proliferation . Moreover , the reduced survival and differentiation of mitfa::GFP+ cells in kita mutants was coupled with a 70-fold increase in melanophore proliferation . These findings highlight the remarkably regulative nature of zebrafish pigment pattern development [36] , [41] , [42] and also the importance of direct imaging for understanding cellular behaviors that would not be predicted from terminal phenotypes alone . A particularly dramatic example of pattern regulation occurs during regeneration . Zebrafish larval melanophores regenerate following laser ablation or the administration of melanocytotoxic drugs [74]–[76] , adult fin melanophores regenerate along with other tissues after fin amputation [77]–[79] , and hypodermal body melanophores regenerate after localized laser ablations [80] , [81] . To test the capacity of latent precursors to supply new hypodermal melanophores in the adult , we used fish doubly mutant for kita and csf1rTS in which loss of residual melanophores at restrictive temperature likely reflects the withdrawal of trophic support provided by csf1r-dependent xanthophores [42] , [60] . Our finding that progressively fewer hypodermal melanophores were recovered after repeated ablations implies a limited re-population potential for latent precursors that give rise to these kita-independent , csf1r-dependent late metamorphic melanophores . Whether the same is true of kita-dependent early metamorphic melanophores remains to be determined . Nevertheless , our finding that scale melanophores regenerated repeatedly-in contrast to late metamorphic , hypodermal melanophores-suggests a more highly regulative precursor pool associated with the adult scales , and highlights the possibility of spatially and temporally distinct pools of precursors having different morphogenetic and differentiative potentials . That scale melanophores developed de novo in these fish , despite their absence from kita single mutants , likely reflects a priority effect; e . g . , initially abundant xanthophores may repress melanophore development in kita mutants [42] , [82] , [83] but simultaneous development of both melanophores and xanthophores during regeneration allows for a stable pattern comprising both cell types . The similar re-population potential of scale and fin melanophores , and the previous observation that basonuclin-2 mutants , which are deficient for hypodermal melanophores , nevertheless retain both scale and fin melanophores [84] , also suggest the possibility of more extensive similarities between scale- and fin-associated precursor pools . Finally our study provides new insights into the temporally distinct populations of melanophores in zebrafish . We found severe deficiencies in the numbers of extra-hypodermally derived mitfa::GFP+ cells in erbb3b and tuba8l3a mutants , illustrating the critical role of such cells in supplying metamorphic melanophores overall . By contrast , we found no evidence for reduced numbers of extra-hypodermally derived mitfa::GFP+ cells in kita , csf1r , or ednrb1 mutants , indicating that different genetic requirements of early and late metamorphic melanophores do not reflect differences in the establishment or maintenance of these cells . These findings suggest either of two interpretations . Early and late metamorphic melanophores could arise from a common precursor pool with differences in residual melanophore complements among mutants reflecting specific requirements for kita , csf1r , and ednrb1 in downstream events of morphogenesis and differentiation . For example , our data indicate that mitfa::GFP+ cells require kita for their surival , and presumably , terminal differentiation: the development of late metamorphic melanophores in kita mutants could thus reflect the late appearance of factors able to substitute for kita activity in promoting melanophore differentiation . Conversely , we found that mitfa::GFP+ cells were only marginally dependent on csf1r: the failure of late metamorphic melanophores to develop in csf1r mutants could , in turn , reflect a late-arising , post-differentiation requirement for trophic support from csf1r-dependent xanthophores [38] , [42] , [60] . An alternative interpretation is that early and late metamorphic melanophores do arise from distinct , but still cryptic , precursor pools that simply were not revealed by our methods . In other systems , niches initially assumed to have just one type of stem or progenitor cell have sometimes been found to harbor distinct classes of cells with disparate proliferative or differentiative potentials [85]–[87] . Additional time-lapse imaging analyses at later stages and genetically based lineage analyses that are now being conducted should provide further insights into these possibilities . The identification of extra-hypodermal nerve-associated precursors to adult melanophores in zebrafish ( this study ) and amniotes [62] indicates that a fuller understanding of adult pigment cell development and pattern formation requires a focus on post-embryonic precursors , as distinct from embryonic neural crest cells . The existence of genetically separable populations of embryonic pigment cells , derived directly from neural crest cells , and adult pigment cells , derived from post-embryonic precursors , further suggests that species-differences in adult pigment patterns may be explicable by evolutionary changes in the establishment , maintenance or recruitment of post-embryonic latent precursors , with few if any consequences for earlier pigment patterns [88] . Lastly , a peripheral nerve origin for adult pigment cells also raises the possibility that the frequent and generally fatal metastases of melanoma cells to the central nervous system [26]–[29] , [89] , [90] may reflect the continued or reiterated expression of genes that favor proliferation and migration in a nerve microenvironment .
Fish were reared at 28–29°C , 14L:10D . Post-embryonic stages are reported as standardized standard length ( SSL ) measurements , which indicate developmental progress of free-feeding larvae more reliably than days post-fertilization [33] . Tg ( mitfa::GFP ) w47 and Tg ( −4 . 9sox10:egfp ) ba2 fish were generously provided by D . Raible and R . Kelsh , respectively . Tg ( TrDct::mCherry ) wp . r . t3 and Tg ( hsp70::kitla ) wp . r . t2 fish were produced using tol2kit Gateway vectors and Tol2-mediated transgenesis [91] , [92] and heat shocks with the latter strain were administered at 37°C for 1 hr three times daily for two days . Experiments with erbb3b mutants used either of two presumptive null alleles , erbb3but . r2e1 or erbb3bwp . r2e2 [93] . Experiments with csf1r mutants used either the presumptive null allele csf1rj4blue or the temperature-sensitive allele csf1rut . r1e174 [60] . In temperature shift experiments , fish were shifted repeatedly between restrictive temperature ( 33°C ) and permissive temperature ( 24°C ) . All experiments with the kita mutant used the presumptive null allele kitab5 [59] . tuba8l3aj115e1 encodes a missense substitution with temperature-sensitive effects . Experiments with this allele were performed at standard rearing temperature , intermediate between restrictive ( 33°C ) and permissive ( 24°C ) temperatures [36] , [44] , [45] , to allow analyses of a fuller complement of mitfa::GFP+ cells . Quantitative analyses here are thus likely to underestimate effects of the tuba8l3a mutation . Animal use conformed to University of Washington IACUC guidelines . Fish were viewed with Olympus SZX-12 or Zeiss Discovery epifluorescence stereomicroscopes or with a Zeiss Observer inverted compound epifluorescence microscope with Apotome . Images were collected in Axiovision software using Axiocam HR or MR3 cameras . For thick specimens , stacks of images were collected and processed using Zeiss Axiovision 6D Acquisition or Extended Focus modules and some fluorescence images were deconvolved using the Zeiss Axiovision Deconvolution module . Alternatively , specimens were viewed and images collected on Zeiss 510 META or Olympus FV1000 laser confocal microscopes . For immunohistochemistry , larvae were fixed in 4% paraformaldehyde containing 1% DMSO in PBS , rinsed , embedded in agarose , then sectioned by vibratome at 150–200 µm . Sections were washed in PBS/1% DMSO/0 . 3% Triton-X pH 7 . 4 ( PDTX ) , blocked in PDTX containing 10–20% heat inactivated goat serum then incubated overnight at 4°C with primary antibody . We used polyclonal antisera raised in rabbit against zebrafish sox10 ( 1:500; provided by B . Appel [94] ) , zebrafish foxd3 ( 1:400; D . Raible [47] ) , zebrafish mbp ( 1:50; W . Talbot [95] ) , and GFP ( 1:200; A11122 , Invitrogen ) as well as monoclonal antibodies against GFP ( 1:200; 3E6 A11120 , Invitrogen ) and acetylated α-tubulin ( 1:200; 6-11B-1 , T6793 Sigma ) . After washing , sections were incubated with secondary antibodies ( AlexaFluor 405 , 488 , 568 , 647; Invitrogen ) , washed and imaged . In situ hybridization of post-embryonic zebrafish followed [44] , [84] . For some analyses larvae were sectioned at 100–300 µm by vibratome . Detailed methods for in situ hybridization are available online at http://protist . biology . washington . edu/dparichy/ . Cell Tracker CM-DiI ( Invitrogen ) was prepared as a stock solution in DMSO then diluted to 0 . 025–0 . 05% in 0 . 3 M sucrose just before use . Larvae were anesthetized briefly and injected with 1–2 nl of DiI using a borosilicate needle , imaged immediately to ascertain the specificity of staining in target tissues , then reared individually until analyzed . Larvae were rinsed with 10% Hanks medium , anesthetized and then sacrificed by decapitation using a razor blade . After removing the anterior portion of the trunk and discarding the tails , larval trunks were placed on 0 . 4 µm transwell membranes ( Millipore ) in glass-bottom dishes containing L-15 medium , 3% fetal bovine serum , and penicillin/streptomycin . Trunks were equilibrated at 28 . 5°C for 3 h then imaged for 18–24 h ( 20 or 30 minute intervals between images ) on a Zeiss Observer inverted epifluorescence microscope . Comparisons between isolated trunks imaged continuously for 24 h and repeatedly anesthetized intact larvae did not reveal gross differences in the survival of mitfa::GFP+ cells , though average maximal estimated velocities of mitfa::GFP+ cells were reduced by ∼22% in cultured trunks as compared to intact larvae ( P<0 . 05; N = 67 cells examined ) . Imaging over longer durations resulted in increased rates of cell death throughout the explant and thus were not used for analyses . A stock solution of AG1478 [4- ( 3-chloroanilino ) -6 , 7-dimethoxyquinazoline; Calbiochem] was diluted in DMSO . Embryos were treated with 3 µM AG1478 in 10% Hanks for through either 72 or 96 hours post-fertilization . To facilitate penetration , 0 . 5% DMSO was added to all media and embryos were dechorionated prior to treatment . Fish were reared in glass Petri dishes and solutions were changed daily . Larvae were incubated with 0 . 005% 5-ethynyl-2′-deoxyuridine ( EdU; Intvitrogen ) in 10% Hank's medium containing 1% DMSO for 36 h . Larvae were then sacrificed , fixed with 4% PFA/1% DMSO , and vibrotome sectioned ( 150–200 µm ) for immunohistochemistry followed by histochemical detection of EdU according to manufacturer's recommendations . Quantitative data were analyzed with JMP 8 . 0 . 2 ( SAS Institute , Cary NC ) . Frequency data were examined using multiple logistic regression or contingency table analyses , and tested for effects of genotype , stage , and genotype x stage interactions . Significance of effects were assessed by likelihood ratio tests and non-significant factors were removed from the final models . Analyses of variance were used for continuous variables including counts . Residuals were examined for normality and homogeneity of variances , conditions that were achieved for some variables after transformation by square root or natural logarithm . Further details of statistical analyses are available upon request . | Understanding the biology of post-embryonic stem and progenitor cells is of both basic and translational importance . To identify mechanisms by which stem and progenitor cells are established , maintained , and recruited to particular fates , we are using the zebrafish adult pigment pattern . Previous work showed that embryonic and adult pigment cells have different genetic requirements , but little is known about the molecular or proliferative phenotypes of precursors to adult pigment cells or where these precursors reside during post-embryonic development . We show here that post-embryonic pigment cell precursors are associated with peripheral nerves and that these cells migrate to the skin during the larval-to-adult transformation when the adult pigment pattern forms . We also define morphogenetic and differentiative roles for several genes in promoting these events . Finally , we demonstrate that latent precursor pools persist into the adult and that different pools have different capacities for supplying new pigment cells in the context of pattern regeneration . Our study sets the stage for future analyses to identify additional common and essential features of pigment stem cell biology . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetic",
"mutation",
"cell",
"differentiation",
"gene",
"function",
"developmental",
"biology",
"organism",
"development",
"stem",
"cells",
"molecular",
"development",
"morphogenesis",
"adult",
"stem",
"cells",
"stem",
"cell",
"niche",
"biology",
"regeneration",
"signaling",
"cell",
"migration",
"genetics",
"genetics",
"and",
"genomics",
"cell",
"fate",
"determination"
] | 2011 | Post-Embryonic Nerve-Associated Precursors to Adult Pigment Cells:
Genetic Requirements and Dynamics of Morphogenesis and
Differentiation |
The activity of neural populations in the brains of humans and animals can exhibit vastly different spatial patterns when faced with different tasks or environmental stimuli . The degrees of similarity between these neural activity patterns in response to different events are used to characterize the representational structure of cognitive states in a neural population . The dominant methods of investigating this similarity structure first estimate neural activity patterns from noisy neural imaging data using linear regression , and then examine the similarity between the estimated patterns . Here , we show that this approach introduces spurious bias structure in the resulting similarity matrix , in particular when applied to fMRI data . This problem is especially severe when the signal-to-noise ratio is low and in cases where experimental conditions cannot be fully randomized in a task . We propose Bayesian Representational Similarity Analysis ( BRSA ) , an alternative method for computing representational similarity , in which we treat the covariance structure of neural activity patterns as a hyper-parameter in a generative model of the neural data . By marginalizing over the unknown activity patterns , we can directly estimate this covariance structure from imaging data . This method offers significant reductions in bias and allows estimation of neural representational similarity with previously unattained levels of precision at low signal-to-noise ratio , without losing the possibility of deriving an interpretable distance measure from the estimated similarity . The method is closely related to Pattern Component Model ( PCM ) , but instead of modeling the estimated neural patterns as in PCM , BRSA models the imaging data directly and is suited for analyzing data in which the order of task conditions is not fully counterbalanced . The probabilistic framework allows for jointly analyzing data from a group of participants . The method can also simultaneously estimate a signal-to-noise ratio map that shows where the learned representational structure is supported more strongly . Both this map and the learned covariance matrix can be used as a structured prior for maximum a posteriori estimation of neural activity patterns , which can be further used for fMRI decoding . Our method therefore paves the way towards a more unified and principled analysis of neural representations underlying fMRI signals . We make our tool freely available in Brain Imaging Analysis Kit ( BrainIAK ) .
Functional magnetic resonance imaging ( fMRI ) measures the blood-oxygen-level-dependent ( BOLD ) signals [1] , which rise to peak ∼6 seconds after neuronal activity increases in a local region [2] . Because of its non-invasiveness , full-brain coverage , and relatively favorable trade-off between spatial and temporal resolution , fMRI has been a powerful tool to study the neural correlates of cognition [3–5] . In the last decade , research has moved beyond simply localizing the brain regions selectively activated by cognitive processes and the focus has been increasingly placed on the relationship between the detailed spatial patterns of neural activity and cognitive processes [6 , 7] . An important tool for characterizing the functional architecture of the brain is representational similarity analysis ( RSA ) [8] . This classic method first estimates the neural activity patterns from fMRI data recorded as participants observe a set of stimuli or experience a set of task conditions , and then calculates the similarity ( e . g . , by Pearson correlation ) between each pair of the estimated patterns . The rationale is that if two stimuli are represented with similar codes in a brain region , the spatial patterns of neural activation in that region would be similar when processing these two stimuli . When using Pearson correlation as a similarity metric , the activity profile of each voxel to all the task conditions is essentially viewed as one independent sample from a multivariate normal distribution in a space spanned by the experimental conditions , which is characterized by its covariance matrix . Recently , it has been pointed out that RSA and two other approaches for understanding neural representational structure , namely encoding model [9] and pattern component modeling ( PCM ) [10] , are closely related through the second moment statistics ( the covariance matrix ) of the true ( unknown ) activity patterns [11] . After the similarity matrix between all pairs of estimated activity patterns is calculated in a region of interest ( ROI ) , it can be compared against similarity matrices predicted by candidate computational models . Researchers can also convert the similarity matrix into a representational dissimilarity matrix ( RDM , e . g . , 1 − C , for similarity C based on correlation ) and visualize the structure of the representational space in the ROI by projecting the dissimilarity matrix to a low dimensional space [8] . Researchers might also test whether certain experimental manipulations change the degrees of similarity between neural patterns of interest [12 , 13] . To list just a few applications from the field of visual neuroscience , RSA has revealed that humans and monkeys have highly similar representational structures in the inferotemporal ( IT ) cortex for images across various semantic categories [14] . It also revealed a continuum in the abstract representation of biological classes in human ventral object visual cortex [15] and that basic categorical structure gradually emerges through the hierarchy of visual cortex [16] . Because of the additional flexibility of exploring the structure of neural representation without building explicit computational models , RSA has also gained popularity among cognitive neuroscientists for studying more complex tasks beyond perception , such as decision making . While RSA has been widely adopted in many fields of cognitive neuroscience , a few recent studies have revealed that the similarity structure estimated by standard RSA might be confounded by various factors . First , the calculated similarity between two neural patterns strongly depends on the time that elapsed between the two measured patterns: the closer the two patterns are in time , the more similar they are [17] [18] . Second , it was found that because different brain regions share some common time courses of fluctuation independent of the stimuli being presented ( intrinsic fluctuations ) , RDMs between regions are highly similar when calculated based on patterns of the same trials of tasks but not when they are calculated based on separate trials ( thus the intrinsic fluctuation are not shared across regions ) . This indicates that RSA can be strongly influenced by intrinsic fluctuation [17] . Lastly , Diedrichsen et al . ( 2011 ) pointed out that the noise in the estimated activity patterns can add a diagonal component to the condition-by-condition covariance matrix of the spatial patterns . This leads to over-estimation of the variance of the neural pattern and underestimation of correlation between true patterns , and this underestimation depends on signal-to-noise ratio in each ROI , making it difficult to make comparison of RDMs between regions [10] . Recognizing the first two issues , several groups have recently suggested modifications to RSA such as calculating similarity or distance between activity patterns estimated from separate fMRI runs [18 , 19] , henceforth referred to as cross-run RSA , and using a Taylor expansion to approximate and regress out the dependency of pattern similarity on the interval between events [18] . For the last issue , Diedrichsen et al . ( 2011 ) proposed PCM which models the condition-by-condition covariance matrix between estimated neural patterns as the sum of a diagonal component that reflects the contribution of noise in the estimated neural patterns to the covariance matrix and components reflecting the researcher’s hypothetical representational structure in the ROI [10] . These methods improve on traditional RSA , but are not explicitly directed at the source of the bias , and therefore only offer partial solutions . Indeed , the severity of confounds in traditional RSA is not yet widely recognized . RSA based on neural patterns estimated within an imaging run is still commonly performed . Furthermore , sometimes a study might need to examine the representational similarity between task conditions within an imaging run , such that cross-run RSA is not feasible . The Taylor expansion approach to model the effect of event-interval can be difficult to set up when a task condition repeats several times in an experiment . There also lacks a detailed mathematical examination of the source of the bias and how different ways of applying RSA affect the bias . Researchers sometimes hold the view that RSA of raw fMRI patterns , instead of activity patterns ( β ) estimated through a general linear model ( GLM ) [20] , does not suffer from the confounds mentioned above . Last but not least , the contribution of noise in the estimated neural patterns to the sample covariance matrix between patterns may not be restricted to the diagonal elements , as we will demonstrate below . In this paper , we first compare the result of performing traditional RSA on a task-based fMRI dataset with the results obtained when performing the same analysis on white noise , to illustrate the severe bias and spurious similarity structure that can result from performing RSA on pattern estimates within imaging runs . By applying task-specific RSA on irrelevant resting-state fMRI data , we show that spurious structure also emerges when RSA is performed on the raw fMRI pattern rather than estimated task activation patterns . We observed that the spurious structure can be far from a diagonal matrix , and masks any true similarity structure . We then provide an analytic derivation to help understand the source of the bias in traditional RSA . Previously , we have proposed a method named Bayesian RSA ( BRSA ) , which significantly reduced this bias and allows analysis within imaging runs [21] . BRSA is related to PCM in the sense that they both treat the true and unknown activity profiles of each voxel as a sample from a multivariate normal distribution and marginalize the true activity pattern in their analysis . The critical difference is that PCM models the estimated activity patterns of each trial or task condition , in which complex spurious correlation structure could have already been introduced during the estimation , while BRSA directly models the raw imaging data . Here , we further extend BRSA to explicitly model spatial noise correlation , thereby mitigating the second issue identified by Heriksson et al . [17] , namely the intrinsic fluctuation not modelled by task events in an experiment . Furthermore , inspired by the methods of hyper-alignment [22] and shared response models [23] , we extend our method to learn a shared representational similarity structure across multiple participants ( Group BRSA ) and demonstrate improved accuracy of this approach . Since our method significantly reduces bias in the estimated similarity matrix but does not fully eliminate it at regimes of very low signal-to-noise ratio ( SNR ) , we further provide a cross-validation approach to detecting over-fitting to the data . Finally , we show that the learned representational structure can serve as an empirical prior to constrain the posterior estimation of activity patterns , which can be used to decode the cognitive state underlying activity observed in new fMRI data . The algorithm in this paper is publicly available in the Python package Brain Imaging Analysis Kit ( BrainIAK ) , under the brainiak . reprsimil . brsa module . Our previous version of Bayesian RSA method [21] with newly added modeling of spatial noise correlation is in the BRSA class of the module . The new version described in this paper is implemented in the GBRSA class and can be applied to either a single participant or a group of participants .
Traditional RSA [8] first estimates the response amplitudes ( β ) of each voxel in an ROI to each task condition , and then calculates the similarity between the estimated spatial response patterns of that ROI to each pair of task conditions . The estimation of β is based on a GLM . We denote the fMRI time series from an experiment as Y ∈ R n T × n V , with nT being the number of time points and nV the number of voxels . The GLM assumes that Y = X · β + ϵ . ( 1 ) X ∈ R n T × n C is the “design matrix , ” where nC is the number of task conditions . Each column of the design matrix is constructed by convolving a hemodynamic response function ( HRF ) with a time series describing the onsets and duration of all events belonging to one task condition . The regressors composing the design matrix express the hypothesized response time course elicited by each task condition . Each voxel’s response amplitudes to different task conditions can differ . The response amplitudes of one voxel to all conditions forms that voxel’s response profile . All voxels’ response profiles form a matrix of spatial activity patterns β ∈ R n C × n V , with each row representing the spatial pattern of activity elicited by one task condition . The responses to all conditions are assumed to contribute linearly to the spatio-temporal fMRI signal through the temporal profile of hemodynamic response expressed in X . Thus , the measured Y is assumed to be a linear sum of X weighted by response amplitudes β , corrupted by zero-mean noise ϵ . The goal of RSA is to understand the degree of similarity between each pair of spatial response patterns ( i . e . , between the rows of β ) . But because the true β is not accessible , a point estimate of β , derived through linear regression , is usually used as a surrogate: β ^ = ( X T X ) - 1 X T Y ( 2 ) Similarity is then calculated between rows of β ^ . For instance , one measure of similarity that is frequently used is Pearson correlation . The similarity between patterns of condition i and j is assessed as C i j = ( β ^ i - β ^ i ¯ ) ( β ^ i - β ^ j ¯ ) T n V σ β ^ i σ β ^ j ( 3 ) where β ^ i ¯ and σ β ^ i are the mean and standard deviation of the estimated pattern of condition i across voxels . To demonstrate the spurious structure that may appear in the result of traditional RSA , we first performed RSA on the fMRI data in one ROI , the orbitofrontal cortex , in a previous dataset involving a decision-making task [24] . The task included 16 different task conditions , or “states . ” In each state , participants paid attention to one of two overlapping images ( face or house ) and made judgments about the image in the attended category . The transition between the 16 task states followed the Markov chain shown in Fig 1A , thus some states often preceded certain other states . The 16 states could be grouped into 3 categories according to the structure of transitions among states ( the exact meaning of the states , or the 3 categories , are not important in the context of the discussion here . ) We performed traditional RSA on the 16 estimated spatial response patterns corresponding to the 16 task states . To visualize the structure of the neural representation of the task states in the ROI , we used multi-dimensional scaling ( MDS ) [25] to project the 16-dimensional space defined by the distance ( 1—correlation ) between states onto a 3-dimensional space ( Fig 1B ) . This projection appears to show clear grouping of the states in the orbitofrontal cortex consistent with the 3 categories , suggesting that this brain area represent this aspect of the task . However , a similar representational structure was also observed in other ROIs . In addition , when we applied the same GLM to randomly generated white noise and performed RSA on the resulting parameter estimates , the similarity matrix closely resembled the result found in the real fMRI data ( Fig 1C ) . Since there is no task-related activity in the white noise , the structure obtained from white noise is clearly spurious and must reflect a bias introduced by the analysis . In fact , we found that the off-diagonal structure obtained from white noise ( Fig 1C ) explained 84 ± 12% of the variance of the off-diagonals obtained from real data ( Fig 1B ) . This shows that the bias introduced by traditional RSA can dominate the result , masking the real representational structure in the data . To help understand this observation , we provide an analytic derivation of the bias with a few simplifying assumptions [21] . The calculation of the sample correlation of β ^ in traditional RSA implies the implicit assumption that an underlying covariance structure exists that describes the distribution of β , and the activity profile of each voxel is one sample from this distribution . Therefore , examining the relation between the covariance of β ^ and that of true β will help us understand the bias in traditional RSA . We assume that a covariance matrix U ( of size nC × nC ) captures the true covariance structure of β across all voxels in the ROI: β ∼ N ( 0 , U ) . Similarity measures such as correlation are derived from U by normalizing the diagonal elements to 1 . It is well known that temporal autocorrelation exists in fMRI noise [26 , 27] . To capture this , we assume that in each voxel ϵ ∼ N ( 0 , Σϵ ) , where Σ ϵ ∈ R n T × n T is the temporal covariance of the noise ( for illustration purposes , here we assume that all voxels have the same noise variance and autocorrelation , and temporarily assume the noise is spatially independent ) . By substituting the expression for Y from Eq ( 1 ) into the point estimate of β ( 2 ) , we obtain β ^ = ( X T X ) - 1 X T X β + ( X T X ) - 1 X T ϵ = β + ( X T X ) - 1 X T ϵ ( 4 ) which means the point estimate of β is contaminated by a noise term ( XT X ) −1 XTϵ . Assuming that the signal β is independent from the noise ϵ , it is then also independent from the linear transformation of the noise , ( XT X ) −1 XTϵ . Thus the covariance of β ^ is the sum of the covariance of true β and the covariance of ( XT X ) −1 XTϵ: β ^ ∼ N ( 0 , U + ( X T X ) - 1 X T Σ ϵ X ( X T X ) - 1 ) ( 5 ) The term ( XT X ) −1 XT∑ϵ X ( XT X ) −1 is the source of the bias in RSA . This bias originates from the structured noise ( XT X ) −1 XTϵ in estimating β ^ . It depends on both the design matrix X and the temporal autocorrelation of the noise ϵ . Fig 1F illustrates how structured noise can alter the correlation of noisy pattern estimates in a simple case of just two task conditions . Even if we assume the noise is temporally independent ( i . e . , Σϵ is a diagonal matrix , which may be a valid assumption if one “pre-whitens” the data before further analysis [27] ) , the bias structure still exists but reduces to ( XT X ) −1 σ2 , where σ2 is the variance of the noise . Since the covariance matrix of β ^ is biased , its correlation is also distorted from the true correlation structure . This is because correlation is merely a rescaling of rows and columns of a covariance matrix . Fig 1C essentially illustrates this bias structure after being converted to correlation matrix ( in this case , σ = 1 and β = 0 ) as this RSA structure , by virture of being derived for white noise , can only result from structure in the design matrix X . In reality , both spatial and temporal correlations exist in fMRI noise , which complicates the structure of the bias . But the fact that bias in Fig 1C arises even when applying RSA to white noise which itself has no spatial-temporal correlation helps to emphasize the first contributor to the bias: the timing structure of the task , which is exhibited in the correlations between the regressors in the design matrix . Whenever the interval between events of two task conditions is shorter than the length of the HRF ( which typically outlasts 12 s ) , correlation is introduced between their corresponding columns in the design matrix . The degree of correlation depends on the overlapping of the HRFs . If one task condition often closely precedes another , which is the case here as a consequence of the Markovian property of the task , their corresponding columns in the design matrix are more strongly correlated . As a result of these correlations , XT X is not a diagonal matrix , and neither is its inverse ( XT X ) −1 . In general , unless the order of task conditions is very well counterbalanced and randomized across participants , the noise ( XT X ) −1 XTϵ in β ^ is not i . i . d between task conditions . The bias term B = ( XT X ) −1 XT Σϵ X ( XT X ) −1 then deviates from a diagonal matrix and causes unequal distortion of the off-diagonal elements in the resulting correlation matrix of β ^ . These unequal distortions alter the order of ranking of the values of the off-diagonal elements . Therefore , rank correlation between the similarity matrix from traditional RSA and the similarity matrix of any candidate computational model is necessarily influenced by the bias . Conclusion based on such comparison between two similarity matrices or based on comparing a pair of off-diagonal elements within a neural similarity matrix becomes problematic , as long as the bias causes unequal distortion . Furthermore , if the design matrices also depend on participants’ performance such as errors and reaction time , the bias structure could depend on their performance as well . Comparison between neural representational structure and participants’ behavioral performance may also become problematic in such situations . It is worth pointing out that the bias is not restricted to using correlation as metric of similarity . Because structured noise exists in β ^ , any distance metrics between rows of β ^ estimated within imaging runs of fMRI data are likely biased . We can take Euclidean distance as an example . For any two task conditions i and j , the expectation of the distance between β i ^ and β j ^ is ∑ k = 1 n V ( β i k - β j k ) 2 + n V ( B i i 2 + B j j 2 - 2 B i j 2 ) , where B is the bias in the covariance structure . Therefore , the bias n V ( B i i 2 + B j j 2 - 2 B i j 2 ) in Euclidean distance also depends on the task timing structure and the property of noise . ( See Fig 1D ) . In our derivations above , point estimates of β ^ introduce structured noise due to the correlation structure in the design matrix . One might think that the bias can be avoided if a design matrix is not used , i . e . , if RSA is not performed after GLM analysis , but directly on the raw fMRI patterns . Such an approach still suffers from bias , for two reasons that we detail below . First , RSA on the raw activity patterns suffers from the second contributor to the bias in RSA that comes from the temporal properties of fMRI noise . To understand this , consider that estimating activity pattern by averaging the raw patterns , for instance 6 sec after each event of a task condition ( that is , at the approximate peak of the event-driven HRF ) is equivalent to performing an alternative GLM analysis with a design matrix X6 that has delta functions 6 sec after each event . Although the columns of this design matrix X6 are orthogonal and ( X 6 T X 6 ) - 1 becomes diagonal , the bias term is still not a diagonal matrix . Because of the autocorrelation structure Σϵ in the noise , the bias term ( X 6 T X 6 ) - 1 X 6 T Σ ϵ X 6 ( X 6 T X 6 ) - 1 essentially becomes a sampling of the temporal covariance structure of noise at the distances of the inter-event intervals . In this way , timing structure of the task and autocorrelation of noise together still cause bias in the RSA result . To illustrate this , we applied RSA to the raw patterns of an independent set of resting state fMRI data from the Human Connectome Project [28] , pretending that the participants experienced events according to the 16-state task in Fig 1A . As shown in Fig 1E , even in the absence of any task-related signal spurious similarity structure emerges when RSA is applied to the raw patterns of resting state data . We then calculated the theoretical bias structure ( X 6 T X 6 ) - 1 X 6 T Σ ϵ X 6 ( X 6 T X 6 ) - 1 for each task sequence based on X6 of that sequence and Σϵ estimated as the average noise temporal correlation matrix of the resting state data of three other participants ( right figure of Fig 1E ) . The off-diagonal elements of all the similarity matrices based on raw patterns were significantly correlated with the theoretical bias structure ( the largest Bonferroni-corrected p-value of Pearson correlation is 0 . 0007 ) and 51 ± 18% of the variance in the off-diagonal elements can be explained by the theoretical bias . Second , averaging raw data 6 sec after events of interest over-estimates the similarity between neural patterns of adjacent events , an effect independent of the fMRI noise property . This is because the true HRF in the brain has a protracted time course regardless of how one analyzes the data . Thus the estimated patterns ( we denote by β ^ 6 ) in this approach are themselves biased due to the mismatch between the implicit HRF that this averaging assumes and the real HRF . The expectation of β ^ 6 becomes E [ β ^ 6 ] = E [ ( X 6 T X 6 ) - 1 X 6 T Y ] = E [ ( X 6 T X 6 ) - 1 X 6 T ( X β + ϵ ) ] = ( X 6 T X 6 ) - 1 X 6 T X β instead of β . Intuitively , X temporarily smears the BOLD patterns of neural responses close in time but ( X 6 T X 6 ) - 1 X 6 T only averages the smeared BOLD patterns without disentangling the smearing . β ^ 6 thus mixes the BOLD activity patterns elicited by all neural events within a time window of approximately 12 sec ( the duration of HRF ) around the event of interest , causing over-estimation of the similarity between neural patterns of adjacent events . If the order of task conditions is not fully counterbalanced , this method would therefore still introduce into the estimated similarity matrix a bias caused by the structure of the task . Similar effect can also be introduced if β ^ is estimated with regularized least square regression [29] . Regression with regularization of the amplitude of β ^ trades off bias in the estimates for variance ( noise ) . On the surface , reducing noise in the pattern estimates may reduce the bias introduced into the similarity matrix . However , the bias in β ^ itself alters the similarity matrix again . For example , in ridge regression , an additional penalization term λβT β is imposed for β of each voxel . This turns estimates β ^ to β ^ = ( X T X + λ I ) - 1 X T Y . The component contributed to β ^ by the true signal Xβ becomes ( XT X + λI ) −1 XTXβ . As λ increases , this component increasingly attributes neural activity triggered by other task events near the time of an event of interest to this event’s activity . Therefore , this method too would overestimate pattern similarity between adjacent events . In all the derivations above , we have assumed for simplicity of illustration that the noise in all voxels has the same temporal covariance structure . In reality , the autocorrelation can vary over a large range across voxels ( Fig 1G ) . So the structured noise in each voxel would follow a different distribution . Furthermore , the spatial correlation in noise means the noise in β ^ is also correlated across voxels . Because noise correlation between voxels violates the assumption of Pearson correlation that observations ( activity profiles of different voxels ) are independent , the p-values associated with the correlation coefficients will not be interpretable . Although we made these simplified assumption for ease of illustration , in the model development below , variation of auto-correlation across voxels and spatial noise correlation are both considered in our proposed method . As shown above , the covariance structure of the noise in the point estimates of neural activity patterns β ^ leads to bias in the subsequent similarity measures . The bias can distort off-diagonal elements of the resulting similarity matrix unequally if the order of task conditions is not fully counterbalanced . In order to reduce this bias , we propose a new strategy that aims to infer directly the covariance structure U that underlies the similarity of neural patterns , using raw fMRI data . Our method avoids estimating β ^ altogether , and instead marginalizes over the unknown activity patterns β without discarding uncertainty about them . The marginalization avoids the structured noise introduced by the point estimates , which was the central cause of the bias . Given that the bias comes not only from the experimental design but also from the spatial and temporal correlation in noise , we explicitly model these properties in the data . We name this approach Bayesian RSA ( BRSA ) as it is an empirical Bayesian method [30] for estimating U as a parameter of the prior distribution of β directly from data . Although Fig 3 shows that BRSA reduces bias , it does not eliminate it completely . This may be due to over-fitting to noise . Because it is unlikely that the time course of intrinsic fluctuation X0 and the design matrix X are perfectly orthogonal , part of the intrinsic fluctuation cannot be distinguished from task-related activity . Therefore , the structure of β0 , the modulation of intrinsic fluctuation , could also influence the estimated U ^ when SNR is low . For instance , in Fig 3F , at the lowest SNR and least amount of data ( top left subplot ) , the true similarity structure is almost undetectable using BRSA . Is this due to large variance in the estimates , or is it because BRSA is still biased , but to a lesser degree than standard RSA ? If the result is still biased , then averaging results across subjects will not remove the bias , and the deviation of the average estimated similarity structure from the true similarity structure should not approach 0 . To test this , we simulated many more subjects by preserving the spatial patterns of intrinsic fluctuation and the auto-regressive properties of the voxel-specific noise in the data used in Fig 3 , and generating intrinsic fluctuations that maintain the amplitudes of power spectrum in the frequency domain . To expose the limit of the performance of BRSA , we focused on the lower range of SNR and simulated only one run of data per “subject” . Fig 4A shows the quality of the average estimated similarity matrix with increasing number of simulated subjects . The average similarity matrices estimated by BRSA do not approach the true similarity matrix indefinitely as the number of subjects increase . Instead , their correlation saturates to a value smaller than 1 . This indicates that the result of BRSA is still weakly biased , with the bias depending on the SNR . It is possible that as the SNR approaches 0 , the estimated U ^ is gradually dominated by the impact of the part of X0 not orthogonal to X . This bias is not due to underestimating the number of regressors in X0 ( see Part 6 The effect of the number of nuisance regressors on BRSA performance of S1 Material ) . We leave investigation of the source of this bias to future work . Empirically , the algorithm [35] we use to estimate the number of regressors in X0 yields more stable and reasonable estimation than other methods we have tested ( e . g . , [36] ) . It should be noted that BRSA still performs much better than standard RSA , for which the correlation between the estimated similarity matrix and the true similarity matrix never passed 0 . 1 in these simulations . The expected bias structure when spatial noise correlation exists is difficult to derive . We used ( XT X ) −1 as a proxy to evaluate the residual bias in the estimated similarity using BRSA . As expected , when the SNR approached zero , the model over-fit to the noise and the bias structure increasingly dominated the estimated structure despite increasing the number of simulated participants ( Fig 4B ) . This observation calls for an evaluation procedure to detect over-fitting in applications to real data , when the ground truth of the similarity structure is unknown . One approach to assess whether a BRSA model has over-fit the noise is cross-validation . In addition to estimating U , the model can also estimate the posterior mean of all other parameters , including the neural patterns β of task-related activity , β0 of intrinsic fluctuation , noise variances σ2 and auto-correlation coefficients ρ . For a left-out testing data set , the design matrix Xtest is known given the task design . Together with the parameters estimated from the training data as well as the estimated variance and auto-correlation properties of the intrinsic fluctuation in the training data , we can calculate the log predictive probability of observing the test data . The unknown intrinsic fluctuation in the test data can be marginalized by assuming their statistical property stays unchanged from training data to test data . The predictive probability can then be contrasted against the cross-validated predictive probability provided by a null model separately fitted to the training data . The null model would have all the same assumptions as the full BRSA model , except that it would not assume any task-related activity captured by X . When BRSA over-fits the data , the estimated spatial pattern β ^ would not reflect the true response pattern to the task and is unlikely to be modulated by the time course in Xtest . Thus the full model would predict signals that do not occur in the test data , and yield a lower predictive probability than the null model . The result of the full BRSA model on training data can therefore be accepted if the log predictive probability by the full model is higher than that of the null model significantly more often than chance . Over-fitting might also arise when the assumed design matrix X does not correctly reflect task-related activity . When there is a sufficient amount of data but the design matrix does not reflect the true activity , the estimated covariance matrix U ^ in BRSA would approach zero , as would the posterior estimates of β ^ . In this case as well , the full model would perform worse than the null model , because the form of the predictive likelihood automatically penalize more complex models . We tested the effectiveness of relying on cross-validation to reject over-fitted results using the same simulation procedure as in Fig 3 , and repeated this simulation 36 times , each time with newly simulated signals and data from a new group of participants in HCP [37] as “noise” . Each such simulated group represents one replication study . The ROI to extract “noise” from HCP dataset and the region to add signals were the same as in Fig 3 . Fig 5A shows the rate of correct acceptance when both training and test data have signals . We counted each simulation in which the cross-validation score ( log predictive probability ) of the full BRSA model was significantly higher than the score of the null model ( based on a one-sided student’s t-test at a threshold of α = 0 . 05 ) as one incidence of correct acceptance . When the SNR is high ( at least 0 . 27 in the active voxels when there are two runs of training data , or at least 0 . 54 when there is one run of data ) , warranting reliable estimation of the similarity structures as indicated in Fig 3I , the cross-validation procedure selected the full model significantly more often than chance across simulation ( the highest p = 0 . 03 , binomial test with Bonferroni correction [38] ) . At low SNRs and with less training data ( SNR below 0 . 27 when there is only one run of data or at 0 . 14 when there are two runs of data ) , the full model was almost never selected ( p<9e-9 ) , although in some cases the true similarity structure is still visible in the result of Fig 3F . This indicates that the cross-validation procedure is relatively conservative . The means and standard deviations of the t-statistics across simulated groups for all simulation configurations are displayed in Fig 5B . The differences in cross-validation scores between full and null models are displayed in Fig 5C . The reason that the null model can have a higher cross-validation score than the full model at low SNR is not only because of potential overfitting of the full model to the noise , but also because the full model’s log likelihood function has an additional log determinant term due to the inclusion of β , which naturally penalizes for the extra complexity of the model compared to the null model ( see Eqs ( 12 ) , ( 13 ) and ( 15 ) in Materials and methods ) . The cross-validation procedure also helps avoid false acceptance when activity patterns are not consistently reproducible across runs . To illustrate this , we simulated the case when signals are only added to the training data but not to the test data . Now , the full model was always rejected across the simulated SNR and amounts of data . Finally , when neither training data nor testing data included signal , the cross-validation procedure also correctly rejected the full model in all cases . Fig 5D and 5E illustrate the difference between cross-validation scores of full and null models for the two simulations , respectively . The advantage of the null model was smaller with 2 runs of training data in Fig 5E ( t = -4 . 1 , p = 2e-4 , paired t-test ) , because in the full model the magnitudes of the posterior estimates of task-related activity patterns β ^ ( post ) are smaller with more training data , and this causes less mis-prediction for the test data . BRSA has a relatively rich model for the data: it attempts to model both the task-related signal and intrinsic fluctuation , and to capture voxel-specific SNR and noise properties . The covariance matrix U and the pseudo-SNR of each voxel serve as structured prior that make the estimation of β ^ more precise . In addition to allowing cross-validation , the more precise estimation of activity patterns and the richer model of data also enable decoding of signals related to task conditions from new data . Similarly to the procedure of calculating cross-validated log likelihood , but without pre-assuming a design matrix for the test data , we can calculate the posterior mean of X ^ test and X ^ 0test in the testing data . Fig 6A shows the decoded design matrix X ^ test for one task condition ( condition 6 in Figs 1B and 6B ) and one participant , using one run of training data with the second-highest SNR . Although our method decodes some spurious responses when there is no event of this task condition and that there is slight shift of baseline , overall the result captures most of the true responses in the design matrix . The average correlation between the decoded design matrix and the true design matrix is displayed in Fig 6B . High values on the diagonal elements indicate that overall , the decoder based on BRSA can recover the task-related signals well . These values are mostly beyond the range of the null distribution of correlations if task conditions were randomly shuffled ( Fig 6C . The 7 percentile of the distribution of the values in the diagonal elements of Fig 6B is beyond the 93 percentile of the distribution of the shuffled correlations ) . The structure of the off-diagonal elements appears highly similar to those of the correlation structure between corresponding columns in the original design matrix ( r = 0 . 82 , p<6e-30 ) and also similar to the true pattern similarity structure we simulated ( r = 0 . 30 , p<1e-3 ) . The resemblance to correlation in design matrix means that the signals corresponding to task conditions which often occur closely in time in training data are more likely to be confused when they are decoded from testing data . This is likely because the overlapping time courses between frequently co-occurring conditions make it difficult to distinguish which of two nearby events triggered the BOLD response during training , and reduced the accuracy of the posterior estimation of response patterns . We suspect that such confusion is not limited to decoding based on BRSA , but should be a general limitation of multi-variate pattern analysis of fMRI data: due to the slow smooth BOLD response , the more often the events of two task conditions occur closely in time in the training data , the more difficult it becomes for the classifier to discern their patterns . The resemblance to the true pattern similarity is not surprising since activity in conditions with similar neural patterns are expected to be more difficult to discern . The diagonal values were higher in conditions 1-8 , because these conditions occurred for more times in the task design . This is expected , since more measurements lead to more reliable estimation of activity patterns .
In this paper , we demonstrated that bias can arise in the result of representational similarity analysis , a popular method in many recent fMRI studies . By analytically deriving the source of the bias with simplifying assumptions , we showed that it is determined by both the timing structure of the experiment design and the correlation structure of the noise ( and task-unrelated neural activity ) in the data . Traditional RSA is based on point estimates of neural activation patterns which unavoidably include high amounts of noise . The task design and noise property induce covariance structure in the noise of the pattern estimates . This structure , in turn , biases the covariance structure of these point estimates , and a bias persists in the similarity matrix . Such bias is especially severe when the SNR is low and when the order of the task conditions is not fully counterbalanced . The bias demonstrated in this paper does not necessarily question the validity of all previous results generated by RSA . It does call for more caution when applying RSA to higher-level brain areas in which SNR of fMRI is typically low , and when the order of different task conditions cannot be fully counterbalanced . Counterbalancing a design could be done at different levels . The lowest level is randomizing the order of task conditions within a run . At higher levels , task sequences can also be randomized across runs and participants . If there are few measurements of each task condition within a run , randomizing trial order only within a run does not guarantee a perfectly counterbalanced design . The bias structure can then deviate from a diagonal matrix . If the same random sequence is used for all runs , or for all participants , small biases can persist across runs and participants and become a confound . Therefore , it is also important to use different task sequences across runs and participants when possible . Although such counterbalancing at different levels is desirable , it may not be achievable when studying high-level cognition , for instance in decision making tasks that involve learning or structured sequential decisions . These tasks often require ( or impose ) a specific relationship among conditions and events cannot be randomly shuffled . To reduce this bias , especially for cases when counterbalancing task conditions is not possible , we proposed a Bayesian framework that interprets the representational structure as reflecting the shared covariance structure of activity levels across voxels . Our BRSA method estimates this covariance structure directly from data , bypassing the structured noise in the point estimates of activity profiles , and explicitly modeling the spatial and temporal structure of the noise . This is different from many other methods that attempt to correct the bias after it has been introduced . Interestingly , Lindquist et al . recently pointed out that when applying a sequence of processing steps to fMRI data , a later step can reintroduce nuisance effects that earlier steps attempted to remove . They therefore suggest combining all the processing steps into one single filtering step [40] , which is similar in spirit to the method we propose here . Although the issues of performing multiple analyses steps sequentially arise in different context , these works together raise caution for developing new analyses on quantities estimated by existing processing pipelines . Furthermore , in the view of [40] , regression analysis is one but many possible projections of the original data to a subspace . Therefore , other preprocessing steps can also potentially alter the covariance structure of noise in the point estimates of patterns , and consequently alter the bias structure in standard RSA . In addition to inferring the representational similarity structure , our method also infers activation patterns ( as an alternative to the traditional GLM ) , SNR for different voxels , and even the “design matrix” for data recorded without knowledge of the underlying conditions . The inferred activation patterns are regularized not only by the SNR , but also by the learned similarity structure . The inference of an unknown “design matrix” allows one to uncover uninstructed task conditions ( e . g . , in free thought ) using the full Bayesian machinery and all available data . In a realistic simulation using real fMRI data as background noise , we showed that BRSA generally outperforms standard RSA and cross-run RSA , especially when the SNR is low and when the amount of data is limited , making our method a good candidate in scenarios of low SNR ( e . g . , higher order cortices ) and difficult-to-balance tasks ( e . g . , learning and sequential decision making ) . Because temporal and spatial correlations also exist in the noise of data from other neural recording modalities , the method can also be applied to other types of data when the bias in standard RSA is of concern . To detect overfitting to noise , the difference between the cross-validated score of the full model of BRSA and a null model can serve as the basis for model selection . We further extended the model to allow for estimating the shared representational structure across a group of participants . This shared structure can be very useful in model testing . As suggested by [41] , the correlation between similarity matrices estimated from individual participants’ data and the group average similarity matrix reflects the “noise ceiling”: the expected correlation one can achieve between the similarity matrix from an unknown “true” computational model and the similarity matrix from data , given the amount of noise in the data . The shared similarity matrix obtained through the joint fitting of GBRSA can be used in place of the group average similarity matrix for the purpose of calculating this noise ceiling . Prior proposals suggested the use of similarity between patterns estimated from separate scanning runs ( cross-run RSA ) [18 , 19 , 42] , in order to overcome “pattern drift” [18] , which can be seen as being caused by an interaction between the study design and auto-correlated fMRI noise . The inner product between noise pattern estimates from separate runs is indeed theoretically unbiased . Cross-run RSA was also proposed for assessing the extent to which brain patterns can discriminate between different conditions [43] . In our simulations , cross-run RSA generally performs better than within-run RSA , but worse than BRSA . However , after spatial whitening of the point estimates of activity patterns , cross-run RSA outperforms BRSA at high SNR , with the results of both methods very close to the true similarity structure in this case . On the other hand , at low SNR , BRSA performs better . The limitation of cross-run RSA is that even though the cross-run covariance matrix is unbiased , the magnitude of cross-run correlation is underestimated . Additionally , the high noise in the pattern estimates can also lead to results that are difficult to interpret , such as an anti-correlation between estimated patterns of the same condition from different runs . In general , it is not guaranteed that cross-run RSA will obtain a measure of distance that satisfies the triangle inequality . In contrast , although BRSA is not fully unbiased , it does guarantee that the estimated similarity matrix is positive semi-definite , and that an interpretable distance metric can be derived by subtracting the similarity matrix from 1 . In sum , it is difficult to predict whether BRSA or cross-run RSA with spatial whitening will be more suitable for any specific study and brain area of interest since their performance depends on SNR and both methods have limitations . Nonetheless , based on our results , both approaches should always be favored over traditional within-run RSA . While spatial whitening might be desirable for the purpose of cross-run RSA since it improves the performance , one should take caution when interpreting activity pattern after whitening . This is because spatial whitening remixes data across voxels and risks moving task-related signals from voxels responding to a task to non-responsive voxels . Instead of performing spatial whitening , BRSA estimates several time series X0 that best explain the correlation of noise between voxels , and marginalizes over their modulations of activity in each voxel . Therefore , BRSA can capture spatial noise correlation without the risk of misattributing signals across voxels . In our study , we did not directly compare BRSA to cross-validated Mahalanobis distance [19] because these two methods are fundamentally different measures: BRSA aims to estimate the correlation between patterns , which is close to the cosine angle between two patterns vectors [44–46]; in contrast , Mahalanobis distance aims to measure the distance between patterns . Nonetheless , given the theoretical soundness of the cross-validated Mahalanobis distance and the possibility of statistically testing the distances against 0 , it could also be a good alternative to BRSA when there are multiple runs in a study . Our BRSA method is closely related to pattern component modeling ( PCM ) [10 , 47] . A major difference is that PCM models the point estimates β ^ after a GLM analysis while BRSA models fMRI data Y directly . The original PCM [10] in fact considered the contribution of noise in pattern estimates to the similarity matrix , but assumed that the noise in β ^ is i . i . d across task conditions . This means that the bias in the covariance matrix was assumed to be restricted to a diagonal matrix . We showed here that when the order of task conditions cannot be fully counterbalanced , such as in the example in Fig 1 , this assumption is violated and the bias cannot be accounted for by methods such as PCM . Another difference is that BRSA explicitly models spatial noise correlation , which can improve the results ( see Figure 3 of S1 Material ) . If the covariance structure of the noise Σϵ were known , the diagonal component of the noise covariance structure assumed in PCM [10] could be replaced by the bias term ( XT X ) −1 XTΣϵ X ( XT X ) −1 to adapt PCM to estimate the covariance structure U ^ that best explains β ^ [47] , ignoring spatial noise correlation . However , as shown in Fig 1G , different voxels can have a wide range of different autocorrelation coefficients , therefore assuming a single Σϵ for all voxels may be over-simplifying . PCM also assumes all voxels within one ROI have equal SNR . However , typically only a small fraction of voxels exhibit high SNR [29] . Therefore , it is useful to model the noise property and SNR of each voxel individually . Relatedly , Alink et al . proposed to model the bias structure in the similarity matrix as a polynomial function of the temporal distance between each pair of conditions and regress out such structure [18] . This regression approach is plausible when each task condition contains a single event . If one were to accept the simplifying assumptions above , a more principled choice for regressing out the bias structure would use the analytic form of the bias structure we derived above as a regressor . BRSA comes with the ability to select between a full model and null model based on cross-validated log likelihood , and can be applied to fMRI decoding . PCM can evaluate the likelihood of a few fixed candidate representational structures given by different computational models . It can also estimate the additive contributions of several candidate pattern covariance structures to the observed covariance structure . These options are not yet available in the current implementation of BRSA . Combining the strengths of PCM and BRSA is an interesting direction for future research . Another future direction is to cross-validate full and null models on the data of left-out subjects , instead of cross-validating within subjects on a left-out run of data . Many aspects of flexibility may be incorporated to BRSA . For example , the success of the analysis hinges on the assumption that the hemodynamic response function ( HRF ) used in the design matrix correctly reflects the true hemodynamics in the region of interest , but the HRF in fact varies across people and across brain regions [48 , 49] . Jointly fitting the shape of the HRF and the representational structure using BRSA may thus improve the estimation . In addition , it is possible that even if the landscape of activity patterns for a task condition stays the same , the global amplitude of the response pattern may vary across trials due to repetition suppression [50–52] and attention [53 , 54] . Allowing global amplitude modulation of patterns associated with a task condition to vary across trials might capture such variability and increase the power of the method . Finally , our simulations revealed that BRSA is not entirely unbiased , that is , the results cannot be improved indefinitely by adding more subjects . This bias is not a consequence of the number of components estimated by the algorithm we chose [35] ( see Part 6 of S1 Material ) , and further investigation is needed to understand the source of this remaining bias . The residual bias occurs when SNR is very low and may be due to overfitting of the model to noise . Fortunately , the cross-validation procedure we provided helps to detect overfitting when the SNR is too low . When this happens , it is advisable to focus on taking measures to improve the design of study . Ultimately , task designs that are not fully counterbalanced and low SNR in fMRI data are two critical factors that cause bias in traditional RSA and impact the power of detecting similarity structure in neural representations . Carefully designing tasks that balance the task conditions as much as possible , using different randomized task sequences across runs and across participants , and increasing the number of measurements , are our first line of recommendations . In the analysis phase of the project , one can then use BRSA .
The fMRI data used in the current manuscript came from two sources . Part of them came from a previously published study , which was approved by the Princeton Institutional Review Board and all subjects gave informed written consent prior to participation . The other part came from open access data of the Human Connectome Project ( https://www . humanconnectome . org/ ) . The detailed derivation of the generative model , the model fitting procedure , model selection based on cross-validation and decoding task-related signals are in the Parts 1 , 2 and 3 of S1 Material . Here we provide the major assumptions and the formula of the likelihood of fMRI data in our model . Our generative model of fMRI data follows the general assumption of GLM . In addition , we model spatial noise correlation by a few time series X0 shared across all voxels . The contribution of X0 to the k-th voxel is β0⋅k . Thus , for voxel k , we assume that Y k = X β · k + X 0 β 0 · k + ϵ k ( 9 ) Yk is the time series of voxel k . X is the design matrix shared by all voxels . β⋅k is the response amplitudes of the voxel k to all the task conditions . ϵk is the residual noise in voxel k which cannot be explained by either X or X0 . We assume that ϵ is spatially independent across voxels , and all the correlation in noise between voxels are captured by the shared intrinsic fluctuation X0 . We use an AR ( 1 ) process to model ϵk: for the k-th voxel , we denote the noise at time t > 0 as ϵt , k , and assume ϵ t , k = ρ k ϵ t - 1 , k + η t , k , η t , k ∼ N ( 0 , σ k 2 ) ( 10 ) where σ k 2 is the variance of the “shock” ( innovation ) , the component at each time point t that is independent from ϵt−1 , k , and ρk is the autoregressive coefficient for the k-th voxel . We assume that the covariance of the multivariate Gaussian distribution from which the activity amplitudes βk are generated has a scaling factor that depends on its pseudo-SNR sk: β · k ∼ N ( 0 , ( s k σ k ) 2 U ) . ( 11 ) This is to reflect the fact that not all voxels in an ROI have equal SNR . We further use Cholesky decomposition to parametrize the covariance structure U: U = LLT , where L is a lower triangular matrix . With the above assumption of the distributions of activity profiles and noise in each voxel , after marginalizing β0⋅k and β⋅k , we have p ( Y k |X , X 0 , L , σ k , ρ k , s k ) =∫ ∫ p ( Y k | β · k , β 0 · k , X , X 0 , σ k , ρ k ) p ( β 0 · k ) p ( β · k | s k , σ k , L ) d β 0 · k d β · k ∝ ( 2 π ) - n T - n 0 2 | Σ ϵ k - 1 | 1 2 | X 0 T Σ ϵ k - 1 X 0 | - 1 2 | Λ k * | 1 2 · exp [ - 1 2 ( 1 σ k 2 Y k T A k * Y k - μ k * T Λ k * - 1 μ k * ) ] ( 12 ) In the equation above , Σ ϵ k - 1 is the temporal autocorrelation matrix of the AR ( 1 ) noise in voxel k . A k * = σ k 2 ( Σ ϵ k - 1 - Σ ϵ k - 1 X 0 ( X 0 T Σ ϵ k - 1 X 0 ) - 1 X 0 T Σ ϵ k - 1 ) = A k - A k X 0 ( X 0 T A k X 0 ) - 1 X 0 T A k , where A k = A ( ρ k ) = σ k 2 Σ ϵ k - 1 . Λ k * = ( i + s k 2 L T X T A k * X L ) - 1 and μ k = s k σ k Λ k * L T X T A k * Y k . σk can be further analytically marginalized . sk and ρk cannot be analytically marginalized . But we can numerically marginalize them by weighted sum of the likelihood at nl × nm discrete grids {ρkl , skm} ( 0 < l < nl , 0 < m < nm ) with each grid representing one area of the parameter space of ( ρ , s ) . The weights w ( ρkl , skm ) reflect the prior probabilities of the two parameters in the area represented by {ρkl , skm} . We assume uniform prior of ρ in ( -1 , 1 ) . All the simulations in this paper used an exponential distribution as prior for s . Alternative forms of priors such as uniform in ( 0 , 1 ) , log normal distribution , and Delta distribution of fixed pseudo-SNR are also implemented in the tool . A comparison of the impact of different forms of prior assumption on BRSA’s performance is provided in Part 5 Comparison of BRSA with different assumptions of the prior of pseudo-SNR of S1 Material . This procedure yields the marginalized log likelihood for each voxel: p ( Y k |X , X 0 , L ) ≈∑ l = 1 n l ∑ m = 1 n m p ( Y k | X , X 0 , L , ρ k l , s k m ) w ( ρ k l , s k m ) ∝∑ l = 1 n l ∑ m = 1 n m ( 2 π ) - n T - n 0 2 ( 1 - ρ k l 2 ) n r 2 | X 0 T A k l X 0 | - 1 2 | Λ k l m * | 1 2 Γ ( n T - n 0 2 - 1 ) · [ Y k T A k l * Y k - s k m 2 Y k T A k l * X L Λ k l m * L T X T A k l * Y k 2 ] 1 - n T - n 0 2 w ( ρ k l , s k m ) ( 13 ) where A k l * is A k * assessed at parameter grid ρkl , Λ k l m * is Λ k * assessed at grid {ρkl , skm} . Because we made the assumption that ϵk is independent across voxels , the log likelihood for all data is the sum of the log likelihood for each voxel . logp ( Y | X , X 0 , L ) = ∑ k = 1 n V logp ( Y k | X , X 0 , L ) . ( 14 ) For the null model , the likelihood for each voxel after marginalizing β0 · k and σ k 2 can be similarly derived , p ( Y k| X 0 , ρ k ) ∝ ( 2 π ) - n T - n 0 2 ( 1 - ρ k 2 ) n r 2 | X 0 T A k X 0 | - 1 2 · Γ ( n T - n 0 2 - 1 ) [ Y k T A k * Y k 2 ] 1 - n T - n 0 2 ( 15 ) and the total log likelihood can be calculated similarly by numerically marginalizing ρk and summing the log likelihood for all voxels . Data used in Fig 1B are from the experiments of Schuck et al . [24] , following the same preprocessing procedure as the original study . The fMRI data were acquired at TR = 2 . 4s . Data of 24 participants were used . Their design matrices were used for all the following analyses and simulations . Data in Figs 1E , 1G and 3 were preprocessed data obtained from Human Connectome Project ( HCP ) [37] . The first 24 participants who have completed all 3T protocols and whose data were acquired in quarter 8 of the HCP acquiring period without image quality issues were selected for analysis in Fig 3 . Data from 864 participants without image quality issues in HCP were used in the analysis in Fig 5 . Each participants in the HCP data have 2 runs of resting state data with left-right phase encoding direction and 2 runs with right-left phase encoding direction . Denoising pipeline ICA-AROMA [55] with default parameters was performed on each run of data . The time series of the aggressively denoised data were resampled at the same TR as the design matrix and truncated to the length of each run in that design matrix before further analysis ( all results were similar without ICA-AROMA denoising ) . Each run contained 182 ± 12 time points . β^ point estimates in Fig 1 were obtained with AFNI’s 3ddeconvolve [56] . The design matrices were set up by convolving the stereotypical double-Gamma HRF in SPM [57] with event time courses composed with impulses lasting for the duration of the participants’ reaction time . AR ( 1 ) coefficients in Fig 1G were estimated after upsampling the fMRI time series in the HCP data to the TR in Schuck et al . [24] and linear detrending . Upsampling is to reflect the lower temporal resolution more typically employed in task-related fMRI studies . In the experiments of Fig 3 , lateral occipital cortex was chosen as the ROI , which included 4802 ± 31 ( mean ± standard deviation ) voxels . Task related signals were only added to voxels within a bounding box ( Fig 3C ) of which the coordinates satisfy 25 < x < 35 , -95< y < −5 and -15< z <5 . 181 . 9±0 . 3 voxels fell within this bounding box . To generate activity patterns at different SNR , β were sampled independently from a multivarite Gaussian distribution with the covariance matrix in Fig 3A for each voxel within the bounding box mentioned above ( resting state data from HCP ) , and were scaled by one values in 1 , 2 , 4 , 8 times the standard deviation of the detrended noise in that voxel . The design matrix X mentioned above were then multiplied with β for each voxel and added to the noise . Different random seeds were used in the simulation for different subjects , different SNR levels and different amounts of simulated data . But the same simulated data were used across different RSA analysis methods . Due to the small magnitude of the design matrix , the resulting ratios of the standard deviation of the simulated signals Xβ⋅k to that of the detrended noise were on average 0 . 14 , 0 . 27 , 0 . 54 and 1 . 08 within the bounding box ( Fig 3C ) . To estimate the similarity matrix of this task involving 16 conditions , the fitting time on an Intel Xeon processor employing 12 CPUs is 751±105 s , 1274±414 s and 2206±564 s , for data of 1 run , 2 runs and 4 runs , respectively . This suggests that our method is practically feasible even for relatively large ROIs . To evaluate the performance of the recovered correlation structure by different methods , the off-diagonal elements of the similarity matrix recovered from data of each simulated participant was correlated with those elements of the ideal similarity matrix to yield the top panel of Fig 3I . The top panel reflects the correlation of individual results . The bottom panel reflects the correlation of average results over simulated participants . In order to make fair comparison with BRSA which considers temporal auto-correlation in noise , all the point estimates of β ^ by other methods in Fig 3 were performed with restricted maximum likelihood estimation , which model the auto-correlation in noise . AR ( 1 ) parameters of each voxel were estimated after initial regular regression . The AR ( 1 ) parameters were used to re-compute the temporal noise covariance matrices for each voxel and β ^ were calculated again assuming these noise covariance matrices . To account for task-irrelevant time courses in the data , extra nuisance regressors were included in all methods based on point estimates of activity patterns . These included Legendre polynomial functions of volume index , up to the fourth order , to model slow drift of fMRI signals , and the first three principal components of the fMRI data in each of white matter and ventricles , to capture intrinsic fluctuations . The principal components were also included as nuisance regressors for BRSA . When spatial whitening of β ^ was performed , residuals of fitting from all runs of data from the same simulated subject were concatenated in time . One covariance matrix Σspatial across voxels was then estimated from these residuals using the optimal shrinkage method [33] implemented in scikit-learn [58] . This covariance matrix was then used to whiten the estimated pattern of all the runs . That is , β ^ Σ s p a t i a l - 1 2 of each run were treated as the spatially whitened pattern estimates . When performing within-run RSA , the estimated patterns of each run ( being spatially whitened or not ) were averaged over the simulated runs before subjecting to a Pearson correlation . When performing cross-run RSA , the correlations were calculated between the estimated patterns corresponding to any two conditions from two different runs . This calculation was repeated for each combination of two runs in the data . And finally all the correlation coefficients between the two conditions calculated over all pairs of runs were averaged ( e . g . , 12 cross-run similarity matrix would be calculated from 4 runs of data and averaged to generate one matrix ) . To simulate the fMRI noise in Fig 4 , we first estimated the number of principal components to describe the spatial noise correlation in the 24 resting state fMRI data from HCP database using the algoritm of Gavish and Donoho [35] . The spatial patterns of these principal components were kept fixed as the modulation magnitude β0 by the intrinsic fluctuation . AR ( 1 ) parameters for each voxel’s spatially indepndent noise were estimated from the residuals after subtrating these principal components . For each simulated subject , time courses of intrinsic flucutations were newly simulated by scrambling the phase of the Fourier transformation of the X0 estimated from the real data , thus preserving the amplitudes of their frequency spectrum . AR ( 1 ) noise were then added to each voxel with the same parameters as estimated from the real data . To speed up the simulation , only 200 random voxels from the ROI in Fig 3B were kept for each participant in these simulations . Among them , 100 random voxels were added with simulated task-related signals . Thus , each simulated participant has different spatial patterns of β0 due to the random selection of voxels . 500 simulated datasets were generated based on the real data of each participant , for each of the three SNR levels . In total 36000 subjects were simulated . The simulated pool of subjects were sub-divided into bins with a fixed number of simulated subjects ranging from 24 to 1200 . The mean and standard deviation of the correlation between the true similarity matrix and the average similarity matrix based on the subjects in each bin were calculated , and plotted in Fig 4A . All SNRs in Figs 3 and 4 were calculated post hoc , using the standard deviation of the added signals in the bounding box region devided by the standard deviation of the noise in each voxel , and averaged across voxels and simulated subjects for each level of simulation . | We show the severity of the bias introduced when performing representational similarity analysis ( RSA ) based on neural activity pattern estimated within imaging runs . Our Bayesian RSA method significantly reduces the bias and can learn a shared representational structure across multiple participants . We also demonstrate its extension as a new multi-class decoding tool . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"learning",
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"radiology",
"functional",
"magnetic",
"resonance",
"imaging",
"engineering",
"and",
"technology",
"signal",
"processing",
"social",
"sciences",
"random",
"variables",
"mathematical",
"models",
"neuroscience",
"covariance",
"learning",
"and",
"memory",
"magnetic",
"resonance",
"imaging",
"simulation",
"and",
"modeling",
"cognitive",
"psychology",
"mathematics",
"brain",
"mapping",
"white",
"noise",
"neuroimaging",
"research",
"and",
"analysis",
"methods",
"imaging",
"techniques",
"resting",
"state",
"functional",
"magnetic",
"resonance",
"imaging",
"mathematical",
"and",
"statistical",
"techniques",
"probability",
"theory",
"psychology",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"signal",
"to",
"noise",
"ratio",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"cognitive",
"science"
] | 2019 | Representational structure or task structure? Bias in neural representational similarity analysis and a Bayesian method for reducing bias |
Histone deacetylase Rpd3 is part of two distinct complexes: the large ( Rpd3L ) and small ( Rpd3S ) complexes . While Rpd3L targets specific promoters for gene repression , Rpd3S is recruited to ORFs to deacetylate histones in the wake of RNA polymerase II , to prevent cryptic initiation within genes . Methylation of histone H3 at lysine 36 by the Set2 methyltransferase is thought to mediate the recruitment of Rpd3S . Here , we confirm by ChIP–Chip that Rpd3S binds active ORFs . Surprisingly , however , Rpd3S is not recruited to all active genes , and its recruitment is Set2-independent . However , Rpd3S complexes recruited in the absence of H3K36 methylation appear to be inactive . Finally , we present evidence implicating the yeast DSIF complex ( Spt4/5 ) and RNA polymerase II phosphorylation by Kin28 and Ctk1 in the recruitment of Rpd3S to active genes . Taken together , our data support a model where Set2-dependent histone H3 methylation is required for the activation of Rpd3S following its recruitment to the RNA polymerase II C-terminal domain .
Histone acetylation was the first covalent histone modification shown to be involved in transcription regulation . Indeed , histones at the promoter of active genes tend to be hyper-acetylated while repressed genes have promoters with hypo-acetylated nucleosomes . It is now well established that this is due to the recruitment of histone acetyltransferases ( HATs ) and histone deacetylases ( HDACs ) by transcriptional activators and repressors , respectively [1] . The first described and best characterized HDAC is Rpd3 . The repressive effect of yeast Rpd3 on transcription has been well studied over the last 15 years , paving the way for the characterization of its mammalian orthologs [2] . Yeast Rpd3 is recruited to the promoters of specific genes by DNA-binding repressors , leading to the repression of many important pathways such as stress response , meiosis , the cell cycle and others [3]–[19] . Moreover , Rpd3 also plays roles in silencing [20]–[27] , DNA replication [28]–[31] and recombination [32] , [33] . Recent proteomic studies have determined that Rpd3 is found in two distinct complexes: the large ( Rpd3L ) and the small ( Rpd3S ) complex [34] , [35] . Both complexes share a core composed of Rpd3 , Sin3 and Ume1 . The large complex is composed of 11 additional proteins whereas the small complex contains only two additional subunits , namely Rco1 and Eaf3 . While Rpd3L is likely responsible for the repressive function of Rpd3 , the function of Rpd3S remains far less understood . Recent work by several groups has shown that Rpd3S is involved in the suppression of cryptic transcription [34] , [36] , [37] and that its activity is linked to the Set2 histone methyltransferase ( HMT ) . Furthermore , in vitro studies have shown that Rpd3S is recruited to H3K36-methylated nucleosomes and that its Rco1 and Eaf3 subunits are essential for this recruitment [34] , [35] , [37] , [38] . Rco1 mediates interactions with histones in a modification-independent manner through a PHD zinc finger domain , while Eaf3 contains a methyl-lysine binding chromodomain ( CHD ) that is essential for recognition of H3K36-trimethylated ( H3K36me3 ) nucleosomes in vitro . These studies also indicate that genome-wide histone acetylation levels on promoters and coding regions are altered when either H3K36 methylation or Rpd3S is disrupted . Taken together , these data lead to a model where Rpd3S is recruited to coding regions through the interaction of Eaf3 with H3K36me3 in order to deacetylate nucleosomes after the passage of the transcriptional machinery [39] . Deacetylation would allow chromatin disrupted by elongating RNA polymerase II ( RNAPII ) to return to a more ordered and compact structure , thereby restoring an environment hostile to cryptic transcription initiation within the coding region . However , this model relies heavily on the in vitro observation that the Eaf3 CHD binds preferentially to H3K36me3 peptides or nucleosomes [34] , [35] , [37] , [38] . It was never formally demonstrated that this interaction is required for the targeting of Rpd3S to coding regions in vivo . In addition , whether Rpd3S is recruited to all transcribed genes or to only a subset of them has never been assessed . In order to address these questions , we performed genome-wide ChIP-chip experiments looking at both Rpd3S- and Rpd3L-specific subunits in wild type cells and in various mutants , including set2Δ and H3K36A ( Table S1 and S2 ) . Quite interestingly , our data show that Rpd3S specifically binds to the coding region of actively transcribed genes whose promoters are also bound by Rpd3L . Surprisingly , the binding of Rpd3S to active genes is not dependant on Set2-mediated H3K36 methylation . However , methylation by Set2 is required for the activity of Rpd3S , as assayed by histone acetylation and RNAPII levels . We also provide in vivo evidence that Rpd3S is recruited to active genes via the phosphorylation of the RNAPII C-terminal domain ( CTD ) . Finally , we show that the yeast DSIF transcription elongation factor negatively regulates Rpd3S recruitment . Based on these results , we propose that the recruitment and activity of Rpd3S on ORFs depend on a two steps mechanism: an initial recruitment to the elongation complex -coordinated by DSIF and RNAPII phosphorylation- , followed by the H3K36me3-dependant modulation of Rpd3S activity through the Eaf3 chromodomain .
While it is generally accepted that Rpd3L negatively regulates specific sets of genes , the specific function of Rpd3S remains largely unaddressed . However , it is expected to be ubiquitously recruited to active genes since it interacts with methylated H3K36 . In order to address the specificity of Rpd3S and Rpd3L in a systematic manner , we performed ChIP-chip experiments on myc-tagged Rpd3 , Rco1 and Sds3 , the last two being specific subunits of Rpd3S and Rpd3L , respectively . As shown in Figure 1A , Rpd3 binds to a large subset of promoters ( Figure 1A , clusters 1 and 2 ) . In addition , a subset of these genes also exhibit Rpd3 binding on their coding regions ( Figure 1A , cluster 2 ) . Figure 1B and 1C show the average signal of Rpd3 , Rco1 and Sds3 over the genes from cluster 1 and cluster 2 , respectively . The data for cluster 3 , representing the genes not bound by Rpd3 , are also shown . Cluster 1 , which is enriched for genes previously demonstrated to be repressed by Rpd3 ( genes involved in M phase ( p-value 10−8 ) , cell cycle ( p-value 10−9 ) , etc . ) , shows binding of both Rpd3 and Sds3 to the promoter . The presence of Rco1 is not detectable on these genes . This cluster therefore represents genes repressed by Rpd3L , which is consistent with the fact that these genes have low level of RNAPII ( Figure S1A ) . Cluster 2 , however , shows evidence for the presence of both Rpd3L and Rpd3S since all three subunits tested for are detected . Rco1 is restricted to the coding region of these genes , consistent with the fact that they are actively transcribed ( Figure S1A ) . Sds3 is present at the promoter , which is expected since Rpd3 binds to these promoters . More surprisingly , however , some level of Sds3 is also detected on the coding region of these genes . These data - also observed with another subunit of Rpd3L ( Rxt2; Figure S1B ) - suggest that the large complex may play some role during transcriptional elongation , perhaps in conjunction with the small complex ( see Discussion ) . Nevertheless , the data presented here clearly show that Rpd3S binds to the coding region of active genes . Strikingly , these experiments also show that Rpd3S preferentially associates with genes that are also bound by Rpd3L . In fact , we found no clusters of genes where Rpd3 binds in the coding region but not in the promoter . This suggests that Rpd3S , contrary to what has been expected , does not ubiquitously bind to active genes but rather targets some of them , namely a subset of those that are bound by Rpd3L at their promoter . In order to test if Rpd3S ubiquitously binds active genes , we performed ChIP-chip experiments of RNAPII and H3K36me3 , two proxies for active gene expression . Figure 1D ( cluster 4 ) clearly shows that many genes , despite having strong enrichment for RNAPII and H3K36me3 , show no evidence of Rpd3S binding . Three conclusions can be drawn from these results; firstly , they confirm that the recruitment of Rpd3S is not a general phenomenon occurring on all transcribed genes; secondly , they suggest that the methylation of H3K36 by Set2 is not providing the specificity for the recruitment of Rpd3S; and finally they suggest that the large complex may play a role in the recruitment of the small complex as Rpd3S appears to target primarily genes also bound by Rpd3L . The idea that H3K36me3 recruits Rpd3S through the Eaf3 subunit is well established in the literature [39] . Work done in vitro by several groups has clearly shown this using peptides and nucleosomal substrates [34] , [35] , [37] , [38] . Our ChIP-chip experiments , however , clearly demonstrate that many genes harboring high levels of H3K36me3 are free of Rpd3S . We have therefore endeavored to examine whether the well characterized interaction between the Eaf3 CHD and H3K36me3 is responsible for the targeting of Rpd3S in vivo . First , we looked at Rco1 binding in a set2Δ mutant . Since this mutant cannot methylate H3K36 , the current model predicts that Rpd3S should not bind to ORFs under these conditions . To our considerable surprise , Rco1 enrichment on ORFs in this mutant is not significantly altered for about two-thirds of Rpd3S target genes ( Figure 2A , cluster 5 ) . For other genes , Rpd3S occupancy is decreased significantly , although not completely abolished ( Figure 2A , cluster 6 ) . We next repeated these experiments in a H3K36A mutant ( where lysine 36 is mutated into an alanine ) with similar results ( Figure 2A ) . The deletion of the Rco1 PHD domain had no effect on Rpd3S occupancy ( despite the fact that it destabilizes the Rco1 protein; Figure 2E ) , while deletion of the Eaf3 CHD domain phenocopied set2Δ and H3K36A . Interestingly , a set1Δ/set2Δ/dot1Δ triple mutant ( abolishing all histone methylation activity in yeast ) was similar to wild type , suggesting that deletion of SET1 and/or DOT1 can partially suppress the set2Δ phenotype . Taken together , these data demonstrate that H3K36 methylation is not required for the recruitment of Rpd3S to most genes . Since Rpd3S occupancy seems to be more dependent on H3 methylation at some genes than others , we looked more closely at clusters 5 and 6 . As shown in Figure 2B and 2C , clusters 5 and 6 are markedly different with regards to transcription levels . While cluster 5 is highly transcribed ( as shown by the presence of high levels of both RNAPII and H3K36me3 ) , cluster 6 is less so . Next we looked at the distribution of Rco1 on genes contained within these clusters in all strains shown in Figure 2A . As expected from data shown in Figure 2A , Rco1 occupancy on ORFs is not ( or only slightly ) affected in these mutants for the cluster 5 genes , but it is reduced for the genes from cluster 6 ( Figure 2D and Figure S2 ) . In addition , for all Rpd3S-bound genes , a redistribution of Rco1 to the promoter region was observed in all mutants tested . This redistribution towards the promoter remains unexplained but correlates with our observation that histone acetylation is decreased at promoters in these same mutants ( Figure S3 ) . Collectively , these data clearly demonstrate that H3K36 methylation has no impact on Rpd3S occupancy at highly transcribed genes ( genes from cluster 5 ) . However , the methyl mark , or the ability to recognize it through the Eaf3 chromodomain , is important for optimal Rpd3S association to ORFs with lower levels of RNAPII ( genes from cluster 6 ) . It is known that set2Δ mutants , along with null mutants of Rpd3S subunits Eaf3 or Rco1 , exhibit a cryptic transcription phenotype [34] , [36] , [37] . This phenotype is thought to be due to the improper deacetylation of transcribed ORFs by Rpd3S after each round of transcription [39] because of a lack of Rpd3S recruitment . Other groups that have characterized acetylation levels on coding regions in Set2 and Rpd3S mutants either looked at bulk chromatin by western blotting [37] , or at specific genes by ChIP [34] , [35] , [38] , [40] and have come to the conclusion that acetylation levels increase on ORFs when Set2 or Rpd3S is disrupted . Since we—quite surprisingly—observed , however , that Rpd3S binding to genes is mostly independent of histone methylation by Set2 , we decided to test whether the activity of Rpd3S requires methylation of H3K36 by Set2 . To do so , we looked at H4K5 acetylation ( H4K5ac ) by ChIP-chip . We used H4K5 acetylation to score for Rpd3S activity because it was shown previously to be a robust read out for Rpd3 activity in ChIP-chip assays [41] . Similar to other groups [34] , [35] , [38] , [40] , we observed decreased acetylation on promoters in set2Δ , H3K36A or Rpd3S mutants ( Figure S3 ) . Histone acetylation is also dramatically affected across ORFs in these mutants . As shown in Figure 3A , we observed a loss of acetylation for normally highly acetylated ORFs , and a gain in acetylation for ORFs that exhibit low levels in the wild type . Because Set2 and Rpd3S are both known to prevent cryptic initiation within ORFs , we repeated the same analyses on RNAPII ChIP-chip results , and found a similar pattern to that observed for H4K5ac , namely that ORFs with high RNAPII enrichment show decreased RNAPII levels in the absence of Set2 or Rpd3S , whereas ORFs with low RNAPII tend to display higher levels of polymerase ( Figure 3B ) . These results clearly show that the activity of the Rpd3S complex requires methylation of H3K36 by Set2 . The effect of the loss of Rpd3S activity on histone acetylation and RNAPII distribution on ORFs is more complex than previously described . Our data indeed suggest that both histone acetylation and RNAPII occupancy are redistributed in a more even manner across the genome than expected . A plausible explanation of the genome-wide averaging of RNAPII and histone acetylation levels in Set2 and Rpd3S mutants would entail aberrant recruitment of the transcriptional apparatus to low-expression genes whose coding regions were not reset properly by a , now inactive , Rpd3S complex . Assuming a limited pool of transcription machinery in a cell , this would result in a lower abundance of RNAPII at the more active ORFs , and aberrant genomic acetylation levels . These results , combined with the Rpd3S occupancy profiles in rco1Δ , rco1-PHDΔ and eaf3-CHDΔ mutants , suggest that the loss of histone H3 methylation at lysine 36 affects ORF identity through a modulation of Rpd3S deacetylase activity rather than through altered recruitment of the Rpd3S complex on coding regions as previously thought . Since the interaction between the Eaf3 CHD and H3K36me3 does not account for the initial recruitment of Rpd3S to active genes , we decided to look for a factor that would fulfill that role . Quan and Hartzog have shown genetic interactions between H3K36 methylation and Rpd3S with Spt5 [42] . Their data suggest that Rpd3S opposes the function of the elongation factor Spt4/5 , which is the yeast ortholog of the human elongation factor DSIF [43] . DSIF negatively regulates elongation in its non-phosphorylated form , but is turned into a positive elongation factor upon phosphorylation by P-TEFb ( Bur1 in yeast ) [44]–[46] . The genetic interaction between Rpd3S and Spt5 led us to test whether Spt4/Spt5 is involved in the recruitment of Rpd3S . We therefore performed ChIP-chip experiments of Rco1 in spt4Δ cells ( the deletion of SPT5 is lethal ) . As shown in Figure 4A , deletion of SPT4 leads to massive changes in Rco1 binding across the genome . Notably , the effect is far more dramatic compared to the deletion of SET2 ( Figure 4A ) . Importantly , the level of RNAPII observed on these genes is not significantly affected in the mutant , ruling out the possibility that the effect is solely due to a reshuffling of the transcriptome ( Pearson correlation = 0 . 94 , Figure S4A ) . The deletion of SPT4 causes a decrease of Rco1 binding at some transcribed genes normally strongly associated with Rco1 ( Figure 4A , cluster 7 ) , as well as an increase at others where Rco1 is otherwise only found at low levels ( cluster 8 ) . Deletion of SPT4 even causes a slight increase of Rco1 occupancy at genes where it is normally undetectable ( cluster 9 ) . Overall , these effects lead to a distribution of Rpd3S that correlates better with RNAPII occupancy than in wild type cells ( Figure 4B ) . To test the possibility that , in the absence of Spt4 , Rpd3S is recruited via H3K36 methylation , we profiled Rco1 binding in a spt4Δ/set2Δ double mutant . As shown in Figure 4A , deleting both SPT4 and SET2 leads to a similar Rpd3S localization phenotype compared to the single spt4Δ mutant , giving further evidence that Set2 does not play a large role in Rpd3S recruitment , even in the absence of Spt4 . To distinguish the direct effect of the loss of SPT4 from eventual indirect effects of the mutation , we localized Spt4 in wild type cells by ChIP-chip using a strain carrying a myc-tagged SPT4 gene . Spt4 associates with genes in a manner that correlates with levels of RNAPII ( Pearson correlation = 0 . 84 ) suggesting that DSIF acts as a general elongation factor . Moreover , it is present across the whole ORF , indicating that it travels with RNAPII , but is further enriched in the 3′ end of genes ( Figure S4B ) , suggesting that it may also regulate the elongation-termination transition , as shown by others [47]–[49] . This is also consistent with the fact that Spt5 interacts with components of the capping and termination machineries [50] . Even more interestingly , the more a gene is occupied by Spt4 in wild type cells , the more Rco1 we detect in the spt4Δ mutant ( Figure 4C ) , suggesting that the direct effect of the loss of Spt4 is an increase in Rpd3S binding ( as observed for cluster 8 ) . Consequently , the decrease in occupancy observed in cluster 7 is most likely indirect since Spt4 is barely detectable at these genes ( Figure 4A ) . Similarly , the level of Rco1 increases dramatically in the spt4Δ mutant on cluster 4 genes ( from Figure 1D ) , representing transcribed genes highly occupied by Spt4 where Rco1 is absent in wild type cells ( Figure S4C ) . In general , genes with a higher Spt4/RNAPII ratio tend to have less Rco1 than genes with lower Spt4/RNAPII ratios ( Pearson correlation = −0 . 42 , Figure S4D ) . Taken together , these data suggest that Spt4 acts as a negative regulator of Rpd3S binding and that its presence prevents the HDAC from freely associating with transcribed genes . This model is in agreement with genetic data showing that Rco1 opposes the function of Spt4/5 [42] . We then looked at the distribution of Rpd3S on transcribed genes where Spt4 is also bound ( cluster 8 ) and found strong differences between the wild type and spt4Δ mutants . While Rco1 occupies the whole ORF at a constant level in wild type cells , it accumulates to abnormally high levels towards the 3′ end of the gene in spt4Δ cells ( Figure 4D , dashed line ) . This binding pattern is also found in the spt4Δ/set2Δ double mutant ( compare dashed and dotted lines in Figure 4D ) . The occupancy profile of Rco1 in a spt4Δ mutant shows a strong similarity to the occupancy profile of RNAPII with a CTD phosphorylated at Ser2 ( Figure 4D , red solid line ) . This led us to hypothesize that CTD phosphorylation by Ctk1 ( the major serine 2 kinase ) might be implicated in Rpd3S recruitment in the absence of Spt4/5 . To test this hypothesis , we profiled Rco1 occupancy in a spt4Δ/ctk1Δ background . Surprisingly , ctk1Δ partially suppressed the spt4Δ Rpd3S binding pattern phenotype ( Figure 4A ) leading to an intermediate binding profile between wild type and spt4Δ . Looking at it more closely , we observed that short genes fail to accumulate Rco1 in spt4Δ/ctk1Δ cells ( Figure 4E ) , whereas Rco1 is observed at genes irrespective of their length in wild type ( Figure S5 ) or spt4Δ cells ( Figure 4F ) , suggesting that Rpd3S accumulates more slowly in spt4Δ/ctk1Δ cells . We therefore conclude that in the absence of Spt4 , Rpd3S binds to the RNAPII CTD and that the phosphorylation of the CTD at serine 2 contributes to that phenomenon . The data presented above demonstrate that Spt4 negatively regulates the association of Rpd3S to highly transcribed genes . The data also suggest that phosphorylation of the RNAPII CTD , notably at serine 2 , plays some role in the recruitment of Rpd3S to active genes in the absence of Spt4 . We next tested whether the phosphorylation of the CTD is implicated in the association of Rpd3S with transcribed genes in wild type cells and performed ChIP-chip experiments of Rco1 in mutants for the known CTD kinases . As shown in Figure 5A , deletion of CTK1 , the major serine 2 kinase , has a clear effect on Rco1 occupancy ( compare solid line with dashed line ) . To test the effect of serine 5 and 7 phosphorylation , we used a strain carrying ATP-analog-sensitive alleles of KIN28 since the deletion of the KIN28 gene is lethal . As shown in Figure 5A ( dotted line ) , inhibition of Kin28 has a dramatic effect on Rco1 occupancy . Indeed , no Rco1 can be detected on ORFs in that mutant . This clearly demonstrates that phosphorylation of serine 5 and/or 7 by Kin28 is a major element in the recruitment of Rpd3S to active genes . These results are supported by data from the Hinnebusch lab who have shown that Rpd3S interacts with the phosphorylated form of RNAPII and has high affinity for doubly phosphorylated ( serine 2/5 ) CTD peptides in vitro [51] . Interestingly , CTD peptides carrying a single phosphate group on serine 5 or serine 2 respectively have a much weaker or no affinity to Rpd3S compared to doubly phosphorylated CTD peptides . Also noteworthy is the fact that Rpd3S is redistributed to promoter regions when Kin28 is inactive ( Figure 5A ) . As we will be discuss below , this might be due to a decrease in H3K36 methylation in that mutant , most likely due to a defective Bur1/2 recruitment , as suggested by [52] and [53] , leading to a defect in Rad6 phosphorylation . Our genome-wide data , together with these in vitro experiments , suggest that phosphorylation of the RNAPII CTD stimulates the recruitment of Rpd3S to transcribed genes while the elongation factor DSIF counteracts this recruitment . In mammalian cells , DSIF is phosphorylated by Cdk9 , a cyclin-dependent kinase associated with the elongation factor P-TEFb [54] . Cdk9 also phosphorylates the RNAPII CTD on serine 2 as well as other proteins including NELF . In yeast , the function of Cdk9 is fulfilled by two distinct kinases . Ctk1 mainly phosphorylates the RNAPII CTD and Bur1 targets Spt5 and Rad6 [55]–[57] . Inactivation of Bur1 has modest effects on phosphorylation of the RNAPII CTD , as shown by western blot and by ChIP-chip experiments ( data not shown; see also [52] , [57] , [58] ) . Because Bur1 phosphorylates Spt5 , the partner of Spt4 , we tested the effect of inactivating Bur1 activity on Rco1 recruitment . Not surprisingly , inhibiting Bur1 using an ATP-analog-sensitive strategy ( bur1AS ) has a profound effect on Rco1 occupancy . In the absence of a functional Bur1 , Rco1 is depleted from the coding region and redistributed to promoter regions ( Figure 5B , dashed line ) as was observed in the Kin28 mutant while the RNAPII level is mostly unchanged ( data not shown ) . Deleting the CTD of Spt5 ( spt5ΔC ) , the region phosphorylated by Bur1 , caused a similar , although milder , phenotype ( Figure 5B , dotted line ) . The stronger effect observed in the Bur1 inactivation experiment , compared to truncation of Spt5 , is most likely due to the fact that Bur1 has additional targets . For example , Bur1 phosphorylates Rad6 , an event that is required for the methylation of H3K4 and K79 by Set1 and Dot1 respectively [59]–[61] . In a triple mutant for Set1 , Set2 and Dot1 , we observed a shift of Rpd3S toward intergenic regions ( Figure S2B ) , suggesting that the redistribution of Rpd3S to promoter regions in Bur1-impaired cells is at least in part a consequence of Bur1's activity on Rad6 . Nevertheless , on the coding region , where our spt4Δ data showed that DSIF negatively regulates Rpd3S binding , we observed a clear decrease in Rco1 occupancy in both bur1AS and spt5ΔC . These data strongly suggest that phosphorylation of Spt5 by Bur1 negatively regulates the activity of DSIF on Rpd3S recruitment . While we cannot completely rule out the possibility that some of the effect observed in the bur1AS strain is due to an effect of Bur1 on the RNAPII CTD , our data on spt5ΔC rather suggest that Bur1 regulates Rpd3S recruitment by regulating DSIF . Interestingly , phosphorylation of Spt5 by Bur1 was previously shown to stimulate its activity as an elongation factor while our data suggest that it inhibits its activity as a negative regulator of Rpd3S recruitment . It will therefore be interesting to see whether these activities are linked .
Because H3K36 methylation correlates with transcription , it is generally accepted to be present at all transcribed genes; Rpd3S was therefore expected to follow the same pattern . Our finding that many transcribed genes do not show any signs of Rpd3S binding despite the presence of H3K36me3 was therefore a considerable surprise . This has important implications since it suggests that not all genes are equally protected against cryptic initiation . Other mechanisms exist that prevent cryptic transcription [62] , so it will be interesting to investigate how these various activities share the labor of protecting the genome from aberrant transcription . Another peculiarity of Rpd3S is that it is preferentially enriched on the coding regions of genes where Rpd3L is found on the promoter: we rarely find Rpd3S on genes where Rpd3L is not present . The presence of Rpd3L on the promoter of active genes is counterintuitive given its role as a co-repressor but it has nevertheless being observed before [17] . The co-occurrence of Rpd3L at promoter and Rpd3S on the ORFs of the same genes suggests that Rpd3L may play a role in targeting Rpd3S . One possibility would be that Rpd3S emerges from Rpd3L during the transition from initiation to elongation . The transition to elongation may trigger the exchange of subunits to transform Rpd3L into Rpd3S . Our finding that DSIF and RNAPII CTD phosphorylation are involved in the proper recruitment of Rpd3S to chromatin suggests that they may play a role in that process . Unexpectedly , we also observe Rpd3L on the coding regions of the genes where Rpd3S is also present . This last result was surprising in that it does not agree with the generally accepted function of Rpd3L as a promoter-recruited co-repressor . These data argue for a new model where subunits of both Rpd3L and Rpd3S coexist on actively transcribed coding regions . According to the model described above , the presence of Rpd3L subunits may reflect an imperfect exchange of subunits during the transition from initiation to elongation . Alternatively , it may reflect the presence of an Rpd3 “super large” complex that contains subunits of both Rpd3S and Rpd3L . In such a model , Rpd3S ( as we normally see it ) would only exist in solution and would represent a module that joins Rpd3L after initiation . This hypothesis is supported by data provided by Collins et al . [63] who combined and re-analyzed previously published mass spectrometry data [64] , [65] to generate a high-accuracy yeast protein interaction dataset . They found that several subunits of Rpd3L can be co-purified using Rco1 as bait . Reciprocally , subunits of Rpd3S can be co-purified with Rpd3L subunits used as baits . These data argue that some interactions between Rpd3S and Rpd3L do exist in the cell . Since the complexes analyzed in these studies were purified from the soluble cellular fraction , the relative low abundance of these inter-complex interactions may be explained by the fact that these complexes normally exist together only on chromatin . The recent development of techniques to purify protein complexes from chromatin [66] , [67] may help testing these intriguing possibilities . We show that the recruitment of Rpd3S to active genes does not require H3K36 methylation , Set2 or the Eaf3 chromodomain . Interestingly , Rco1 binding is not abolished in a triple deletion mutant for all the known yeast HMTs ( Set1 , Set2 and Dot1 ) , ruling out the possibility that the previously described interaction between H3K4me3 and Eaf3 [38] ( or an eventual interaction with methylated H3K79 ) is compensating for the loss of H3K36 methylation in set2Δ and H3K36A mutants . Moreover , although the methylation state of H3K36 does not affect Rpd3S binding at highly transcribed genes , it has an effect on the occupancy of Rpd3S at ORFs showing lower transcription levels . This suggest that , despite not providing the main recruitment signal , H3K36 methylation provides some stabilizing effect on the association of Rpd3S with chromatin . While Rpd3S recruitment is largely independent on H3K36 methylation , the activity of Rpd3S , however , depends on the integrity of the Set2/Rpd3S pathway , namely Set2-dependant H3K36 methylation , the Rco1 PHD and the Eaf3 CHD . We therefore propose that the main function of H3K36 methylation is to “activate” Rpd3S after it has been recruited by virtue of its association with the RNAPII CTD . This may involve the anchoring of Rpd3S on chromatin via the Eaf3-H3K36me interaction . Our data clearly show that phosphorylation by Kin28 is absolutely required for the association of Rpd3S with the coding region of transcribed genes . This suggests that phosphorylation of serine 5 provides the signal for the association of Rpd3S with early elongating RNA polymerases II molecules . Since Kin28 was recently shown to phosphorylate serine 7 in addition to serine 5 , we cannot rule out the possibility that serine 7 phosphorylation also contributes to the binding of Rpd3S to RNAPII . Using a ctk1Δ strain , we also show that phosphorylation of serine 2 also contributes -although to a lesser extend than serine 5- to the association of Rpd3S with transcribed genes . These results are in perfect agreement with a recent paper from the Hinnebusch group who showed that RNAPII CTD peptides harboring a phosphate group on both serines 5 and 2 have a high affinity for Rpd3S in vitro while a single phosphate on serine 5 has a reduced affinity [51] . More importantly , they were not able to detect any interaction of Rpd3S with non-phosphorylated CTD peptides , and peptides carrying a single phosphate on serine 2 barely show any affinity with Rpd3S [51] . We therefore propose that Rpd3S is recruited to active genes via interaction with the phosphorylated RNAPII CTD . However , this complex remains inactive until it has been “anchored” on chromatin via H3K36 methylation . As discussed above , this phenomenon does not appear to occur equally on all genes . Negative regulation of these interactions by the DSIF elongation complex modulates association of Rpd3S with genes , therefore creating situations where active genes with high level of Spt4 carry much less Rpd3S than expected from their transcriptional level . Interestingly , we found that deletion of SPT4 , a subunit of DSIF , has a profound effect on Rpd3S occupancy in vivo . In the absence of Spt4 , Rpd3S associates with the genome in a manner that correlates better with transcription than in wild type cells . This suggests that DSIF is involved in regulating the amount of Rpd3S on transcribed genes . DSIF appears to prevent the association of Rpd3S to a subset of transcribed genes , and even at genes occupied by Rpd3S , DSIF also plays a role since it prevents the hyper-accumulation of the HDAC in the 3′end of the gene . How exactly an elongation factor can regulate the association of a HDAC with elongating RNAPII remains obscure but we envision several mechanisms by which it may operate: 1 ) Rpd3S may directly or indirectly interact with DSIF; 2 ) The association of DSIF with the elongation complex may prevent the association of the HDAC; 3 ) DSIF may modulate the speed of the elongation complex in a way that makes it less favorable for Rpd3S binding; 4 ) DSIF may also impinge on the phosphorylation of the RNAPII CTD . More complex mechanisms may also be envisioned . For instance , Pin1 , a proline isomerase that may modify the RNAPII CTD , binds to both , phosphorylated DSIF [68] and Rpd3/Sin3 [69] . CTD isomerisation may therefore be involved in the regulation of the recruitment of Rpd3S . In human and yeast , DSIF is regulated by phosphorylation of the CTD of its Spt5 subunit . In yeast , this phosphorylation is mediated by the Bur1 cyclin-dependent kinase [56] , [57] . We therefore tested whether phosphorylation of Spt5 by Bur1 affects Rpd3S occupancy in vivo . As expected , both the catalytic inactivation of Bur1 and the removal of its substrate ( by deleting the CTD of Spt5 ) have a dramatic impact on Rpd3S occupancy . Both these mutants indeed show a decreased in Rpd3S occupancy along genes . These data suggest that phosphorylation of Spt5 by Bur1 negatively regulate the activity of DSIF on Rpd3S recruitment . In addition , the inactivation of Bur1 also causes Rpd3S to redistribute to promoter regions , a phenomenon that we also observed in set2Δ cells . Since Bur1 phosphorylates Rad6 , which is also required for H3K36 methylation by Set2 , it appears likely that this redistribution in Bur1 mutants is due to its effect on H3K36 methylation via Rad6 . Why a lack in H3K36 methylation leads to an association of Rpd3S with promoters remains unknown , but is in agreement with the previous observation that histone acetylation decreases at promoters in these mutants . Taken together , our in-depth analysis of Rpd3S genomic occupancy has revealed several key insights about its recruitment to genes in vivo . Our study highlights a complex network of protein-protein interactions mediated by phosphorylation of several substrates by at least three kinases . Interestingly , there is previous evidence in the literature linking these kinases to the function of DSIF . First , P-TEFb ( Cdk9 ) and Bur1 can phosphorylate DSIF [56] , [57] , [70] . Second , Kin28 , Ctk1 and Bur1 exhibit synthetic genetic interactions with Spt4 and Spt5 [71] . And third , the recruitment of Bur1 is stimulated by Kin28 [52] , [72] . Finally , both DSIF and Kin28 have been shown to stimulate the recruitment of the Paf1 complex to the elongation complex [57] , [73] . A lot more work will be required before we completely understand the interplay between these factors and Rpd3S .
All strains used in this study are listed in Table S1 . All strains were grown in 50mL of YPD to an OD600 of 0 . 6–0 . 8 before crosslinking , unless otherwise indicated . For ChIP-chip , most strains were crosslinked with 1% formaldehyde for 30 min at room temperature on a wheel . The Rco1-9myc strains were crosslinked with 1% paraformaldehyde for 30 min at room temperature followed by 90 min at 4°C on a wheel . ATP analog-sensitive strains were treated with 6 µM of NAPP1 for 15 minutes prior to crosslinking . ChIP experiments were performed as per [74] , with minor modifications . For myc-tag ChIP , we used 5µg of 9E11 antibody coupled to 2×107 pan-mouse IgG DynaBeads ( Invitrogen ) per sample . For histone H4K5 acetylation ChIP , we used 4µL of a rabbit serum ( Upstate 07-327 ) coupled to 2×107 protein G DynaBeads per sample . Histone H3K36me3 was immunoprecipitated with 4µL of antibody ( Abcam Ab9050 ) coupled to 2×107 protein G DynaBeads per sample . Histone H4 levels were assayed with 2µL of an antibody raised against recombinant yeast histone H4 ( a gift from Alain Verreault ) coupled to 2×107 protein G DynaBeads per sample . RNAPII ChIPs were done using 2µL of 8WG16 antibody coupled to 2×107 pan-mouse IgG DynaBeads per sample . Note that in our ChIP-chip assays , 8WG16 generates profiles that are nearly identical as using a tagged RNAPII ( data not shown ) . 8WG16 is therefore used here to measure total RNAPII levels on genes . RNAPII CTD serine 2 phosphorylation was assayed using 5µL of H5 antibody ( Covance Research MMS-129R-200 ) coupled to 2×107 protein G DynaBeads per sample . The microarrays used for location analysis were purchased from Agilent Technologies ( Palo Alto , California , United States ) and contain a total of 44 , 290 Tm-adjusted 60-mer probes covering the entire genome for an average density of one probe every 275 bp ( ±100 bp ) within the probed regions ( catalog # G4486A and G4493A ) . Myc-tag ChIPs were hybridized against ChIPs from isogenic strains that did not contain the tag as controls . Acetylation , RNAPII and histone H4 ChIPs were hybridized against a sample derived from 400ng of input ( non-immunoprecipitated ) DNA . Acetylation levels were normalized to histone H4 levels by subtracting Log2 ( Histone H4/input ) from Log2 ( H4K5 acetyl/input ) . All microarray experiments described in this work are listed in Table S2 , the processed data are available in Datasets S1 , S2 , S3 , S4 , and the raw data have been deposited into the GEO database ( Accession # GSE22636 ) . The data were normalized and biological replicates were combined using a weighted average method as described previously [74] . The log2 ratio of each spot of combined datasets was then converted to Z-score , similar to Hogan et al . [75] , to circumvent the large differences in the immunoprecipitation efficiencies of the different factors . Visual inspection of the Z-scores was carried out on the UCSC Genome Browser ( http://genome . ucsc . edu/ ) . All data analyses described here were done using data from protein-coding genes longer than or equal to 500 bp . Median Z-score values for promoter and complete length of each annotation from SGD ( version Feb . 02 2008 ) were calculated without interpolation ( Dataset S5 ) and used in our clustering and Pearson correlation analyses . Promoters are defined as the shortest of either 250bp or half the intergenic region ( half-IG ) relative to the reference gene's 5′ boundary . Self-organizing map ( SOM ) clustering was done with the Cluster software [76] and visualized with Java Treeview [77] . Only genes with no missing value were used for clustering . Gene mapping was performed as in Rufiange et al . [78] on selected groups of genes described in the text . Briefly , data were mapped onto the 5′ and 3′ boundaries in 50 bp windows for each half-gene and adjacent half-IG regions . A sliding window of 300 bp was then applied to the Z-scores to smooth the curve . | Acetylation of histone N-terminal tails occurs on nucleosomes as a gene is being transcribed , therefore helping the RNA polymerase II traveling through nucleosomes . Histone acetylation , however , has to be reversed in the wake of the polymerase in order to prevent transcription from initiating at the wrong place . Rpd3S is a histone deacetylase complex recruited to transcribed genes to fulfill this function . The Rpd3S complex contains a chromodomain that is thought to be responsible for the association of Rpd3S with genes since it interacts with methylated histones , a feature found on transcribed genes . Here , we show that the recruitment of Rpd3S to transcribed genes does not require histone methylation . We found that Rpd3S is actually recruited by a mechanism implicating the phosphorylation of the RNA polymerase II C-terminal domain and that this mechanism is regulated by a transcriptional elongation complex called DSIF . We propose that the interaction between the Rpd3S chromodomain and methylated histones helps anchoring the deacetylase to its substrate only after it has been recruited to the elongating RNA polymerase . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/histone",
"modification",
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/functional",
"genomics",
"genetics",
"and",
"genomics/chromosome",
"biology",
"molecular",
"biology/transcription",
"elongation",
"genetics",
"and",
"genomics/epigenetics",
"molecular",
"biology/chromatin",
"structure"
] | 2010 | DSIF and RNA Polymerase II CTD Phosphorylation Coordinate the Recruitment of Rpd3S to Actively Transcribed Genes |
Working memory ( WM ) is the part of the brain's memory system that provides temporary storage and manipulation of information necessary for cognition . Although WM has limited capacity at any given time , it has vast memory content in the sense that it acts on the brain's nearly infinite repertoire of lifetime long-term memories . Using simulations , we show that large memory content and WM functionality emerge spontaneously if we take the spike-timing nature of neuronal processing into account . Here , memories are represented by extensively overlapping groups of neurons that exhibit stereotypical time-locked spatiotemporal spike-timing patterns , called polychronous patterns; and synapses forming such polychronous neuronal groups ( PNGs ) are subject to associative synaptic plasticity in the form of both long-term and short-term spike-timing dependent plasticity . While long-term potentiation is essential in PNG formation , we show how short-term plasticity can temporarily strengthen the synapses of selected PNGs and lead to an increase in the spontaneous reactivation rate of these PNGs . This increased reactivation rate , consistent with in vivo recordings during WM tasks , results in high interspike interval variability and irregular , yet systematically changing , elevated firing rate profiles within the neurons of the selected PNGs . Additionally , our theory explains the relationship between such slowly changing firing rates and precisely timed spikes , and it reveals a novel relationship between WM and the perception of time on the order of seconds .
Various mechanisms have been proposed to model the main aspect of neural activity — elevated firing rates of a cue-specific population of neurons — observed during the delay period of a working memory ( WM ) task [1]–[4] . These include reentrant spiking activity [5] , intrinsic membrane currents [6] , NMDA currents [7]–[10] , and short-term synaptic plasticity [7] , [11] , [12] . These mechanisms , however , fail to explain other aspects of neural correlates of WM [13] , and they have been demonstrated to work only with a limited memory content where the number of items represented in long-term memory is small , i . e . , they hold in WM a few items ( limited capacity [14] ) out of only a conceivable few ( limited memory content ) . Memories in these simulated networks are often represented by carefully selected , largely non-overlapping groups [15] of spiking neurons [11] . Indeed , extending the memory content in such networks increases the overlap between the memory representations ( unless the size of the network is increased , too ) , and activation of one representation spreads to others , resulting in uncontrollable epileptic-like “runaway excitation” . The narrow memory content is at odds with experimental findings that neurons participate in many different neural circuits ( see e . g . [16]–[18] ) and , therefore , are part of many distinct representations that form a vast memory content for WM . These limitations may be overcome by a model that accounts for the precise spike-timing nature of neural processing . We propose a model in which memories are represented by extensively overlapping neuronal groups that exhibit stereotypical time-locked , but not necessarily synchronous , firing patterns called polychronous patterns [19] ( see also [20] ) . In Figure 1 , we use a small network to illustrate this concept: Two distinct patterns of synaptic connections ( red and black connections in Figure 1A–1C ) with appropriate axonal conduction delays form two distinct polychronous neuronal groups ( PNGs ) . Notice that these PNGs are defined by distinct patterns of synapses , and not by the neurons per se , which allows the neurons to take part in multiple PNGs and enables the same set of neurons to generate distinct stereotypical time-locked spatiotemporal spike-timing patterns ( see Figure 1B and 1C ) . PNGs arise spontaneously [19] , [21] in simulated realistic cortical spiking networks shaped by spike-timing dependent plasticity [22] ( STDP ) . Another distinctive feature of our theory is that synaptic efficacies are subject to associative short-term changes , that is , changes that depend on the conjunction of pre- and post-synaptic activity ( see [23]–[25] for experimental findings supporting this postulation ) . We simulated two different mechanisms: ( 1 ) associative short-term synaptic plasticity via short-term STDP , where short-term synaptic changes — that decay to baseline within a few seconds — are induced by the classical STDP protocol ( Figure 2A ) ; and ( 2 ) the short-term amplification of synaptic responses via simulated NMDA spikes [8] at the corresponding dendritic sites ( Figure 2B–2D ) . The latter mechanism is also pre- and postsynaptic activity dependent: Pre-synaptic spikes alone activate postsynaptic NMDA receptors , yet only generate small excitatory postsynaptic potentials ( EPSPs ) at the dendritic compartment ( Figure 2D , red trace ) because of the magnesium block of the NMDA receptors . Postsynaptic spikes , however , induce dendritic membrane potential depolarization and removal of the magnesium block . Hence , the dendritic compartment flips into up-state . While in the up-state , each presynaptic spike results in a large-amplitude response ( often called an NMDA spike ) that can propagate from the dendritic compartment to the soma and enhance the efficacy of synaptic transmission in eliciting somatic spikes . The short-term enhancement of synaptic efficacy is similar to that recorded in vitro [26] and in detailed simulations of Hodgkin-Huxley-type conductance-based models [27] . ( See Figure 2B–2D and Methods for details . ) We found that the exact form of such short-term synaptic changes is not important for the WM functionality presented in this paper ( see Results ) , as long as these changes selectively affect synapses according to the relative spike timing of pre- and post-synaptic neurons . For example , activation of the red PNG in Figure 1 temporarily potentiates the red synapses and not the black ones ( Figure 1B and 1C ) . This differs from the standard short-term synaptic facilitation or augmentation used in previous WM models [7] , [11] , which are not associative , and hence non-selectively affect all synapses belonging to the same presynaptic neuron . In the model presented here , PNGs get spontaneously reactivated due to stochastic synaptic noise . Short-term strengthening of the synapses of selected PNGs can bias these reactivations , i . e . , increase the reactivation rate of the selected PNGs , which results in activity patterns similar to those observed in vivo during WM tasks [1]–[4] , [13] . Additionally , even though PNGs share neurons with other PNGs , the activity of one PNG does not spread to the others . Therefore , frequent reactivation of a selected PNG does not initiate uncontrollable activity in the network . In this way , the WM mechanism presented here can work in finite networks with large memory content . This is different from previous models [11] , [28]–[31] where large memory content and maintenance of several memory items can only be achieved by a drastic increase in the size of the network or the number of connections between neurons .
We implemented our model in a simulated network of 1000 spiking neurons [32] , where 80% of the neurons are regular spiking pyramidal neurons and 20% are GABAergic fast spiking interneurons . The probability that any pair of neurons are connected equals 0 . 1 . Excitatory synaptic connections have a random distribution of axonal conduction delays in the [1…20] ms range [19] , [33]–[35] . Excitatory synaptic efficacy is subject to both associative short-term plasticity and long-term STDP [22] . Maximum synaptic strengths are set so that three simultaneously arriving pre-synaptic spikes are needed to reliably elicit a post-synaptic spike . ( The Methods section has detailed description of the network , neuron model , and synaptic plasticity . ) Approximately 8000 strongly overlapping PNGs emerge spontaneously in such network ( Figure 3 ) and we select a few to demonstrate how these mechanisms ( PNG formation and associative short-term plasticity ) can serve to maintain WM , and how they can account for the other related experimental findings . To initiate sustained neuronal activity that characterizes WM , we select ( cue ) a random PNG and stimulate its neurons in the sequence that characterizes the PNG's polychronous pattern . That is , we stimulate the intra-PNG neurons sequentially with the appropriate polychronous pattern 10 times during a one second interval ( see e . g . Figure 4A ) to temporarily increase the intra-PNG synaptic efficacy ( see Methods ) . The red dots in the spike raster in Figure 4A indicate spikes of the selected target PNG . The initial stimulation of the target PNG resulted in short-term strengthening of the intra-PNG synapses via associative short-term plasticity , but had little effect on the other synapses in the network ( Figure 4A , “short-term synaptic change” curves ) . Upon termination of the stimulation , the temporarily facilitated intra-PNG synapses and the noisy synaptic inputs resulted in sporadic reactivations of different segments of the target PNG , often leading to the reactivation of the rest of the polychronous sequence ( seen as red vertical stripes in the raster in Figure 4A and magnified in Figure 4B ) . Each such reactivation of the target PNG triggers further strengthening of its synapses , thereby maintaining the target PNG in the active state for tens of seconds . Notably , the active maintenance of a PNG in WM does not depend on a reverberant/looping circuit , but it emerges as a result of the interplay between non-specific noise ( which spontaneously triggers activation of PNGs ) and short-term strengthening of the appropriate synapses ( that makes subsequent reactivations of the target PNG more likely ) . There are frequent gaps of hundreds of milliseconds between spontaneous reactivations of the target PNG , clearly seen in Figure 4A , but occasional reactivation is necessary to maintain the PNG in WM . Without the reactivations , the initial short-term strengthening of intra-PNG synapses decays quickly ( illustrated in Figure 4A , “decay without replay” curve ) . Figure 4F shows that almost all of the thousands of emerged PNGs , if stimulated , remained activated for more then ten seconds in WM ( average seconds ) . Since spontaneous reactivations of the target PNG in WM are stochastic , timing of the spiking activity of each neuron in a PNG also looks random when considered in isolation . The coefficient of variation ( CV ) of inter-spike intervals ( ISIs ) , i . e . , the variability of ISIs ( see Methods ) , is higher for individual intra-PNG neurons when the PNG is in WM [36] ( Figure 4C and Figure S6 ) . This phenomenon is due to the systematically changing and non-stationary mean firing activities and mean ISIs of the intra-PNG neurons during replay ( see section below ) . Relative intra-PNG timing at the millisecond timescale is , however , maintained during replay , as can be seen in the magnified spike rasters in Figures 4B and 5C . This is a major feature that distinguishes our approach from earlier approaches that posit synchronous [11] or totally asynchronous [7] spiking , and this feature allows our model to have a vast repertoire of overlapping PNGs , i . e . , large memory content . Cross-correlograms ( CCG ) of simulated intra-PNG neuronal pairs also reveal the precisely timed nature of their spiking activity , as well as the context-dependent changes in functional connectivity linking these neurons: The red CCG in Figure 4D is recorded while the target PNG is in WM , and it has a peak around 5 ms , whereas the blue CCG ( recorded later in a different session , when the PNG is not activated ) is flat . A similar dependence of CCGs of spiking activity on the behavioral state of the network biased by sensory cues was reported in medial prefrontal neurons [37] . The average multiunit firing rate of the neurons forming the target PNG following activation is around 4 Hz , much higher than that of the rest of the network , which is about 0 . 3 Hz ( Figure 4A , “multiunit firing rate” red vs . blue solid lines ) . The average firing rate histograms of most intra-PNG neurons show distinct temporal profiles that repeat from trial to trial ( Figure 4E and Figure S4 ) : Some neurons only respond to the initial stimulation ( Figure 4E n392 ) ; some have ramping or decaying firing rates ( n652 ) ; whereas others have their peak activity seconds after the stimulus offset ( n559 ) . Neurons that are not part of the target PNG show uniform low firing rate activity across the whole trial ( n800 ) . These systematically varying , persistent temporal firing profiles are similar to those observed experimentally in vivo in frontal cortex during the delay period of the WM task [1] , [3] , [38] , [39] , but no previous spiking model of WM could reproduce them . To get the results presented in Figures 4E , only an initial segment of the target PNG is activated during the selection ( cueing ) process ( see Methods ) . Therefore , only the synapses forming the initial segment of the target PNG get temporarily potentiated . Hence , directly after stimulation/cuing only the neurons in the initial segment of the target PNG get more frequently reactivated as propagation of activation along the PNG dies out somewhere in the middle of the PNG without activating the neurons at the back . As frequent spontaneous reactivations persist , more and more synapses undergo short-term STDP , and more and more neurons from the end of the target PNG start to participate in the reactivations . Activities of such neurons show ramping up firing rates ( Figure 4E n559 ) . Conversely , neurons in the initial segment of the PNG may not participate in enough reactivations and , therefore , synapses to those neurons decay back to their baseline strength , resulting in a ramping down firing profile ( n392 Figure 4E ) . In general , the slowly changing firing rates are generated by spontaneous incomplete activations within the target PNG: Neurons that are initially stimulated typically do not get reactivated or get reactivated only shortly after the target PNG stimulation and , therefore , exhibit ramping down firing profile ( n392 , n652 ) ; In contrast , those that join just later the wave of reactivation ( Figure S4E ) express ramping up ( and later down ) firing activity ( n559 ) . These stereotypical firing rate profiles may be utilized to encode time intervals [38] , [40] . For example , a motor neuron circuit that needs to execute a motor action 10 seconds after a GO signal might have strong connections from neurons such as n559 in Figure 4E , and be inhibited by the activity of neurons such as n652 . Moreover , a sequence of behaviors could be executed by potentiating connections from multiple subsets of the PNG to multiple motor neuron circuits ( e . g . , via dopamine-modulated STDP [41] ) . Activations of multiple representations in WM , as illustrated in Figure 5 , could implement multiple timing signals and multiple sequences of actions . In a single network , multiple PNGs , i . e . , multiple memories , can be loaded and maintained in WM simultaneously despite large overlap in their neuronal composition . In Figure 5A we stimulate two PNGs sequentially ( out of the thousands available PNGs ) . The target PNGs consisted of 220 and 191 neurons each , and have 66 neurons in common . The intra-PNG neurons , however , fire with different timings relative to the other neurons within each PNG ( Figures 5C and 5D ) . Therefore , there is little or no interference , and both PNGs are simultaneously kept in WM for many seconds . The model can hold several items in WM but eventually its performance deteriorates with increased load ( note the sub-linear histogram in Figure 5B ) . To demonstrate that a novel cue can be loaded and kept in WM , we stimulated the network with a novel spike-timing pattern repeatedly every 15 seconds ( Figure 6 ) . Notice that this spiking pattern — triggered by the novel external cue — did not correspond to any of the existing PNGs' firing pattern . Each time the new pattern is presented to the network , the synapses between the stimulated neurons that fire with the appropriate order are potentiated due to long-term STDP . In addition , synapses to some other post-synaptic neurons that were firing by chance and have synaptic connections with converging conduction delays that support appropriate spike timing , are also potentiated [19] . Thus , the expansion of the network's memory content , i . e . , the formation of a new PNG representing the novel cue , occurs via the interplay of long-term STDP and repeated firing of neurons with the right spatiotemporal pattern . This pattern can be triggered by stimulation ( as shown in [19] ) , or it could result from autonomous reactivations due to WM mechanism ( as shown in Figure 6A and 6D ) . Therefore , the WM mechanism , by facilitating the reactivations of the new PNG , facilitates the formation of the new PNG . Despite that the new PNG consists both of neurons that received ( red dots in Figure 6D ) and of neurons that did not receive ( marked black in Figure 6D ) direct stimulation during the cue presentations/learning , in order to load and keep the cue in WM it is sufficient to stimulate those neurons that were directly stimulated during learning . The reactivation rate of the new PNG , 4 Hz , is similar to those observed in Figures 4 and 5 .
Results of our simulations are robust with respect to the mechanisms of associative short-term change of synaptic efficacies and to parameters of the model , such as short-term synaptic decay time constants ( see Figures 4 and 5; and Figure S1 ) ; probability of random synaptic inputs; or choice of the target PNGs ( Figure 4 and 5; see also Figures S3 and S4 , where we replicate the results of Figures 4 and 5 using PNGs that were manually generated and inserted in the network ( see Methods ) ) . The underlying currency of information in the theory presented here is the activation of a PNG . This , combined with an associative form of short-term changes of synaptic efficacies results in spontaneously emerging WM functionality: short-term synaptic changes bias the competition between PNG reactivations , and give rise to frequent spontaneous reactivations of the selected PNGs ( relative to the reactivation rate of the other PNGs ) , which are expressed as short polychronous events with preserved intra-PNG spike-timings . The simulations result in a network with large memory content , and produce neural activity consistent with those observed experimentally [1] , [3] . Our theory predicts that polychronous structures are essential for cognitive functions like WM , and such structures may be the basis for complex activity patterns observed in neocortical assemblies [42] and for memory replays involving , for example , prefrontal cortex , visual cortex , and hippocampus [43]–[45] . Additionally , this theory makes a testable prediction that changes in functional connectivity ( as in Figures 4D and 5D ) should be observed experimentally in vivo during WM tasks .
We use a model of spiking neurons [32] , [46] that was developed to satisfy two requirements: It is computational simple and efficient to implement in large-scale simulations , and it exhibits most of the types of the firing patterns recorded in animals in vitro and in vivo . We use the differential equations in the formwith the auxiliary after-spike resettingwhere v and are the membrane potential and recovery variables , respectively; and are parameters: , time scale of the recovery variable ; , sensitivity of the recovery variable to the sub-threshold fluctuations of the membrane potential ; , after-spike reset value of the membrane potential caused by the fast high-threshold conductances; , after-spike reset of the recovery variable caused by slow high-threshold and conductances . Various choices of these parameters result in various intrinsic firing patterns , including those exhibited by the known neocortical neurons . Here for regular spiking pyramidal neurons , and for GABAergic fast spiking interneurons . Derivation of these equations/parameters are explained in [32] , [46] . 80% of the neurons in our network are regular spiking pyramidal neurons and 20% of them are GABAergic fast spiking interneurons . A careful measurement of axonal conduction delays in the mammalian neocortex [33] , [35] showed that these delays could be as small as 0 . 1 ms and as large as 44 ms , depending on the type and location of the neurons . Moreover , the propagation delay between any individual pair of neurons is precise and reproducible with a sub-millisecond precision [33] , [34] . In our network ( similar to the network in [19] ) , excitatory synaptic connections have random axonal conduction delays in the [1…20] ms range , therefore , it can be considered as a subnetwork embedded into a large part of the prefrontal cortex . All inhibitory connections are set to have 1 ms delays . The probability that any pair of neurons are connected equals 0 . 1 . After running the simulation for five hours , providing only random synaptic input to the network , we analyzed the evolved network data — synaptic connections , axonal conductance delays , and synaptic strengths — using the methods described in [19] and found a total of spontaneously generated , strongly overlapping distinct PNGs; See Figure 3 for details on the emerging PNGs . We used these spontaneously emerging PNGs for the results shown in Figures 4 and 5 . Embedded in the noisy spike train are occasional precise spiking patterns corresponding to spontaneous reactivations of PNGs [19] . Since each such PNG has a distinct pattern of polychronous spiking activity , we use the pattern as a template to find the reactivation of the PNG in the spike train . A PNG is said to be activated when more than percent of its neurons fire according to the prescribed polychronous pattern with jitter . To select a specific PNG in WM , i . e , to temporarily increase the intra-PNG synaptic efficacy , we transiently stimulate its neurons sequentially with the appropriate spatiotemporal spike-timing pattern [48]–[50] . What enters WM is possibly gated by attention . To avoid modeling attentional mechanisms , we provide two different gating implementations: We also performed stochastic stimulations ( for both types of stimulations ) where the firing response probability of individual neurons to external stimulation was smaller then 1 and found the qualitative behavior of the network to be similar . For example , the response probability for the target neurons in Figure S2 is 0 . 8 . For the results presented in Figures 4E ( and 6 ) , and Figures S3 and S4 , not all the neurons of the target PNG were stimulated ( with the appropriate polychronous pattern ) but only the initial segment of the target PNGs ( 80 percent in Figure 4E; 10 percent in Figures S3 and S4 ) . The rest of the target neurons ( i . e . , neurons that were not stimulated but are part of the target PNG ) systematically joined the reactivation process . ( For detailed description , see the figure legends for Figures S3 and S4 . ) For the results presented in Figures S3 and S4 we inserted additional synapses in the randomly connected network in order to form 100 new PNGs . Activity of each such PNG lasted for 200 milliseconds and it consisted of 40 neurons . Each intra-PNG neuron has at least three converging synapses from other pre-synaptic intra-PNG neurons ( except for the first three neurons in the PNG ) . Throughout the whole simulation the network is stimulated with stochastic miniature synaptic potentials ( called “minis” ) , and it exhibits asynchronous noisy spiking activity . The average background multiunit firing rate is around 0 . 3 Hz for the simulations presented in the article . Qualitative behavior of the network is similar to a wide range of noisy background firing activity , which , however , cannot be too small , as some background activity is necessary to initiate spontaneous PNG reactivations ( see Figure 6 and Figure S5 ) , or too high , as that would interfere with neural activity propagation within the PNG . Spontaneously emerging PNGs in the simple network we used tend to be prone to noise . This means that the initiated activity in the PNG is less likely to propagate along the whole PNG in the presence of high background noise ( for excitatory neurons ) . This is because neurons that should respond ( fire ) to presynaptic activity and pass that activity to postsynaptic intra-PNG neurons are likely to be inhibited or be in their refractory period if there is too much background activity present in the network . Manually inserted PNGs can be engineered to have redundant connections , i . e , postsynaptic neurons have more presynaptic connections ( from multiple presynaptic neurons ) than minimally required to fire these postsynaptic neurons . This redundancy can make these PNGs much more robust to noise: the inability of a presynaptic neuron to fire ( e . g . due to inhibition ) is less likely to prevent the propagation of activity in the PNG , as there are likely other presynaptic intra-PNG neurons firing and passing the activity to the same postsynaptic target . The first 20 seconds after stimulus presentation offset of the spike trains of the target PNG were used for inter-spike interval ( ISI ) analysis presented in Figure 4C and Figure S6 , red histograms . The data was collected over multiple trials . The coefficient of variation ( CV ) measures the variation in the neurons' ISIs: , i . e . , CV equals the standard deviation of ISIs divided by the mean ISI . , a local measure for coefficient of variation , used for Figure S6 is less biased by non-stationary ISIs . is computed by comparing each ISI ( ) to the subsequent ISI ( ) to evaluate the degree of variability of ISIs in a local manner: , where . These measures are identical to those used in [36] . | Working memory ( WM ) is the part of the brain's vast memory system that provides temporary storage and manipulation of the information necessary for complex cognitive tasks , such as language comprehension , learning , and reasoning . Despite extensive neuroscience research , its mechanism is not clearly understood . We exploit a well-known feature of the brain — its ability to use precisely timed spiking events in its operation — to show how working memory functionality can emerge in the brain's vast memory repertoire . Our neural simulations explain many features of neural activity observed in vivo during working memory tasks , previously thought to be unrelated , and our results point to a relationship between working memory and other mental functions such as perception of time . This work contributes to our understanding of these brain functions and , by giving testable predictions , has the potential to impact the broader neuroscience research field . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"neuroscience/theoretical",
"neuroscience",
"computational",
"biology/computational",
"neuroscience"
] | 2010 | Spike-Timing Theory of Working Memory |
The opportunistic fungal pathogen Candida albicans frequently causes diseases such as oropharyngeal candidiasis ( OPC ) in immunocompromised individuals . Although it is well appreciated that the cytokine IL-17 is crucial for protective immunity against OPC , the cellular source and the regulation of this cytokine during infection are still a matter of debate . Here , we directly visualized IL-17 production in the tongue of experimentally infected mice , thereby demonstrating that this key cytokine is expressed by three complementary subsets of CD90+ leukocytes: RAG-dependent αβ and γδ T cells , as well as RAG-independent ILCs . To determine the regulation of IL-17 production at the onset of OPC , we investigated in detail the myeloid compartment of the tongue and found a heterogeneous and dynamic mononuclear phagocyte ( MNP ) network in the infected tongue that consists of Zbtb46-Langerin- macrophages , Zbtb46+Langerin+ dendritic cells ( DCs ) and Ly6C+ inflammatory monocytes . Of those , the Langerin+ DC population stands out by its unique capacity to co-produce the cytokines IL-1β , IL-6 and IL-23 , all of which promote IL-17 induction in response to C . albicans in the oral mucosa . The critical role of Langerin+ DCs for the innate IL-17 response was confirmed by depletion of this cellular subset in vivo , which compromised IL-17 induction during OPC . In conclusion , our work revealed key regulatory factors and their cellular sources of innate IL-17-dependent antifungal immunity in the oral mucosa .
As part of the upper gastrointestinal tract , the oral cavity is colonized by microbes and constitutes an important entry point for hazardous pathogens . However , despite the relevance of the oral mucosa as a first site of interaction between microbes and the host , it remains little studied and its cellular composition is not well characterized . Oropharyngeal candidiasis ( OPC ) is a common infection of the oral cavity mediated by the opportunistic fungal pathogen Candida albicans in immunocompromised individuals [1] . It frequently develops as a consequence of impaired immune function due to administration of steroids and other immunosuppressant agents , or because of underlying diseases such as AIDS or primary immunodeficiencies [1] . The recent study of hereditary factors predisposing to OPC and other forms of mucocutaneous candidiasis determined the relevance of the interleukin-17 ( IL-17 ) pathway as a key mechanism for protective immunity against this disease . Genes directly associated with disease include those encoding the IL-17 receptor subunits IL-17RA [2] and IL-17RC [3] , the signaling adaptor Act1 ( also referred to as CIKS or TRAF3IP2 ) [4] and the cytokine family member IL-17F [2] , but also genes encoding transcription factors involved in the regulation of IL-17 production , such as STAT1 [5 , 6] , STAT3 [7–9] and RORγt [10] . Work on experimental mouse models further confirmed the important role of IL-17 , in particular IL-17A and IL-17F , in antifungal defense [11–13] . IL-17 is thought to act by promoting the antimicrobial function and the epithelial integrity of barrier tissues [14 , 15] . Although there is little doubt about the relevance of IL-17 in host-defense against C . albicans , the tissue-specific regulation of IL-17 production during candidiasis remains not well understood . Th17 cells are the major source of IL-17A and IL-17F in response to C . albicans [16–19] . The rapid induction of IL-17A and IL-17F within 24 hours post-infection in experimentally infected mice suggested that innate cells also participate in cytokine production in the infected mucosa . Indeed , previous work from our group demonstrated that Rag1-/- animals , lacking T and B cells , control the fungus and recover from infection within a week [13] . Genetic deletion or antibody-mediated depletion of CD90+ or IL-2 receptor gamma chain-dependent cells in Rag1-/- mice however rendered the animals susceptible to OPC to a degree similar to IL-17RA-deficient mice [13] . Hence , our oberservations strongly argued for the involvement of a RAG-independent cellular source of IL-17 . This result was challenged by a study using an IL-17A reporter mouse strain suggesting that T cells produce IL-17 within 24 hours of primary infection [20] . Here we employed a flow cytometry approach to directly visualize IL-17A protein expression in the tongue of infected mice and identified three separate populations of IL-17-producing CD90+ leukocytes that act in an at least partially redundant manner during acute OPC . Moreover , we dissected and tested the functional relevance of different mononuclear phagocytes and IL-17 instructive cytokines in the infected tissue and thereby revealed the key determinants that orchestrate the innate IL-17 response against C . albicans in the oral mucosa .
We set out to characterize cells with the potential to produce IL-17A and IL-17F in response to C . albicans at the site of infection . To do so , we analyzed tongue leukocytes from RorcCreR26ReYFP reporter mice for the expression of eYFP as a reporter for the lymphocyte-associated transcription factor RORγt . We detected a distinct population of eYFP+ cells that uniformly co-expressed CD90+ in both naïve and infected tongues and comprised αβ T cells , γδ T cells and TCR- ILCs ( Fig 1A ) . Innate IL-17 production in the murine tongue was so far never demonstrated directly and at the single cell level . We therefore established a protocol to visualize IL-17A and IL-17F protein in the C . albicans-infected tongue by combining in vivo Brefeldin A administration and intracellular cytokine staining ( S1A Fig ) . By doing so we detected a well-defined population of CD90+IL-17A+ cells in the tongue of infected but not naïve wild type ( WT ) mice ( Fig 1B ) . Because Il17a and Il17f transcript levels are maximal on day 1 post-infection with C . albicans strain SC5314 [13] , we focused our analysis on this time point . Reminiscent of our analysis of RorcCreR26ReYFP mice , the CD90+IL-17A+ population comprised three subsets: αβ T cells , γδ T cells and TCR- ILCs ( Fig 1B ) . The specificity of our IL-17A staining in CD90+ cells was confirmed by applying the same experimental conditions to Il17af-/- mice , which served as a biological negative control ( S1B Fig ) . In WT mice , the majority of IL-17A+ cells co-produced IL-17F ( S1C Fig ) . NKT cells were not found to contribute significantly to IL-17A production in the infected tongue , as only very few IL-17A+ cells could be stained with CD1d tetramers ( S1D Fig ) . Quantification of the CD90+ and CD90+IL-17A+ subpopulations revealed an expansion of all subsets within the first 24 hours of infection ( Fig 1C ) . Consistent with this , all IL-17A+ cells stained positive for the marker Ki67 ( Fig 1D and S2 Fig ) , suggesting in situ proliferation of the three IL-17A-producing cellular subsets . To characterize the IL-17A-producing cells in the tongue of OPC-infected mice in more detail , we analyzed phenotypic and activation markers on their surface . Consistent with their TCR expression , αβ and γδ T cells , but not TCR- ILCs , co-expressed CD3 ( Fig 1D and S2 Fig ) . All three IL-17A+ subsets displayed an activated phenotype as evidenced by their high expression of CD44 and partial expression of CD69 . None of the IL-17A-expressing cellular subsets expressed CD122 , NCR1 , CCR6 or MHCII ( S2 Fig ) . To further define the ILC compartment , we examined RAG1-deficient mice . RorcCreR26ReYFP fate reporter mice , crossed to a RAG1-deficient background , harbored a clear population of eYFP+ cells in the naïve tongue that were uniformly CD90+ ( Fig 1E ) . Infection of ( reporter-less ) Rag1-/- mice revealed , as expected , an overall reduction of total CD90+ and CD90+IL-17A+ cells due to the absence of TCRβ+ and TCRγδ+ cells . Yet , the number of IL-17A-producing cells was still significantly increased in infected as compared to naïve WT mice , which was attributed to the ILC subset ( Fig 1F and 1G ) . Overall , our data demonstrate that direct visualization of IL-17A protein production by intracellular staining ex vivo revealed the involvement of three distinct CD90+ leukocyte subsets during OPC . IL-23 is a key regulator of IL-17 immunity . To investigate the impact of IL-23 on IL-17A production by the three distinct cellular subsets during OPC , we infected Il23a-/- animals in comparison to WT controls and assessed the changes in absolute numbers of IL-17A producing CD90+ cells and the corresponding TCRβ+ , TCRγδ+ and ILC subsets in the tongue on day 1 post infection . While the CD90+ populations overall remained unchanged in Il23a-/- mice in comparison to WT controls , the number of IL-17A-positive cells was significantly reduced ( Fig 2A and 2B ) . This is consistent with previous studies analyzing the dependence of overall IL-17A and IL-17F expression on IL-23 in the infected organ [13] and the impact of IL-23 on fungal control [11 , 13] . Of the three identfitied IL-17A-producing cellular subsets , the TCRβ+ subset was most strongly affected , but also the TCRγδ+ and ILC populations showed a clear trend towards reduced IL-17A production in IL23-deficient mice compared to WT controls . Instruction of IL-17A production by the adaptive immune system is driven by IL-1β and IL-6 in addition to IL-23 [21–23] . The impact of these cytokines on the regulation of IL-17A production by innate immune cells during OPC was not investigated in detail so far . Therefore , we set out to analyze mice with a nonfunctional IL-1 and/or IL-6 pathway . While genetic deletion of the IL-1 receptor or antibody-mediated neutralization of IL-6 alone had no measurable effect on IL-17A induction ( S3 Fig ) , the combination of both resulted in a drop in CD90+IL-17A+ cells in the tongue of infected mice in comparison to controls ( Fig 2C and 2D ) . Again , the effect was most pronounced for the TCRβ+ subset . Together , these data demonstrate that IL-23 , IL-1 and IL-6 play a critical role for rapid IL-17A induction at the onset of OPC , whereby IL-6 and IL-1 act in a redundant manner . Mononuclear phagocytes ( MNPs ) are a prominent source of IL-1β , IL-6 and/or IL-23 in diverse infectious settings and they may thus represent important players in the regulation of innate IL-17A production during acute OPC . However , the CD11c+MHCII+ MNP network in the tongue of mice remains poorly characterized . We thus set out to dissect this cellular compartment in detail . This required adaptation of our protocol for mouse tongue preparation to assure that the tissue-resident cells were liberated from the dense epithelial network . Based on CD11b and Langerin ( CD207 ) expression , we identified four distinct subsets within the CD11c+MHCII+ population in the naïve tongue ( Fig 3A ) . Unexpectedly , these cells appeared to be clearly distinct from the established DC populations e . g . in the spleen that comprise CD11b-CD24+ conventional DCs group 1 ( cDC1s ) , CD11b+CD24- cDC2s [24] and Langerin+ DCs [25] ( Fig 3A ) . Further characterization of the CD11c+MHCII+ MNPs in the naïve tongue showed that all four MNP subsets expressed low levels of XCR1 and Ly6C and high levels of Sirpα , whereby the highest expression of Sirpα was found in the CD11bhi subsets ( Fig 3B ) . The double-negative subset was strongly positive for F4/80 , while the CD11bhiLangerin- subset was heterogenous for most markers analyzed . Both Langerin+ subsets displayed a similar phenotype with high expression of CD24 , EpCam and CD64 ( Fig 3B ) . Consistent with a previous study on Langerhans cells in the oral mucosa [26] , we found tongue Langerin+ cells to be radiosensitive ( S4A and S4B Fig ) . Moreover , immunofluorescent staining of tissue sections and epithelial sheets from infected animals revealed an intraepithelial localization of the Langerin+/MHCII+ cells in the tongue ( S4C Fig ) . As the phenotypical analysis did not allow categorizing the tongue CD11c+MHCII+ MNP populations into DCs and/or macrophages , we examined their expression of Zbtb46 , a transcription factor selectively expressed by cDCs but no other myeloid and lymphoid cells [27] . The analysis of Zbtb46GFP reporter mice in combination with EpCam expression as a surrogate for Langerin on the cell surface demonstrated that EpCam+ but not EpCam- CD11c+MHCII+ cells were bona fide cDCs , whereas the EpCam- subsets rather represented macrophages in the naïve mouse tongue ( Fig 3C ) . We then assessed the dependence of the tongue CD11c+MHCII+ MNPs on the transcription factors Batf3 and Klf4 , which in other organs are lineage defining for cDC1s and a subset of cDC2s , respectivey [28 , 29] . Hematopoietic deletion of Klf4 had no impact on the tongue CD11c+MHCII+ MNPs ( Fig 3D and 3E ) . In contrast , Batf3 deficiency resulted in a specific loss of the CD11blowLangerin- subset in the tongue , which was surprising since those cells were Zbtb46-negative ( Fig 3C ) . For comparison , spleen samples were analyzed in parallel confirming that CD11b-CD103+ cDC1s were clearly Batf3-dependent whereas CD11b+CD103- cDCs were partially Klf4-dependent ( Fig 3D and 3E ) . Together , our data revealed that the CD11c+MHCII+ compartment in the naïve tongue is unique and heterogenous with its Zbtb46+CD11b+/lowLangerin+ DCs and Zbtb46-CD11b+/lowLangerin- macrophages that display an unprecedented onthological signature . Finally , we investigated the dynamics of these newly defined populations of tongue MNPs during OPC . The presence of C . albicans led to a rapid relocalization and clustering of intraepithelial MNPs in proximity of fungal hyphae ( S5 Fig ) . The population of Langerin+ DCs was reduced in size on day 1 post-infection when compared to the naïve state ( Fig 4A and 4B ) . The lack of Annexin-V+ staining suggested that the Langerin+ DCs were rather emigrating from the tongue epithelium than undergoing apoptosis ( data not shown ) . The loss of Langerin+ cells was accompanied by an increase in Langerin-CD11c+MHCII+ MNPs in the tongue , which co-expressed Ly6C and CCR2 , indicating that they were derived from Ly6Chigh inflammatory monocytes ( Fig 4C ) . In conclusion , these data show that the tongue bears a complex network of MNPs with important tissue-specific pecularities and that acute OPC leads to dynamic changes within this network including the decline in Langerin+ DCs and the infiltration and differentiation of CD11b+Ly6Chigh inflammatory monocytes . Our observation that IL-1β and IL-6 are both important for induction of IL-17 during OPC prompted us to identify the cellular source ( s ) of these cytokines . We first assessed cytokine production by the two major myeloid cell populations in the infected tongue , namely CD11b+CD11c+MHCII+ MNPs ( P1 ) and CD11b+CD11c-MHCII- cells ( P2 ) , the latter comprise pre-dominantly neutrophils . Both populations contributed to the overall IL-1β production ( S6A Fig ) , while—consistent with previous data [30]—no IL-1β was produced by the non-hematopoietic compartment ( S6B Fig ) . Intracellular staining of pro-IL-1β served to identify IL-1β-producing cells . The specificity of the intracellular staining for pro-IL-1β was confirmed by applying the same staining panel to cells obtained from infected Il1ab-/- mice ( S6C Fig ) . In addition to IL-1β , the IL-1 receptor is also engaged by IL-1α , which is released by oral keratinocytes during OPC [30] . IL-6 on the other hand was produced by both , the hematopoietic and non-hematopoietic compartment in response to infection . Among the CD45+ cells , CD11b+CD11c+MHCII+ MNPs provided the main source of IL-6 with very little contribution of CD11b+CD11c-MHCII- neutrophils ( S6A , S6B and S6D Fig ) . Overall , our data defined the presence of multiple hematopoietic and non-hematopoietic cellular compartments providing IL-1β and IL-6 at the onset of OPC . Next , we aimed at investigating the contribution of individual MNP subset ( s ) to the overall production of IL-1β and IL-6 during acute OPC . Therefore , we combined our staining panel for the four MNP subsets ( based on Langerin and CD11b expression ) with intracellular cytokine staining for pro-IL-1β and IL-6 ( S7A Fig ) . This revealed that pro-IL-1β and IL-6 were produced by three out of the four MNP subsets , namely Langerin+CD11b+ and Langerin+CD11b- DCs as well as Langerin-CD11b+ MNPs ( predominantly macrophages ) , while CD11b-Langerin- MNPs did not produce either of the two cytokines ( Fig 5A , 5B , 5D and 5E ) . Of all MNP subsets , the Langerin+ subsets displayed the highest proportion of IL-1β+ and IL-6+ cells ( Fig 5C and 5F ) . Analyzing cytokine production by Ly6C+ inflammatory monocytes showed that these cells contributed only little to pro-IL-1β and not to IL-6 secretion ( S7B and S7C Fig ) . In summary , our data show that tissue-resident MNPs , especially Langerin+ DCs , provide the IL-17A-inducing factors IL-1β and IL-6 during the onset of OPC . Besides IL-1β and IL-6 , IL-23 also contributes critically to innate IL-17A induction during OPC ( Fig 2A and 2B ) . Determining IL-23 production at a cellular level remains difficult due to the lack of a functional antibody for detection of the specific cytokine subunit IL-23p19 by flow cytometry . To overcome this limitation , we FACS-sorted four major cell populations from infected tongues and quantified the expression of IL23a transcripts ( coding for IL-23p19 ) by these cells by RT qPCR . For technical reasons we used EpCam instead of Langerin in the flow cytometry panel in this experiment . Our approach led to the identification of EpCam+CD11c+MHCII+ MNPs ( corresponding to Langerin+ DCs ) and CD11b+Ly6C+CD11c-MHCII- cells ( predominantly neutrophils ) as the main sources of IL23p19 , while EpCam-CD11c+MHCII+ MNPs ( mostly macrophages ) and CD45-EpCam+ epithelial cells contributed only marginally ( Fig 6A and 6B ) . The specificity of the RT qPCR analysis for IL-23p19 was verified by analyzing IL-23p19 expression in the four cell populations sorted from infected Il23a-/- mice ( S8A Fig ) . The adequacy of using EpCam instead of Langerin was confirmed by analyzing Cd207 transcript expression ( coding for Langerin ) in the sorted cell populations ( Fig 6C ) . Overall our data demonstrate that EpCam+/Langerin+ DCs as well as neutrophils provide IL-23p19 at the onset of OPC . We also examined expression of Il1b and Il6 transcripts by the four sorted cell populations and could thereby confirm the results obtained by flow cytometry ( Fig 5 , S6 Fig ) , namely that IL-1β was expressed predominantly by EpCam+ MNPs , EpCam- MNPs and neutrophils , and that IL-6 was expressed by EpCam+ DCs and to a lesser degree by EpCam- macrophages ( S8B and S8C Fig ) . Together , tissue-resident Langerin+/EpCam+ DCs thus stand out by their efficient expression of all three cytokines involved in innate IL-17A induction . After identification of multiple and partially overlapping myeloid cell subsets producing IL-17A-inducing cytokines , we aimed at evaluating their functional relevance for IL-17A production during infection . First , we targeted neutrophils , which represent the majority of the tongue-infiltrating leukocytes during actue OPC ( S9A Fig ) [15 , 30] and contribute to the overall IL-1β and IL-23p19 production in the infected tongue ( Fig 6B , S6A Fig ) . We therefore treated mice with anti-Ly6G and anti-G-CSF antibodies to deplete neutrophils prior and during infection . This had only a limited impact on IL-17A production by the three CD90+ celluar subsets that did not reach statistical significance ( S9B Fig ) . However , the actual contribution of neutrophils to the IL-17 response may likely be underestimated due to the difficulty to fully deplete these rapidly infiltrating cells from the infected tissue despite the administration of two distinct neutralizing/blocking antibodies [15 , 30] . Moreover , additional cellular sources , such as EpCam+ DCs , provide IL-17A-inducing cytokines that may compensate for the compromised neutrophil response . We also assessed the contribution of inflammatory monocytes to innate IL-17A induction during OPC , as these cells have previously been shown to promote the adaptive Th17 response to C . albicans in the oral mucosa [19] . However , infected Ccr2-/- mice , in which CD11b+Ly6Chigh monocytes recruitment to the infected tongue was strongly impaired ( S9C and S9D Fig ) , displayed no defect in IL-17A production on day 1 post-infection by any of the three CD90+ cell subsets ( TCRβ+ , TCRγδ+ , ILCs ) compared to WT control mice ( S9E Fig ) . Albeit not fully conclusive due to the incomplete monocyte depletion in Ccr2-/- mice , these data are in line with our observation that Ly6C+ inflammatory monocytes contribute only weakly to the overall IL-1β production and do not produce IL-6 ( S7C and S7E Fig ) , indicating that in contrast to the later phase of OPC , inflammatory monocytes are dispensible for the early IL-17A response during acute OPC . Finally , we examined the role of the newly identified tongue MNP subsets in the initiation of innate IL-17A production during OPC . For this , we analyzed different genetic and antibody-depletion models with selective MNP defects . The lack of the CD11blowLangerin- MNP population in Batf3-/- mice ( Fig 3D and 3E ) had no impact on the number of IL-17A-producing CD90+ leukocytes during acute OPC ( data not shown ) , which is in line with a previous publication showing that Batf3-deficiency in mice is dispensable for IL-17-mediated antifungal defense during OPC [31] . The same was true for IL-17A production in VAVCreKLF4fl/fl animals ( data not shown ) , which was not surprising given that these mice did not display any changes in the tongue MNP network ( Fig 3D and 3E ) . Homeostasis of tissue-resident MNPs , including macrophages and Langerin+ DCs , depends on CSF1R signaling and consequently these cells can be depleted in vivo upon antibody-mediated blockade of the CSF1R [32 , 33] . We thus aimed at depleting tongue CD11c+MHCII+ MNPs by administering an anti-CSF1R blocking antibody prior to OPC onset . This resulted in the selective loss of Langerin+CD11c+MHCII+ DCs , while Langerin-CD11c+MHCII+ MNPs , Ly6C+CCR2+ monocytes or neutrophils were not significantly affected by the treatment ( Fig 7A and 7B ) . Importantly , the loss of Langerin+ DCs in the tongue was accompanied by a significant reduction in IL-17A-production by all three CD90+ subsets if compared to non-treated WT control mice ( Fig 7C and 7D ) . These results were in line with our findings of Langerin+ DCs being the only tissue-resident MNP population in the tongue producing all three IL-17A-inducing cytokines IL-1β , IL-6 and IL-23 ( Figs 5 and 6 ) and demonstrate that Langerin+ DCs are key players in the IL-17 response during acute OPC .
The cytokine IL-17 has gained much attention due to its association with auto-inflammatory disorders such as psoriasis , psoriatic arthritis or ankylosing spondylitis [34 , 35] and the remarkable success in treating these diseases with IL-17 targeting reagents [36 , 37] . However , IL-17 is also crucial for mediating immune homeostasis in barrier tissues that are continuously exposed to microbes . Over the past decade , the lower gastrointestinal tract has been intensively studied in this context [38–40] . IL-17 production in the gut is determined by the microbiota [41–43] and the Gram positive segmented filamentous bacterium SFB has been identified as a major driver of the response [44 , 45] . The regulation of IL-17 in other mucosal tissues than the gut is less well studied . Only recently , it was shown that the homeostatic Th17 response in the gingiva is independent of the microbiota but rather a consequence of constant tissue damage elicited by mastication [46] . IL-17 immunity plays an essential role in host defense against opportunistic infections with the fungal pathogen C . albicans as evidenced by primary immunodeficiency patients with defects in genes of the IL-17 pathway that suffer from chronic mucocutaneous candidiasis . One of the tissues most frequently affected by C . albicans is the oral cavity . Experimental OPC in mice triggers prominent IL-17 induction and fungal clearance depends on the rapid induction of IL-17 in the infected mucosa during the onset of infection [13] . Here , we demonstrated that upon OPC in mice , IL-17 is produced by a tripartite population of CD90+ leukocytes in the tongue , comprising αβ T cells , γδ T cells and ILCs . Production of innate IL-17 is under the control of CSF1-dependent Langerin+ DCs , which are the major source of the IL-17-inducing cytokines IL-1β , IL-6 and IL-23 in the oral mucosa . Earlier work from our group already proposed ILCs to provide IL-17A at the onset of OPC in experimentally infected mice [13] . This observation was based on the rapid kinetics of IL-17 induction and the fact that Rag1-/- but not Rag1-/-Il2rg-/- or anti-CD90-treated Rag1-/- mice were protected from infection due to their capacity of upregulating IL-17 in the infected mucosa . Furthermore , MHC-II-deficiency did not impair IL-17A expression in the infected mucosa indicating that conventional MHCII-mediated antigen presentation is dispensable for IL-17A production during OPC [13] . However , our findings were challenged by the work from Conti et al . , who reported IL-17 expression by RAG-dependent lymphocytes , foremost αβ and γδ T cells , during OPC [20] . These seemingly contradictory results have arisen not least because of indirect assessments of IL-17A production during infection by either measuring Il17a and Il17f transcripts in crude tongue extracts [13] or monitoring IL-17A promoter activity in Il17aeYFP fate reporter mice [20] . Here , we reconcile the discrepancy and demonstrate the existence of three separate and complementary IL-17-producing cell types by direct visualization of IL-17A and IL-17F cytokines in the infected tongue . These three cellular subsets act in an at least partially redundant manner: selective lack of αβ or γδ T cells does not affect fungal control and only deletion of all three subsets phenocopies the high susceptibility of IL-17RA or IL-17RC-deficient mice to OPC [11 , 13 , 15] , underlining the robustness of the IL-17 response to the fungus . Based on their independence of RAG and their expression of RORγt , the TCR-negative IL-17 producers are part of the family of group 3 ILCs , although they lack expression of CCR6 , NCR1 and MHCII , which are characteristic of at least some ILC3s [47 , 48] . Visualization of IL-17A and IL-17F protein expression by flow cytometry not only allowed us to define the sources of IL-17 but also offered the opportunity to investigate the regulatory mechanisms of IL-17 production during acute OPC . We confirmed the critical role of IL-23 for innate IL-17 induction , an observation that is in line with previous work demonstrating that Il23a-/- mice phenocopy Il17ra-/- and Il17rc-/- mice in their inability to clear C . albicans [11 , 13] . However , it also became evident that the defect in IL-17 production in response to OPC was not complete in Il23a-/- animals , indicating that IL-23 may have ( an ) additional IL-17-independent function ( s ) in antifungal defense . Moreover , IL-23 may share redundancy with other IL-17-inducing cytokines . The dependence on IL-23 was not equally pronounced for all IL-17-producing subsets , suggesting that the relative contribution of different cytokines to IL-17 induction may differ for different cellular sources . In addition , the technical limitations of the experimental system due to the small cell numbers recoverable from the tongue may also mask clearer associations . Both , IL-1 and IL-6 have been implicated in the regulation of IL-17 immunity in the gut [49 , 50] and IL-6 was also implicated in Th17 polarization in the gingiva during steady state [46] . Here , we report that these cytokines also trigger innate IL-17 production in response to C . albicans in the oral mucosa . The impact of IL-1 and IL-6 on innate IL-17 production was somewhat overlooked before when each pathway was examined in isolation [13 , 20] . We now demonstrate that only concurrent blockade of both IL-1 and IL-6 pathways resulted in a significant drop of IL-17 induction in infected mice . Tongue-resident MNP populations have not been characterized in detail in mice . Our phenotypic and transcription factor analysis of CD11c+MHCII+ MNPs revealed the presence of two heterogeneous populations of Zbtb46-Langerin- macrophages and Zbtb46+Langerin+ DCs in the naïve tongue . Whether the Langerin+ DCs are bone fide Langerhans cells or represent the mucosal analogue of Langerin+ dermal DCs remains to be determined [51] . In terms of their phenotype and in situ localization they closely resemble Langerhans cells in the gingiva and the buccal mucosa [26] . Moreover , their independence of Batf3 further supports that they are indeed Langerhans cells [52] . Langerhans cells in the oral mucosa have been shown to differ from their skin counterparts in terms of their ontogeny , as they are derived from circulating radiosensitive precursors instead of radio-resistant embryonic precursors [26] and we confirmed this to be the case in the tongue . That MNPs in the tongue are different from MNPs in other tissues is also exemplified by the Langerin- MNP populations . We found the CD11blowLangerin- subset to depend on Batf3 , a lineage-determining transcription factor for cDC1s , but at the same time to lack Zbtb46 expression , thus calling into question whether it constitutes a Batf3-dependent subset of macrophages or a special population of Zbtb46-independet DCs . While future work will be needed to fully clarify the ontogeny of all four oral MNP subsets , our detailed dissection of the MNP network in the murine tongue sets the stage for interrogating the contribution of individual populations to immune homeostasis and defense . Among the accessory cells supplying IL-1 and IL-6 during acute OPC , we identified non-hematopoietic cells , which have been shown before to release IL-6 and IL-1α in the oral mucosa of mice [30] . In addition , we found ( several ) complementary myeloid cell populations serving as hematopoietic sources of IL-1 , IL-6 and/or IL-23 during infection . Neutrophils that rapidly infiltrate to the site of infection were found to produce IL-1β and IL-23 during OPC . While neutrophils can themselves serve as a source of IL-17A under certain circumstances [53 , 54] , we have no evidence for IL-17 production by neutrophils during OPC . Conversely , neutrophils may support IL-17 production by CD90+ leukocytes as they secrete IL-17-promoting cytokines during infection in the oral cavity . Macrophages and monocytes ( defined as Langerin- MNPs ) contributed to the production of IL-1β and IL-6 , but did not produce IL-23 . Langerin+ MNPs however were the only cellular subset producing all three IL-17-inducing factors . Together with their strategic location in the outermost layer of the epithelium at the onset of the infection , prior to the arrival of infiltrating inflammatory cells , this unique property predisposes Langerin+ DCs as the primary IL-17-inducing cellular subset . Targeting MNPs via antibody-mediated blockade of CSF1R confirmed the crucial role of Langerin+ cells as the primary cellular determinant for IL-17 induction at the onset of OPC . Langerin+ DCs have been implicated in IL-17-mediated immunity against C . albicans previously: during experimental cutaneous candidiasis , the constitutive absence of Langerhans cells in huLangerin-DTA mice [55] resulted in a drastric reduction in Th17 differentiation in skin-draining lymph nodes [56] . Reminiscent of experimental OPC , infection of mice with C . albicans via the epicutaneous route also triggered an immediate local IL-17 response within one day of infection , which is dominated by γδ T cells [12] . The release of IL-17 by γδ T cells in the skin during epicutaneous infection was not dependent on Langerhans cells , but rather on CD301b+ dermal DCs [12] . In the oral mucosa however and in contrast to the skin , adaptive immunity against C . albicans does not rely on Langerin+ cells , but instead depends on CCR2-dependent inflammatory DCs and other Flt3-dependent migratory DCs [19] . Therefore , the contribution of specific MNP subsets to the regulation of IL-17 production in different epithelial tissues and in different phases during infection emphasizes the dynamic and tissue-specific regulation of IL-17 immunity to C . albicans . Here , we revealed a novel role of Langerin+ DCs in the tongue coordinating the acute IL-17 response during OPC .
All mouse experiments in this study were conducted in strict accordance with the guidelines of the Swiss Animals Protection Law and were performed under the protocols approved by the Veterinary office of the Canton Zurich , Switzerland ( license number 201/2012 and 183/2015 ) . All efforts were made to minimize suffering and ensure the highest ethical and humane standards . WT C57BL/6j mice were purchased by Janvier Elevage . Il1r-/- [57] , Rag1-/- [58 , 59] , Ccr2-/- [60] , RorcCre [61] , Il23p19-/- [62] , Il17af-/- [63] ( a kind gift from Immo Prinz , MH Hannover , Germany ) , RorcCreR26ReYFP x Rag1-/- ( a kind gift from Burkhard Becher , University of Zurich , Switzerland ) and Rosa26reYFP animals [64] were bred at the Institute of Laboratory Animals Science ( University of Zurich , Zurich , Switzerland ) . Klf4fl/fl , VavCreKlf4flfl [65] , Batf3-/- [28] and Zbtb46GFP/+ [27] animals were bred at the Department of Biomedicine , University of Basel , Switzerland . Il1ab-/- mice were obtained from Wolf-Dietrich Hardt , ETH Zurich , Switzerland . All mice were on the C57BL/6 background except for Batf3-/-animals , which were on mixed Sv129/B6 background . The animals were kept in specific pathogen-free conditions and used at 6–12 weeks of age in sex- and age-matched groups . The C . albicans strain SC5314 [66] was used for all experiments if not stated otherwise . CAF-yCherry was obtained from Robert Wheeler [67] . Mice were infected with 2 . 5x106 cfu of C . albicans sublingually as described [68] without immunosuppression . Mice were monitored for morbidity and euthanized in case they showed severe signs of pain or distress . All analyses of infected animals in this study were carried out at 24 hours post infection . Mice were anaesthetized with a sublethal dose of Ketamin ( 100mg/kg ) , Xylazin ( 20mg/kg ) and Acepromazin ( 2 . 9mg/kg ) and perfused by injection of PBS into the right heart ventricle prior to removing the tongue and/or the spleen . For most experiments except for the analysis of tongue-resident MNPs , we isolated leukocytes as previously described in detail [69] . Briefly , tongues were cut into fine pieces and digested with DNase I ( 200μg/ml ) and Collagenase IV ( 4 . 8 mg/ml , Invitrogen ) in PBS for 50 minutes at 37°C . Single cell suspensions were obtained by passing the digested tissue through a 70μm strainer using ice-cold PBS supplemented with 1% FCS and 2mM EDTA . Tongue leukocytes were enriched over a 40% Percoll gradient before they were stained for flow cytometry . For the characterization of tongue-resident MNPs , tongues were cut in half and the underlying muscle tissue was carefully removed using a scalpel . The remaining tongue tissue was cut into fine pieces and digested with Trypsin ( 1mg/ml ) , DNase I ( 200mg/ml ) and Collagenase IV ( 2 . 4mg/ml ) in PBS for 45 minutes at 37°C . Single cell suspensions were obtained by passing the digested tissue through a 70μm strainer using ice-cold PBS supplemented with 1% FCS and 2mM EDTA and then stained for flow cytometry . For tongue sections , the tissue was embedded in Tissue-Tek OCT compound ( Sakura ) and snap-frozen in liquid nitrogen . Sagittal cryo-sections were cut at a thickness of 9 μm with a HM525 Microtome Cryostat and were mounted to super frost glass slides ( Thermo Scientific ) . The specimen were allowed to dry at room temperature for 30min prior to immunofluorescence staining . For epithelial sheets , the tongue was cut longitudinally and the muscle tissue was carefully removed with a scalpel . The tissue was placed with the epithelial layer upwards onto a Dispase II solution ( 2 . 85 mg/ml PBS , Roche ) and incubated for 1 hour at 37°C . Epithelial sheets were obtained by separating the lamina propria from the epithelium using two watchman tweezers . For immunofluorescence staining the specimen were fixed either with methanol at 20°C for 20 minutes or with acetone a room temperature for 10 minutes depending on the antibody used for the staining . The following antibodies were used: anti-Langerin ( clone 929F . 3 , hybridoma supernatant ) , anti-MHCII ( clone M5/114 . 15 . 2 , Biolegend ) and anti-CD11c ( clone HL3 , BD Bioscience ) . The stained specimens were mounted with Mowiol and stored at 4°C . Images were acquired with a digital slide scanner ( NanoZoomer 2 . 0-HT , Hamamatsu ) and analyzed with NDP . view2 . To block cytokine secretion , infected mice were treated with Brefeldin A ( Axon Lab AG , 250μg per mouse i . p . ) three hours prior to euthanization . For IL-6 neutralization , animals were injected with anti-IL-6 ( clone MP5-20F3 , BioXCell , 60μg per mouse i . p . ) directly after infection and again eight hours later . For neutrophil depletion , mice were treated with anti-Ly6G ( clone 1A8 , BioXCell , 150μg per mouse i . p . ) on day -1 and with anti-G-CSF ( clone 67604 , R&D Systems , 10μg per mouse i . p . ) on day -1 and day 1 post-infection . Anti-CSF1R ( clone AFS98 , BioXCell or produced and in-house and obtained from M . Greter ) was injected on day -3 ( 2mg per mouse i . p . ) , on day -1 ( 0 . 5mg per mouse i . p . ) and day 0 ( 0 . 5mg per mouse i . p . ) of infection . All antibodies were from BioLegend , if not stated otherwise . For Flow cytometric analysis , single cell suspensions of the tongue and the spleen were stained in PBS supplemented with 1% FSC , 5mM EDTA and 0 . 02% NaN3 . LIVE/DEAD Near IR stain ( Life Technologies ) was used for exclusion of dead cells . The following antibodies were used for surface markers: anti-CD90 . 2 ( 30-H12 ) , anti-CD45 . 2 ( 104 ) , anti-TCRβ ( H57-597 ) , anti-TCRγδ ( GL3 ) , anti-CD11b ( eBioscience , M1/70 ) , anti-CD11c ( N418 ) , anti-MHCII ( M5/114 . 15 . 2 ) , anti-CD3 ( 145-2C11 ) , anti-CCR6 ( 29-2L17 ) , anti-CD24 ( M1/69 ) , anti-NCR1 ( 29A14 ) , anti-CD122 ( TM-β1 ) , anti-CD44 ( IM7 ) , anti-CD69 ( H1 . 2F3 ) , anti-ki67 ( 16A8 ) , anti-Ly6C ( HK1 . 4 ) , anti-CCR2 ( SA203G11 ) , anti-CD64 ( X54-5/7 . 1 ) , anti-Langerin ( 929F3 ) , anti-EpCam ( G8 . 8 ) , anti-Ly6G ( 1A8 ) , anti-XCR1 ( ZET ) , anti-Sirpα ( P84 ) , anti-CD103 ( 2E7 ) . For intracellular cytokine staining , tongue cells were fixed and permeabilized using BD Cytofix/Cytoperm reagent ( BD Bioscience ) and subsequently incubated in Perm/Wash buffer ( BD Bioscience ) containing the following cytokine-directed antibodies or the respective isotype controls: anti-IL-17A ( TC11-18H10 . 1 ) , anti-IL-17F ( 8F5 . 1A9 ) , anti-pro-IL-1β ( NJTEN3 ) and anti-IL6 ( MP5-20F3 ) . CD1d surface expression was stained with anti-CD1d tetramers ( Proimmune ) . All extracellular and intracellular staining steps were carried out on ice . Cells were acquired on a FACS LSR II Fortessa ( BD Biosciences ) or on a FACS Gallios ( Beckman Coulter ) and the data were analyzed with FlowJo software ( Tristar ) . In all the experiments , the cells were pre-gated on viable and single cells for analysis . Absolute cell numbers of CD90+ and CD90+IL-17A+ cells and their respective subpopulations were calculated based on a defined number of counting beads ( BD Bioscence , Calibrite Beads ) , which were added to the samples before flow cytometric acquisition . For sorting cells from the infected tissue , single cell suspensions of the tongue were stained in PBS , supplemented with 1% FSC and 5mM EDTA , using the same antibodies as described in the previous section . Using a FACS Aria III , 50–100 target cells per defined population were sorted per well of a 96-well plate ( Eppendorf ) containing RLT Plus RNeasy® lysis buffer ( Qiagen ) . Lysates were snap-frozen and stored at -80°C until further processing . Whole-transcriptome amplification was performed following the Smart-Seq2 protocol [70] . Briefly , Agencourt RNAClean XP paramagnetic beads ( Beckman Coulter ) in combination with a DynaMag-96 side skirted magnet ( Thermo Fisher ) were applied to purify whole-genome RNA . Subsequently cDNA was generated using the SuperScript II Reverse Transcriptase Kit ( Thermo Fisher ) , and further amplified with HiFi HotStart PCR Mix ( KAPA Biosystems ) . For DNA clean-up , Agencourt AMPure XP beads ( Beckman Coulter ) were used as above . Optimal DNA concentration for real-time qPCR assays was determined by testing sample serial dilution for the expression of the control gene Actb . RT qPCR was performed using SYBR Green ( Roche ) and a QuantStudio 7 Flex ( Life Technology ) instrument . The primers were Actb fwd 5´-CCCTGAAGTACCCCATTGAAC-3´ , Actb rev 5´-CTTTTCACGGTTGGCCTTAG-3´; Il1b fwd 5´-TACAGGCTCCGAGATGAACA-3´ , Il1b rev 5´-AGGCCACAGGTATTTTGTCG-3´; Il6 fwd 5´-GAGGATACCACTCCCAACAGACC-3´ , Il6 rev 5´-AAGTGCATCATCGTTGTTCATACA-3´ , Il23a fwd 5´- CCAGCAGCTCTCTCGGAATC-3´ , Il23a rev 5´-TCATATGTCCCGCTGGTGC-3´; Cd207 fwd 5´-ATGTTGAAAGGTCGTGTGGAC-3´ , Cd207 rev 5´- GTGGTGTTCACTATCTGCATCT-3´; All qRT-PCR assays were performed in duplicates and the relative expression ( rel . expr . ) of each gene was determined after normalization to Actb transcript levels . Cell numbers of CD90+ and CD90+IL-17A+ cells and their respective subpopulations were transformed using the formula Y = y ( Log10+1 ) to plot absolute cell numbers including 0 values . Bar of each data set indicate arithmetic mean . Statistical significance was determined by unpaired Student’s-test with Holm-Sidak correction for multiple comparison , one-way or two-way ANOVA with Tukey’s multiple comparison test using GraphPad Prism software with *p< 0 . 05; **p<0 . 01; ***p<0 . 001; ****p<0 . 0001 . | IL-17 is a key cytokine for immune homeostasis and host defense in barrier tissues , which can also drive inflammatory diseases and immunopathology under certain conditions . Most studies addressing IL-17-mediated processes focus on the lower gastrointestinal tract , while other barrier tissues such as the oral mucosa remain largely understudied , despite their important role for entry of hazardous microbes . The protective role of IL-17 is particularly relevant for host defense against the fungal pathogen Candida albicans , as evidenced by individuals with genetic defects in the IL-17 pathway . Experiments with mice demonstrated that rapid IL-17 expression is essential for fungal control during oropharyngeal candidiasis . How this is regulated remained largely unclear . In this study , we identified a tripartite population of innate lymphocytic cells as the bona fide cellular source of IL-17 in the oral cavity . At the molecular level , we found IL-6 and IL-1 to act synergistically and complementary to IL-23 for instruction of IL-17 in response to infection . In search for the cellular source ( s ) of these IL-17-inducing factors we identified Langerin+ DCs as key players in the coordination of C . albicans-induced innate IL-17 production . Together , this greatly advances our understanding of IL-17 regulatory mechanisms in antifungal immunity in the oral mucosa . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"flow",
"cytometry",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"immune",
"cells",
"pathogens",
"immunology",
"microbiology",
"opportunistic",
"infections",
"organisms",
"developmental",
"biology",
"fungi",
"tongue",
"experimental",
"organism",
"systems",
"molecular",
"development",
"fungal",
"pathogens",
"neutrophils",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"mycology",
"white",
"blood",
"cells",
"infectious",
"diseases",
"animal",
"cells",
"medical",
"microbiology",
"microbial",
"pathogens",
"spectrophotometry",
"immune",
"system",
"yeast",
"cytophotometry",
"candida",
"cell",
"staining",
"eukaryota",
"mouth",
"anatomy",
"cell",
"biology",
"monocytes",
"physiology",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"cellular",
"types",
"spectrum",
"analysis",
"techniques",
"candida",
"albicans"
] | 2018 | Langerin+ DCs regulate innate IL-17 production in the oral mucosa during Candida albicans-mediated infection |
HTLV-I-specific CD8+ T cells have been characterized with high frequencies in peripheral blood and cerebrospinal fluid and production of proinflammatory cytokines , which contribute to central nervous system inflammation in HTLV-I-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) . However , little is known about the differences in CD8+ T cell activation status between asymptomatic carrier ( ACs ) and patients with HAM/TSP . The expression of CD244 , a signaling lymphocyte activation molecule ( SLAM ) family receptor , was significantly higher on CD8+ T cells in HTLV-I-infected patients , both ACs and patients with HAM/TSP , than those on healthy normal donors ( NDs ) . Blockade of CD244 inhibited degranulation and IFN-γ production in CD8+ T cells of patients with HAM/TSP , suggesting that CD244 is associated with effector functions of CD8+ T cells in patients with HAM/TSP . Moreover , SLAM-associated protein ( SAP ) was overexpressed in patients with HAM/TSP compared to ACs and NDs . SAP expression in Tax-specific CTLs was correlated in the HTLV-I proviral DNA loads and the frequency of the cells in HTLV-I-infected patients . SAP knockdown by siRNA also inhibited IFN-γ production in CD8+ T cells of patients with HAM/TSP . Thus , the CD244/SAP pathway was involved in the active regulation of CD8+ T cells of patients with HAM/TSP , and may play roles in promoting inflammatory neurological disease .
HTLV-I infects 20 million people worldwide [1] . While the majority of infected individuals are asymptomatic carriers ( ACs ) of the virus , 5–10% of infected people develop either adult T cell leukemia/lymphoma ( ATL ) [2] or a chronic , progressive neurological disease termed HTLV-I-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) [3] , [4] . HAM/TSP is characterized by infiltration of perivascular inflammatory cells in the spinal cord including HTLV-I-specific CD8+ T cells CTLs [5] , [6] . High frequencies of these effector cells have been demonstrated in peripheral blood with even higher frequencies in cerebrospinal fluid ( CSF ) of patients with HAM/TSP . HTLV-I-specific CTLs produce various factors including IFN-γ and TNF-α that may suppress viral replication and kill infected cells or promote bystander activation and killing of nearby resident glial cells [7]–[15] . These studies suggested that HTLV-specific CTLs might be immunopathogenic in the inflammatory lesions of patients with HAM/TSP . Despite HTLV-I-specific CTL responses , HTLV-I proviral loads are significantly elevated in HAM/TSP patients compared to AC [16] . Increased expression particularly of the trans-activating viral gene encoding HTLV-I Tax has been suggested to play a role in HTLV-I disease progression [10] , [17] . HTLV-I Tax induces the expression of a various cellular genes , including IL-2 [18] , the α-chain of the IL-2 receptor ( IL-2Rα ) [19] , IL-15 [20] , and IL-15Rα [21] . Increased expressions of these critical immune mediators directly contributes to CD8+ T cell activation and the ex vivo T cell proliferation observed in patients with HAM/TSP [22] . Although HTLV-I-specific CTL responses have been demonstrated in ACs and patients with HAM/TSP [23] , [24] , high expression of IFN-γ in CD8+ T cells specifically in HAM/TSP patients compared to ACs have been reported to be induced by interaction with virus-infected CD4+ T cells and CD8+ T cells [8] , [9] , [25] . Recently , CD8+ T cells in patients with HAM/TSP , but not in ACs , were demonstrated to spontaneously degranulate and produce IFN-γ . Importantly , this CTL degranulation was shown to be mediated by HTLV-I infection of mononuclear phagocytes ( MPs ) with the concomitant expression of IL-15 [24] . Thus , the activation of HTLV-I-specific CTLs in HAM/TSP is associated with both virus and cytokines , although the relative contribution of each of these factors to the observed dysregulation of chronically activated virus-specific CD8+ T cells in patients with HAM/TSP remains to be determined . Effector functions of CD8+ T cells are known to be regulated by various cellular receptors and their downstream molecules . The signaling lymphocyte activation molecule ( SLAM ) family of receptors and their associated adaptors play a pivotal role in the control of both innate and adaptive immunity [26] . Recent evidences indicate that the family of receptors and their signaling cascades are also involved in various inflammatory diseases , such as rheumatoid arthritis and inflammatory bowl disease [27]–[29] . SLAM family receptors consist of six immunoglobulin-like molecules named CD150 ( SLAM ) , CD244 ( 2B4 ) , CD229 ( Ly-9 ) , NK-T-B antigen ( NTB-A; Ly-108 ) , CD84 and CD2-like receptor activating cytotoxic T cells ( CRACC ) [26] , [30] . These receptors are differentially expressed on various immune cell types . Most of these receptors recognize self-ligands , but only CD244 is implicated in heterotypic interactions with CD48 . CD244 is present on natural killer ( NK ) cells , γδ T cells , activated CD8+ T cells , monocytes and basophils [31]–[37] . Importantly , the expression of CD244 on CD8+ T cells was correlated with T cell activation , CTL differentiation and exhaustion [31] , [34] , [35] , [38] , [39] . High CD244 expression on CD8+ T cells has been shown in patients with HIV-1 infection [40] , acute infectious mononucleosis [41] and myelodysplastic syndrome [42] . Its ligand , CD48 , is a glyco-phosphatidylinositol ( GPI ) -linked receptor broadly expressed on immune cells . CD244/CD48 interactions between antigen-specific CD8+ T cells and their targets or among CD8+ T cells themselves can augment cytotoxicity of their specific targets , IFN-γ production , and proliferation [43] , [44] . These immune responses are regulated by downstream signals , including the adaptor molecule , SLAM-associated protein ( SAP ) . SAP is a small SH2-domain containing protein , expressed in T cells , NK cells , NKT cells , and some B cells [45]–[47] . Following CD244/CD48 interaction , SAP binds with high affinity to the cytoplasmic tail of CD244 and regulates the signal transduction by recruiting SRC kinases [48] . Patients with X-linked lymphoproliferative syndrome ( XLP ) with SAP deficiency are characterized by a decrease in cytotoxic function of NK cells and EBV-specific CD8+ T cells [49] , suggesting that SAP also has critical role in cytotoxic function of these cells . Since it has been demonstrated that activated HTLV-I specific CTL may play an important role in the pathogenesis of HAM/TSP and that CD244 is a crucial regulator of CTL function , we characterized the expression of CD244 on CD8+ T cells of HTLV-I-infected patients and examined their functions in patients with HAM/TSP . Here for the first time we demonstrate that CD244 was overexpressed on CD8+ T cells of HTLV-I-infected patients compared to healthy normal donors ( NDs ) , and that the upregulation of the adaptor protein , SAP , in CD8+ T cells distinguished patients with HAM/TSP from ACs . Moreover , blocking of CD244 and knockdown of SAP by siRNA resulted in inhibition of CD8+ T cell function , degranulation and IFN-γ expression , of patients with HAM/TSP . These data suggest that CD244/SAP pathway is involved in active regulation of CTLs in patients with HAM/TSP .
To determine whether CD244 expression correlates with activation of CD8+ T cells in HTLV-I-infected patients , CD244 expression on CD8+ T cells was examined by flow cytometry in NDs , ACs and patients with HAM/TSP . A representative histogram demonstrates that CD244 expression was higher on CD8+ T cells of HTLV-I-infected patients , both in an AC and a patient with HAM/TSP , than those of a ND ( Figure 1 A ) . In NDs , group analysis demonstrated that 13–62% of CD8+ T cells expressed CD244 ( n = 14; Figure 1 B ) . This result was comparable to previous results that reported approximately 30–50% of CD8+ T cells were CD244 positive in healthy donors [33] . In contrast , CD8+ T cells of HTLV-I-infected patients , both ACs and patients with HAM/TSP had significantly higher levels of CD244 expression; 35–90% and 45–98% respectively ( Figure 1 B ) . There was no significant difference in CD244 expression between ACs and patients with HAM/TSP ( P>0 . 05 ) . Moreover , CD244 was expressed on Tax11-19-specific and CMV pp65-specific CD8+ T cells of HLA-A*0201 patient with HAM/TSP ( Figure 1 C ) . NK cells of all subjects expressed CD244 ( >95% ) , and did not show any differences in CD244 expression between NDs and HTLV-I-infected patients ( data not shown ) . These results demonstrated that CD8+ T cells of HTLV-I-infected patients , including antigen-specific CTLs , showed significantly high CD244 expression , compared to NDs . As interactions of CD244 and its ligand , CD48 , have been reported to augment cytotoxicity , IFN-γ production and proliferation of CD8+ T cells in mouse [43] , [44] , high expression of CD244 on CD8+ T cells may also augment functional human CD8+ T cell responses . It has been established that in ex vivo cultures of HAM/TSP PBMCs , CD8+ T cells are in close contact with HTLV-I-infected cells and rapidly function to kill these infected cells by secretion of lytic granules and cytokines [9] , [12] , [24] , [25] . Therefore , to confirm the involvement of CD244 in this cytolytic process , we assessed CD8+ T cells of patients with HAM/TSP for their cytotoxic activity as defined by degranulation ( CD107 expression ) and IFN-γ expression by blocking CD244 or its ligand , CD48 . Figure 2 A shows a representative dot plot of CD107a and IFN-γ expressions in CD8+ T cells of ND and patients with HAM/TSP after culture of whole PBMCs for 24 hours . As previously reported [24] , CD8+ T cells of patients with HAM/TSP expressed both CD107a and IFN-γ after culture for 24 hours , whereas CD8+ T cells of ND did not ( Figure 2 A ) . When anti-CD244 or anti-CD48 was titrated and cultured in PBMCs of a patient with HAM/TSP , both antibodies inhibited CD107a and IFN-γ expression in CD8+ T cells of a patient with HAM/TSP in a dose dependent manner ( Figure 2 B ) . Figure 2 C shows the inhibitory effects of anti-CD244 and anti-CD48 ( 1µg/ml ) on degranulation and IFN-γ expression of CD8+ T cells in patients with HAM/TSP ( n = 7 ) . Anti-CD244 significantly inhibited ( 32 . 5% ) degranulation and IFN-γ expression in CD8+ T cells of patients with HAM/TSP compared with a control isotype IgG . Anti-CD48 had an even more pronounced inhibitory effect ( 65 . 5% ) . An additional established measure of HAM/TSP T cell activation ex vivo is the well-described observations of increased spontaneous T cell lymphoproliferation [50] . However , blockade of CD244 did not show any inhibitory effects on spontaneous lymphocyte proliferation of patients with HAM/TSP ( data not shown ) . These results demonstrate that CD244/CD48 interactions might be specifically involved in cytotoxic CD8+ T cells dysregulation , especially degranulation and IFN-γ expression , of patients with HAM/TSP . Although CD48 is broadly expressed on hematopoietic cells including lymphocytes and monocytes , viral infection such as HIV or EBV has been shown to decrease or increase CD48 expression on the infected cell , respectively [51] , [52] . Therefore , we examined CD48 expression on CD4+ T cells and CD14+ cells , which are known to be in vivo reservoirs for HTLV-I in patients with HAM/TSP and gradually express HTLV-I Tax protein after short term in vitro culture [24] , [53] . Figure 2 D shows a representative result of CD48 and HTLV-I Tax expression in CD4+ T cells and CD14+ cells of a patient with HAM/TSP . Both CD4+ T cells and CD14+ cells expressed CD48 before culture ( Figure 2 B ) . CD48 expression on CD4+ T cells did not change over time despite the expression of HTLV-I Tax ( Figure 2 B ) . This result supports previous studies that in vitro infection of HTLV-I in CD4+ T cells did not alter CD48 expression [54] . However , CD48 expression was partially downregulated in CD14+ cells after culture , but did not appear to be directly mediated by HTLV-I Tax expression , because Tax was not detected on the cells expressing low level of CD48 ( Figure 2 B ) . Collectively , these results demonstrate that HTLV-I expression does not directly mediate CD48 expression on infected cells , suggesting that CD244-expressing HTLV-I-specific CTLs have the potential to interact with virus-infected CD48+ target cells . It is well known that interactions between effector CD8+ T cells and their targets establish a distinct immunological synapse organized by the accumulation of various immunoreceptors to cell-cell junctions associated with the polarization of lytic granules , such as perforin , leading to induction of cell death [55] . To support further the involvement of CD244/CD48 signaling on CD8+ T cells in HTLV-I-infected patients , the distribution of CD244 was visualized on cytotoxic lymphocytes ( perforin+ cells ) of patients with HAM/TSP after 8 hours in vitro culture when both perforin+ cells and polarizing perforin+ cells were most frequently visualized . As shown in Figure 3 , immunofluorescence analysis demonstrated colocalization of CD244 on perforin+ cells ( Figure 3 A and B ) . When these perforin+ cells were in contact with their targets , either strongly positive ( Figure 3 A ) or weakly positive ( Figure 3 B ) for CD48 , polarization of perforin with the accumulation of CD244 was visualized at the cell-cell contact area ( Figure 3 A and B , middle ) . The accumulation of CD244 on polarizing perforin+ cells was observed in more than 30% of all the polarizing perforin+ cells in contact with another cell . These results demonstrate that expression and subcellular distribution of CD244 and perforin on cytotoxic lymphocytes from HAM/TSP patients was translocated to the contact area ( immunological synapse ) with CD48+ target cells , suggesting that CD244 on cytotoxic lymphocytes is involved in recognition of target cells or activation of cytotoxic lymphocytes in HTLV-I-infected patients . Given the involvement of CD244/CD48 interaction on CD8+ T cells of patients with HAM/TSP , we asked how CD244/CD48 signaling might regulate CD8+ T cell function since high expression of CD244 was demonstrated on CD8+ T cells of both ACs and patients with HAM/TSP . It is known that NK cells constitutively express CD244 [33] . However , the regulation of these cells , either in a resting or activated state , is related to the expression of the SLAM-related adaptor proteins such as SAP and EAT-2 , which play a role in controlling the active and inhibitory signal transduction in NK cells , respectively [32] , [56]–[58] . To determine whether SLAM-related adaptor proteins were associated with active regulation of CD8+ T cells , the expression of the adaptor proteins , SAP and EAT-2 , were compared in CD8+ T cells of NDs , ACs and patients with HAM/TSP . Representative results of SAP expression in CD8+ T cells are shown in Figure 4 . SAP expression was higher in CD8+ T cells of a patient with HAM/TSP compared to that of ND and AC ( Figure 4 A ) . Group analysis demonstrates that compared with the expression of SAP on CD8+ T cells of NDs and ACs , SAP expression was significantly increased in CD8+ T cells of patients with HAM/TSP ( Figure 4 B ) . Although CD8+ T cells of ACs expressed SAP slightly higher than those of NDs , statistical analysis did not show any significant differences between these groups . The expression of SAP in NK cells were 43 . 9% , 44 . 6% , and 58 . 0% on average in NDs , ACs , and patients with HAM/TSP , respectively , and was not significantly different among groups . Likewise , EAT-2 was highly expressed in CD8+ T cells of both NDs and HTLV-I-infected individuals with no significant differences observed ( Figure 4 C ) . These results demonstrate that CD8+ T cells of patients with HAM/TSP overexpress SAP compared to NDs and ACs , although EAT-2 expression was comparable in all three groups . To assess SAP expression in antigen-specific CTL , we examined HTLV-I Tax11-19 tetramer+ CD8+ T cells in HLA-A*0201+ AC and patient with HAM/TSP . As shown in a representative histogram , SAP expression was higher in Tax tetramer+ CD8+ T cells of patient with HAM/TSP than the AC ( Figure 4 D ) . To address SAP expression in these antigen-specific CD8+ T cells , we analyzed the HTLV-I proviral DNA loads in PBMCs and the frequency of Tax11-19 tetramer+ CD8+ T cells in HTLV-I-infected individuals as a function of SAP expression . As previously reported [9] , [12] , patients with HAM/TSP had higher frequencies of Tax11-19 tetramer+ CD8+ T cells , compared to ACs that correlated with HTLV-I proviral DNA loads . Figure 4E demonstrates that the amount of SAP expression in Tax11-19 tetramer+ CD8+ T cells was significantly correlated with HTLV-I proviral DNA loads in PBMCs ( P = 0 . 0401 , R2 = 0 . 4746 ) and the frequency of Tax11-19 tetramer+ CD8+ T cells in HTLV-I-infected individuals ( P = 0 . 0062 , R2 = 0 . 6811 ) . These results suggest that expansion of HTLV-I-specific CD8+ T cells particularly in patients with HAM/TSP is associated with the expression of SAP and can distinguish patients with neurologic disease from HTLV-I infected asymptomatic carriers . HTLV-I Tax induces the expression of a various cytokine genes , including IL-2 and IL-15 , which has been shown to be associated with CD8+ T cell activation and proliferation in patients with HAM/TSP [22] . IL-15 plays an important role in the prolonged maintenance of memory CD8+ T cell responses [59] . To examine whether IL-2 and IL-15 can upregulate SAP and CD244 expression in CD8+ T cells , purified CD8+ T cells from ND PBMCs ( n = 2 ) were cultured with recombinant human IL-2 ( rhIL-2 ) or rhIL-15 for 7 days , and expression of SAP and CD244 in these cells were analyzed . Both rhIL-2 and rhIL-15 stimulation induced SAP expression in CD8+ T cells of NDs , depending on the concentration of rhIL-2 or rhIL-15 ( Figure 5 A ) . In addition , the magnitude of SAP expression in CD8+ T cells by IL-2 and IL-15 was variable among individuals . By contrast , CD244 expression on CD8+ T cells did not change after culture with rhIL-2 and rhIL-15 ( Figure 5 B ) . Thus , IL-2 or IL-15 stimulated CD8+ T cells showed high expression of SAP , suggesting that the observed upregulation of SAP in CD8+ T cells of patients with HAM/TSP may be a function of increased cytokine production . Upregulation of SAP may be related to dysregulation of chronically activated CD8+ T cells in patients with HAM/TSP . To confirm the involvement of SAP on dysregulation of CD8+ T cells in patients with HAM/TSP , we examined degranulation and IFN-γ expression in CD8+ T cells of patients with HAM/TSP after knockdown of SAP or EAT-2 by siRNA . SAP and EAT-2 expression in transfected CD8+ T cells was determined after culture for 6 hours . Gene knockdowns by siRNA were effective against their respective targets at levels of 50–60% of baseline ( Figure 6 A ) . After coculture with autologous CD14+ cells , CD8+ T cells transfected with SAP siRNA significantly decreased degranulation and IFN-γ expression ( 40 . 5% inhibition , P = 0 . 005 ) compared to CD8+ T cells transfected with control siRNA ( Figure 6 B ) while CD8+ T cells transfected with EAT-2 siRNA had no inhibitory effect . The inhibitory effect of SAP siRNA was similar to those of anti-CD244 on degranulation and IFN-γ expression in CD8+ T cells of patients with HAM/TSP ( Figure 2 C ) . These results demonstrate that decreased SAP expression resulted in inhibition of degranulation and IFN-γ expression in CD8+ T cells of patients with HAM/TSP , supporting the role for SAP in the activation of cytotoxic CD8+ T cell function in patients with HAM/TSP .
Activation and dysregulation of CD8+ T cells in HTLV-I-infected patients have been suggested to be associated with disease progression and pathogenesis of HAM/TSP [7]–[15] . In this study , we have characterized that a member of the SLAM family of receptors , CD244 , was overexpressed on CD8+ T cells of HTLV-I-infected patients compared to NDs , and demonstrated that the upregulation of the adaptor protein , SAP , in CD8+ T cells distinguished patients with HAM/TSP from ACs . The expression of CD244 on CD8+ T cells correlated with T cell activation [31] , [34] , [35] , [38] , as has been reported in patients with HIV-1 infection [40] , acute infectious mononucleosis [41] , and myelodysplastic syndrome [42] . CD244+ CD8+ T cells lack expression of CD45RA , CD62L , CD28 , and CCR7 and acquire expression of perforin , granzyme B , and IFN-γ [31] , [35] , [38] . In addition , 2B4+ cells ( also known as CD244 ) showed higher cytotoxicity than 2B4− T cells [31] , [34] . Since high frequency of CD45RA−CD27+ memory CD8+ T cells and high proliferation rate of CD8+ CD45RO+ T cells have been reported in patients with HAM/TSP [10] , [60] and in HTLV-I-infected patients ( both ACs and patients with HAM/TSP [61] ) , the expression of CD244 on CD8+ T cells in HTLV-I-infected patients is consistent with the interpretation that CD244 is a marker of memory CD8+ T cells and with persistent immune activation in HTLV-I-infected patients . High CD244 expression was demonstrated on HTLV-I Tax11-19-specific CD8+ T cells as well as CMV pp65-specific CD8+ T cells in a patient with HAM/TSP . Previous results have shown that CMV pp65-specific CD8+ T cells also showed high expression of CD244 whereas influenza virus-specific and melanoma antigen-specific CD8+ T cells showed low or negative expression of CD244 , respectively [31] . Furthermore , high expression of CD244 has been recently reported to be related to CD8+ T cell exhaustion during chronic LCMV infection in mouse [39] . In our study , there were no differences in CD244 expression on CD8+ T cells between ACs and patients with HAM/TSP ( Figure 1 B ) and expression of CD244 in HTLV-I-infected patients did not show any direct correlation with HTLV-I infection such as HTLV-I proviral load or expression of Tax ( data not shown ) . Thus , expression of CD244 on CD8+ T cells might indicate the activation state of CD8+ T cells or degree of CTL differentiation in chronic viral infection . As suggested by the analyses of the disorder XLP , which is characterized by SAP deficiency and a decrease in cytotoxic function of NK cells and EBV-specific CD8+ T cells , the apparent activating effect of CD244 in human NK cells and CD8+ T cells may relate to preferential expression of SAP [32] , [56] , [62]–[64] . Here we have demonstrated a disease-specific upregulation of the adaptor protein , SAP , in CD8+ T cells of patients with HAM/TSP but not in NDs or ACs . This observation was specific for SAP since EAT-2 was comparably expressed in both NDs , and HTLV-I-infected patients and carriers . Moreover , the expression of SAP was higher in both total CD8+ T cells and in Tax11-19-specific CD8+ T cells in patients with HAM/TSP compared to ACs . Interestingly , while expression of CD244 in HTLV-I-infected patients did not show any correlation with HTLV-I infection , the upregulation of SAP in Tax11-19-specific CD8+ T cells was significantly correlated with the HTLV-I proviral DNA load in PBMCs and the frequency of Tax11-19-specific CD8+ T cells in HTLV-I-infected patients , suggesting that the expression of SAP is associated with the increase of proviral DNA and the expansion of antigen-specific CD8+ T cells . This is consistent with previous reports that demonstrated the relationship of disease progression with proviral DNA loads and virus-specific CD8+ T cells [17] . Upregulation of SAP was reported in splenocytes of mice infected with LCMV and MCMV [56] , and in PBMCs of patients with infectious mononucleosis , even during early stages of the disease [41] . While SAP expression in NK cells is related to stimulation with NK cell activators , such as IL-2 , IL-12 , IFN-α and poly ( I∶C ) [56] , [58] , regulation of SAP expression in human CD8+ T cell is still unclear [64]–[66] . Since the expansion of HTLV-I-specific CD8+ T cells in HTLV-I infected patients is known to be influenced by various cytokines such as IL-2 and IL-15 as well as by virus antigen , we reasoned that these same stimuli would induce the expression of SAP that we have shown to be elevated in HTLV-I specific CD8+ T cells . In our study , stimulation of NDs CD8+ T cells with IL-2 and IL-15 induced SAP but not CD244 expression on CD8+ T cells . Indeed , while TCR activation was strongly correlated with expression of CD244 on CD8+ T cells , as previously reported [64] , stimulation of cytokines such as IL-2 and IL-15 induced upregulation of SAP in CD8+ T cells . Therefore , compared to CD8+ T cells in ACs with high expression of CD244 and low expression of SAP , high expression of CD244 and SAP in CD8+ T cells of patients with HAM/TSP may be regulated by these cytokine-dependent activation as well as TCR dependent activation . Results presented in this study also demonstrate that CD244/SAP pathway was functionally involved in cytotoxic activity of CD8+ T cells in patients with HAM/TSP . Immunofluorescence analysis visualized CD244+ perforin+ cytotoxic T cells in contact with CD48+ cells where clusters of CD244 and perforin were localized to the cell-cell contact area . These results demonstrated that CD244 on cytotoxic lymphocytes is involved in recognition of target cells or activation of cytotoxic lymphocytes . Recruitment of lytic molecules such as perforin to the cell-cell contact area suggests induction of target cell death . Moreover , analysis of CD8+ T cell function in patients with HAM/TSP as defined by degranulation and IFN-γ expression was inhibited by blockade of CD244 or knockdown of SAP , but not EAT-2 . However , since spontaneous degranulation and IFN-γ expression was not detected in CD8+ T cells of ACs [24] , we suggest that high expression of CD244 alone was not directly responsible for spontaneous degranulation and IFN-γ expression of CD8+ T cells . Rather , the overexpression of SAP in CD8+ T cells of patients with HAM/TSP , with high expression of CD244 , might contribute to the CTL activity observed in this disorder by overproduction of IFN-γ and other inflammatory mediators . These results confirmed that expression of both CD244 and SAP are associated with CD8+ T cell activity , as previously reported [64] . The CD244/SAP pathway has been shown to regulate effector functions of NK and T cells depending on the expression level of CD244 and SAP [67] . Our results extends the significance of the CD244/SAP pathway in activation and regulation of virus-specific CD8+ T cell responses and has important implications for T cell-mediated pathogenesis of inflammatory neurologic disorders associated with an imbalance of immune function .
Blood samples were obtained from thirteen patients with HAM/TSP ( HAM#1-13 ) , eight AC ( AC#1-8 ) and fourteen HTLV-I-seronegative healthy donors ( ND#1-14 ) . Diagnosis of HAM/TSP was based on World Health Organization ( WHO ) diagnostic criteria . Five patients with HAM/TSP and four AC were HLA typed as HLA-A*0201+ . PBMCs were isolated by Ficoll-Hypaque ( Lonza Walkersville , Walkersville , MD ) centrifugation , and were cryopreserved in liquid nitrogen until use . Informed consent was written and obtained from each subject in accordance with the Declaration of Helsinki . The study was reviewed and approved by the National Institute of Neurological Disorders and Stroke Institutional Review Board . For flow cytometory , antibodies for human CD3 , CD4 , CD8 , CD14 , CD48 , CD107a , CD244 , IFN-γ , perforin ( all from BD Biosciences , San Jose , CA ) , SAP ( Cell Signaling , Danvers , MA ) , EAT-2 ( Santa Cruz Biotechnology , Santa Cruz , CA ) , Tax 11–19/HLA-A0201 tetramer ( provided by National Institute of Allergy and Infectious Disease MHC Tetramer Core Facility , Atlanta , GA ) and CMV pp65/HLA-A0201 tetramer ( Beckman Coulter , San Diego , CA ) were used . Anti-Tax monoclonal antibody ( Lt-4 ) was kindly provided by Dr . Y . Tanaka ( University of the Ryukyus , Okinawa , Japan ) . For blocking experiments , both anti-CD244 and anti-CD48 were purchased from eBioscience ( San Diego , CA ) . For immunofluorescence microscopy , primary antibodies used were anti-CD244 ( R&D systems; Minneapolis , MN ) , anti-CD48 ( AbD Serotec; Oxford , UK ) and anti-perforin ( BD Biosciences ) ; and secondary antibodies used were Alexa 546 donkey anti-goat IgG ( H&L ) , Alexa 647 goat anti-mouse IgG1 , and Alexa 488 goat anti-mouse IgG2b ( all from Invitrogen; Carlsbad , CA ) , respectively . siRNAs specific for SAP mRNA , EAT-2 mRNA and control siRNA were purchased from Santa Cruz Biotechnology . Recombinant human ( rh ) IL-2 and rhIL-15 were purchased from Peprotech . ( Rocky Hill , NJ ) . CD107a mobilization assay was performed as previously described [24] . Briefly , PBMCs of ND or HTLV-I-infected patients were suspended in RPMI 1640 media supplemented with 10% FBS , 100 U/mL penicillin , 100 µg/mL streptomycin sulfate , and 2 mM L-glutamine , and cultured in 24 well plate in 5% CO2 incubator at 37°C for 24 hours . In blocking experiments , 0 , 0 . 1 , 1 or 10 µg/mL of each antibody ( control IgG , anti-CD244 and anti-CD48 ) was added . Conjugated CD107a antibody , GoldiStop™ ( BD Biosciences ) , and brefeldin A ( Sigma , St . Louis , MO ) were added into the culture for 5 hours before the time point for detection . Expressions of CD107a , IFN-γ , CD244 , Tax , SAP or EAT-2 in the cultured or uncultured PBMCs were examined by flow cytometoric analysis . PBMCs were surface-stained with specific antibodies . In the case of combination staining with tetrameric complexes , PBMCs were stained with either Tax11-19 or CMV pp65-specific tetramer before surface staining . After being fixed and permeabilized with Fixation/Permeabilization solution ( BD Biosciences ) , the cells were intracellular-stained with antibodies against IFN-γ , Tax , SAP or EAT-2 for each experiment . Flow cytometric analysis was performed using a FACSCalibur flow cytometer ( BD Biosciences ) . The data were analyzed using FlowJo software ( Tree Star , San Carlos , CA ) . To examine CD244 and SAP expression in CD8+ T cells stimulated with IL-2 or IL-15 , CD8+ T cells were magnetically isolated from ND PBMCs by negative selection using CD8+ T cell isolation kit II ( Miltenyi , Bergisch Gladbach , Germany ) and cultured with appropriate concentration of rhIL-2 or rhIL-15 for 7 days . Lab tech chamber glass slides ( Nalge Nunc International , Rochester , NY ) were coated with poly-L-lysine ( Sigma ) before use . PBMCs were plated into the chamber slides with RPMI media , and cultured for 8 hours . After washing with PBS including 1% FBS , the cells were stained with primary antibodies for CD244 and CD48 for 1 hour at room temperature . After fixation with 4% paraformardehide for 15 min and subsequent washing with PBS including 0 . 1% Triton-X , cells were stained with a primary antibody for perforin for 1 hour . After washing , each secondary antibody was applied , and DAPI was finally used for nuclear counterstaining . The stained cells were visualized with Zeiss 200M Axiovert inverted microscope equipped with mercury lamp house HBO-100 and four appropriate dichroic filters for DAPI and Alexa 488 , 546 and 647 ( Carl Zeiss MicroImaging Inc , Thornwood , NY ) . To evaluate the cell-cell contact and the distribution of perforin , CD244 and CD48 , polarizing perforin+ cell and its contacting cell was first identified under the microscope and then the image picture was created using four filters for DAPI , perforin , CD244 and CD48 . Three-dimensional ( 3-D ) reconstructions of each section were assembled using Volocity 3-D imaging analysis software ( Improvision , Waltham , MA ) . The image data was deconvoluted and modified into 3-D image , and the accumulation of CD244 on polarizing perforin+ cells in contact with the other cell was confirmed from 50 sets of the 3-D image data . DNA was extracted from total PBMCs of HLA-A*0201+ HTLV-I-infected patients using QIAamp DNA Blood Mini Kit ( QIAGEN , Valencia , CA ) and HTLV-1 proviral DNA load was measured using TaqMan system as previously described [68] . CD8+ T cells and CD14+ cells were magnetically isolated from the patient's PBMCs by CD8+ T cell isolation kit II and CD14 MicroBeads ( both from Miltenyi ) , respectively , according to the manufacturer's instructions . The purities of CD8+ T cells and CD14+ cells were confirmed as >90% cells and as approximately 95% , respectively . Purified CD8+ T cells ( 3×106 cells ) were transfected with 300 nM of control , SAP or EAT-2 siRNA using the Human T Cell Nucleofector Kit ( AMAXA , Cologne , Germany ) with the AMAXA Nucleofector II according to the manufacturer's instructions . Transfection efficacy for siRNA was determined 55–70% as determined with fluorescein conjugated control siRNA . Each transfected CD8+ T cells were rested for 6 hours and then cocultured with an equal number of CD14+ cells for 24 hours . After culture , CD107a and IFN-γ expression was analyzed in CD8+ T cells by flow cytometry . SAP knockdown by siRNA was determined by western blot . After resting for 6 hours , transfected CD8+ T cells were collected and stored at −80°C until use . The cells were lysed in 5 mM Tris–HCl , pH 8 . 0 , 1% Triton X-100 and a cocktail of protease inhibitors ( Sigma ) . Protein concentration was determined using Quick Start Bradford Protein Assay ( BioRad , Hercules , CA ) . From each protein sample , 10 µg was electrophoresed through a NuPAGE® 12% Bis–Tris gel ( Invitrogen ) . The gel was transferred to a nitrocellurose membrane ( Invitrogen ) . After blocking with 3% BSA in TBS , the membrane was probed with anti-SAP antibodies ( Cell Signaling ) and then probed with horseradish peroxidase-conjugated IgG ( Cell Signaling ) . The membrane was visualized by chemiluminescence using SuperSignal® West Pico Chemiluminescent substrate ( Thermo Scientific , Rockford , IL ) and analyzed the profile on Kodak digital science™ 1D image analysis software ( Kodak , Rochester , NY ) . Scatter plot and simple regression analysis were performed using Prism ( GraphPad software , La Jolla , CA ) . | Human T-lymphotropic virus type I ( HTLV-I ) is a retrovirus that persistently infects 20 million people worldwide . The majority of infected individuals are asymptomatic carriers of the virus , but 5–10% of infected people develop either adult T cell leukemia/lymphoma ( ATL ) or a chronic , progressive neurological disease termed HTLV-I-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) . HAM/TSP is characterized by central nervous system ( CNS ) inflammation including HTLV-I-specific CD8+ T cells where disease progression and pathogenesis is associated with a dysregulation of antigen-specific CD8+ T cells , although the mechanism of this dysregulation remains to be defined . Here we demonstrate that a signaling lymphocyte activation molecule ( SLAM ) family of receptors , CD244 , was overexpressed on CD8+ T cells of HTLV-I-infected patients than those of healthy normal donors , and that the upregulation of the adaptor protein , SAP , in CD8+ T cells distinguished HTLV-I infected individuals with and without neurologic disease . Both CD244 and SAP were associated with effector functions ( high expression of IFN-γ ) of CD8+ T cells in patients with HAM/TSP . This finding has important implication for T cell-mediated pathogenesis in human chronic viral infection associated with imbalance of immune function . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"immunology/immunity",
"to",
"infections",
"infectious",
"diseases/viral",
"infections",
"immunology/leukocyte",
"activation"
] | 2009 | High Expression of CD244 and SAP Regulated CD8+ T Cell Responses of Patients with HTLV-I Associated Neurologic Disease |
While most yeast enzymes for the biosynthesis of glycerophospholipids , sphingolipids and ergosterol are known , genes for several postulated transporters allowing the flopping of biosynthetic intermediates and newly made lipids from the cytosolic to the lumenal side of the membrane are still not identified . An E-MAP measuring the growth of 142'108 double mutants generated by systematically crossing 543 hypomorphic or deletion alleles in genes encoding multispan membrane proteins , both on media with or without an inhibitor of fatty acid synthesis , was generated . Flc proteins , represented by 4 homologous genes encoding presumed FAD or calcium transporters of the ER , have a severe depression of sphingolipid biosynthesis and elevated detergent sensitivity of the ER . FLC1 , FLC2 and FLC3 are redundant in granting a common function , which remains essential even when the severe cell wall defect of flc mutants is compensated by osmotic support . Biochemical characterization of some other genetic interactions shows that Cst26 is the enzyme mainly responsible for the introduction of saturated very long chain fatty acids into phosphatidylinositol and that the GPI lipid remodelase Cwh43 , responsible for introducing ceramides into GPI anchors having a C26:0 fatty acid in sn-2 of the glycerol moiety can also use lyso-GPI protein anchors and various base resistant lipids as substrates . Furthermore , we observe that adjacent deletions in several chromosomal regions show strong negative genetic interactions with a single gene on another chromosome suggesting the presence of undeclared suppressor mutations in certain chromosomal regions that need to be identified in order to yield meaningful E-map data .
All living cells define their boundaries by lipid-containing membranes , which in eukaryotes are mainly made of glycerophospholipids ( GPLs ) , sphingolipids and sterols . Phosphatidic acid ( PA ) is an obligatory intermediate in the biosynthesis of all GPLs [1 , 2] . In eukaryotes , PA is generated through acyl-CoA dependent reactions from glycerol-3-phosphate ( G3P ) , which is acylated twice , first on sn-1 by a G3P acyltransferase ( GPAT ) and subsequently on sn-2 by a 1-acylglycerol-3-phosphate acyltransferase ( AGPAT ) . PA is then used for the biosynthesis of mature GPLs such as phosphatidylserine ( PS ) , phosphatidylcholine ( PC ) , phosphatidylinositol ( PI ) and several others . Similarly , mature sphingolipids are built on ceramides , which are synthesized by acyl-CoA dependent ceramide synthases attaching a fatty acid ( FA ) to the free amino group of sphingoid bases . In eukaryotes most of these biosynthetic reactions occur in the ER , but some are also present in mitochondria and in peroxisomes . For the enzymes required to make PA , sphingolipids and sterols , biochemical assays are available and corresponding genes are identified . There is strong evidence that in certain bacteria GPATs and AGPATs belonging to the large lysophospholipid acyltransferase superfamily ( pfam01553 , COG0204 or COG2937 ) have their active site at the cytosolic side of the cytosolic membrane [3–7] . As for eukaryotes , the traditional view is that all acylation reactions in the biosynthesis of PA and triacylglycerides ( TAGs ) as well as the further elaboration of GPLs occur on the cytosolic side of membranes ( or else , the matrix side of chloroplasts ) [1 , 8 , 9] but some notable exceptions may exist . Recent data indicate that the active site of some membrane bound O-acyltransferases ( MBOATs ) probably resides in the ER lumen . According to the Pfam database ( http://pfam . xfam . org/ ) this pfam03062 family of multispan integral membrane proteins presently comprises 16226 sequences belonging to 9933 bacterial and eukaryotic organisms [10] . Many MBOAT proteins acylate lipids such as cholesterol , diacylglycerol ( DAG ) , or lyso-GPLs , but some acylate ER lumenal secretory proteins such as Hedgehog , Wnt , Ghrelin and yeast glycosylphosphatidylinositol ( GPI ) anchored proteins [11–15] . Likewise biochemical investigations localize the highly conserved putative active site His residue of MBOAT proteins to the ER lumen [10 , 16–20] . Moreover , the GPI biosynthetic intermediate PI-Glucosamine is acylated on the inositol moiety in the ER lumen and lyso-GPI anchors are acylated in the lumen of the Golgi , reactions achieved by two acyltransferases not belonging to the MBOAT family [21 , 22] . Acyl-CoA dependent acylation reactions in the lumen of the secretory apparatus imply that acyl-CoA or its acyl group may have to be transported through organellar membranes . Such transport is well documented for the import of fatty acids into mitochondria and peroxisomes , whereby in this latter case it is unclear if acyl-CoA or only acyls are crossing the membrane [23 , 24] . No acyl-CoA transporters however have been described for the ER or Golgi . Besides , the non-specific GPL flippase of the ER has not yet been identified genetically [6 , 25] although the genes encoding the flippases for aminophospholipids at the plasma membrane and Golgi and their antagonists , the scramblases of the plasma membrane are well characterized [6 , 25] .
Here we report on a genetic screen aiming at the detection of genes required for the transfer of GPLs , acyl-CoAs , other acyls or GPl intermediates across organellar membranes . We argued that such lipid flipping transporters might be redundant , in part explaining why they have not yet been identified genetically . We also hypothesized that any growth phenotype due to a lack of some lipid flipping activity in a double mutant may be enhanced if the corresponding lipid is made in reduced amounts . We further assumed that , similar to already characterized lipid flipping transporters [26 , 27] lipid flippases would need to have multiple transmembrane domains ( TMDs ) , i . e . belong to the so called multispan proteins ( MSPs ) . We thus set out to create an E-MAP from a set of 629 strains harboring a deletion or mutation in a MSP with the intent to compare the fitness of double mutants in the absence or presence of Cerulenin , a relatively specific inhibitor of the FA synthase . In the following these E-MAPs are called MSP-E-MAP and MSP/C-E-MAP ( C for Cerulenin ) , respectively . Indeed , all known membrane lipids except for ergosterol contain FAs or derivatives of them . The MSP set contained numerous proteins with 2–14 and in rare cases up to 22 predicted TMDs ( Fig 1A ) . For 340 genes , the gene product had been localized at a particular subcellular location ( Fig 1B , S1 Table ) . Our MSP-E-MAP set was also subdivided manually into 11 different functional categories and was strongly enriched in transporters and lipid biosynthetic enzymes ( S1 Table ) . After data clean up ( S3_supplemental material , materials and methods ) the MSP- or MSP/C-E-MAPs contained 606 or 654 significant negative , and 877 or 896 significant positive interactions , respectively ( unadjusted P value <0 . 005 , see S3_supplemental material , materials and methods ) ( Fig 2A and 2B; S2B Table ) . When comparing ( in May 2015 ) the genetic interactions with those reported by the Krogan and Boone labs , our MSP-E-MAP tested 92’496 genetic interactions not tested before ( Fig 1C ) . E-MAPs require that the 543 genes of the set be deleted once in the MATa background to generate the array , once in the MATα background to generate the query set . After removal of noisy strains ( see S3 Text , Materials and Methods ) the significant S scores obtained in reciprocal [query A x array B] and [query B x array A] crosses where strongly correlated in the MSP-E-MAP ( Fig 1D ) as well as the MSP/C-E-MAP ( Fig 1E ) . Significant positive interactions were found in both MSP/C- and MSP-E-MAP , but in the former some clustered along x- and y-axes , hence were not found in reciprocal crosses ( Fig 1E; S1A–S1D Fig ( Processing of raw E-MAP data ) ) . This may be caused by additional noise due to the addition of Cerulenin but we can’t exclude that there is a biological significance to this phenomenon . To ascertain reproducibility , part of the E-MAP was repeated , i . e . 84 randomly chosen queries were once more crossed with all arrays and analyzed both in the presence and absence of Cerulenin . As seen in Fig 1F , 1G and S2 Fig ( Reproducibility of correlations of the MSP-E-MAP and MSP/C-E-MAP ) , the S scores and correlations of the first complete and the second partial E-MAP were highly correlated , and became even much more correlated when only significant values were considered . As shown in Fig 1H , there also was good agreement between S scores in our MSP-E-MAP with the corresponding interactions in the early secretory pathway E-MAP ( ESP-E-MAP ) [28 , 29] , which inaugurated the E-MAP approach . In an E-MAP , the genetic interactions of a given mutant with all other mutants generates a characteristic interaction profile and any two mutants get a correlation score describing the similarity of their interaction profiles , whereby a very positive correlation of interaction profiles suggests collaboration of the two genes for a single function [28] . As in the ESP-E-MAP report [28] , in our MSP-E and MSP/C-MAPs the average of S scores as a function of the correlation score gets increasingly negative , up to a correlation value of about +0 . 4 , and abruptly becomes more positive at higher correlation values ( Fig 2A and 2B ) . Similarly , the ratio of the number of interactions with significantly positive S scores over the number of interactions with significantly negative S scores drops up to a correlation of about 0 . 4 and then goes up again at higher correlation values ( S1G and S1H Fig ( Processing of raw E-MAP data ) ) . As typical for E-MAPs , our data showed clustering of functionally related genes upon hierarchical clustering and confirmed some genes as "hyper-interactors" since their deletion generated many more interactions than the average deletion . Data also showed preferential interactions between genes of certain functional classes . This and a general statistical analysis of our data are to be found in S1 Text and S8 Fig ( Heat maps and main clusters of the MSP-E-MAP ) , S9 Fig ( Enlargement of regions in heat maps of S8A and S8B Fig showing frequent interactions or correlations between genes belonging to two different clusters ) , S10 Fig ( Frequency of significant interactions and correlations within and amongst different functional classes of genes ) and S11 Fig ( Interdependence of the number of interactions and correlations generated by the MSP-E-MAP ) . There were significant changes between the MSP- and the MSP/C-E-MAP in the sense that some genetic interactions were aggravated , others alleviated on Cerulenin as described in S2 Text and S12 Fig ( Comparison of E-MAPs with or without Cerulenin ) . We had hoped that potential lipid transporters could surface as pairs of less well characterized genes , which would interact more negatively on Cerulenin than without . However , we could not find many such pairs in our data . Most candidate genes were unattractive because of their known localization in the mitochondria , the vacuole or at the plasma membrane , because their interactions were with well-characterized genes not involved in lipid biosynthesis , or because they failed to generate strongly negative S scores . Therefore , we turned our attention to gene pairs giving strong negative interactions in both the MSP- and the MSP/C-E-MAP , which were not strongly aggravated on Cerulenin and in which the function is not fully understood . These criteria are met by the Flc proteins: The S score of the flc1Δ flc2Δ double mutant on Cerulenin drops from -11 . 8 to -12 . 7 , but the double mutant contains two further paralogs , FLC3 and YOR365c , which still are in their wild type ( WT ) state . Flc proteins are widely conserved in fungi , and have three domains: 1 ) an N-terminal hydrophilic domain of 150–200 amino acids forming a lipid binding pocket ( also present in the human Niemann-Pick type C2 protein required for the egress of cholesterol from late endosomes ) , 2 ) a 450 amino acids long very hydrophobic region with >8 predicted TMDs , which is classified as a TRP ( transient receptor potential , pfam 06011 ) domain and is related to human mucolipin and polycystin2 calcium transporters , and finally 3 ) a hydrophilic , 100 to 200 amino acids long non-conserved C-terminus . Flc1 , Flc2 and Flc3 have been localized in the ER and Golgi , whereas Yor365c was reported to be mitochondrial [30 , 31] . In a previous report the flc1Δ flc2Δ strain was found to be nonviable , but growth was partially rescued by 1 M sorbitol , was hypersensitive to the chitin-binding drug calcofluor white ( CFW ) , and cells had thickened cell walls , diminished N-glycan elongation in the Golgi , reduced β1 , 6 glucan synthesis at the plasma membrane , a delay in the maturation of the vacuolar protease CPY and a reduced FAD import into the ER of permeabilized spheroplasts [30] . Furthermore , a recent study proposes that Flc1 , Flc2 and Flc3 mediate or regulate calcium release from intracellular stores into the cytosol in response to hypotonic shock [31] . As shown in Fig 3B and S3 Fig ( Division times of single or combined flc mutants ) , in our background flc1Δ flc2Δ , flc1Δ flc3Δ and flc2Δ flc3Δ double mutant strains were all viable , even if YOR365c was deleted in addition . Flc1Δ flc2Δ had a reduced growth that confirmed the negative genetic interaction seen in our E-MAP . However , the flc1Δ flc2Δ flc3Δ triple mutant was not viable and this argues that the three Flc proteins are redundant with regard to an essential function . We thus created flc1Δ flc2Δ tetoffflc3 strains , in which the tetracycline repressible tetoff promoter replaced the genomic FLC3 promoter . These mutants showed slower growth than flc1Δ flc2Δ cells and ceased to grow on Doxy ( Fig 3B and S3 Fig ( Division times of single or combined flc mutants ) ) . Further deletion of YOR365c had no negative influence on the growth of flc1Δ flc2Δ tetoffflc3 cells . To resolve whether this essential function is restricted to the maintenance of cell wall integrity ( CWI ) , we placed flc1Δ flc2Δ tetoffflc3 yor365cΔ ( in the following abbreviated as 1Δ2Δ3tyΔ ) on 1 . 4 M sorbitol . Sorbitol greatly increased viability of 1Δ2Δ3tyΔ but could not rescue cells on Doxy ( Fig 3C vs . 3B ) , arguing that the essential function of Flc proteins involves more than granting osmoresistance , one of the major tasks of the cell wall . In comparison to WT , 1Δ2Δ3tyΔ were much more sensitive to CFW , as well as to the β1 , 3glucan synthase inhibitor caspofungin , and the inhibitor of N-glycosylation tunicamycin , but , when placed on sorbitol , they remained only sensitive to CFW ( Fig 3C ) . This argues that sorbitol is efficiently combating the cell wall deficiency of 1Δ2Δ3tyΔ , but does not eliminate it completely since CFW hypersensitivity signals an elevated chitin content of the cell wall , suggesting that chitin synthesis remains elevated since the cells continue to sense a cell wall problem [32–34] . In an attempt to get evidence for a hypothetical lipid flippase activity of Flc proteins we used standard tests to measure lipid biosynthesis in flc mutants . For this 1Δ2Δ3tyΔ cells were labeled with [3H]-C16:0 or [3H]-myo-inositol after 16 h of culture with or without Doxy , the time it takes to see a slow down of the growth rate of Doxy-treated cells in comparison with non-treated cells ( S3A Fig ( Division times of single or combined flc mutants ) ) . When the cells were grown with Doxy in the absence of 1 . 4 M sorbitol , [3H]-C16:0 incorporation into GPLs and sphingolipids in 1Δ2Δ3tyΔ cells was very low ( Fig 4A ) . ( Sphingolipids are the only polar lipids remaining after NaOH treatment ) . When grown in sorbitol , the synthesis rate of GPLs was brought back , although not to WT levels ( Fig 4B ) . This difference could not be attributed to a difference in cell viability since colony forming units ( CFU ) after 16 h of culture on Doxy with and without sorbitol was the same ( 39 and 41% , respectively , compared to cells not treated with Doxy ) . In spite of reduced viability by this criterion , all cells still retained a full redox potential ( see below ) . Differently from GPLs , sphingolipid biosynthesis remained inefficient in Doxy treated 1Δ2Δ3tyΔ cells even on sorbitol , both if [3H]-C16:0 or [3H]-myo-inositol was used to label cells ( Fig 4C and 4D ) . In Doxy treated 1Δ2Δ3tyΔ mutants , the incorporation of [3H]-myo-inositol tended to remain in the form of phosphatidylinositol ( PI ) ( Fig 4D ) . Accumulation of PI is expected when ceramides are not made in sufficient quantity since most PI is normally consumed by Aur1-Kei1 transferring inositol-phosphate from PI to ceramides thus generating inositolphosphorylceramides . Three possible reasons for the reduction of the lipid biosynthesis rate in Doxy treated 1Δ2Δ3tyΔ mutants came to our mind: 1 ) The proposed lack of FAD [30] would lead to a severe dysfunction of Ero1 and Pdi1 and a lack of proper oxidative folding of some ER proteins , also affecting enzymes involved in lipid biosynthesis [35] . 2 ) The reduced activity of the lipid biosynthetic enzymes would be related to the growth arrest the cells undergo when incubated with Doxy and 3 ) The flc mutants would indeed have a defect in flipping acyl-CoA or GPLs from the cytosol into the lumen of the ER . To get evidence for a flippase defect in flc mutants , we measured the acyl transferase activity of microsomes in presence of very low concentrations of 16:0-CoA hoping that low concentrations of acyl-CoA would make the PA synthesis more dependent on flippase activities . When microsomes from 1Δ2Δ3tyΔ cells grown on Doxy were incubated with low concentrations of 16:0-CoA ( 0 . 5 μM ) and a 20 fold excess of [14C]-G3P , they made [14C]-PA at a normal rate ( Fig 5 , lanes 1–12 ) . This suggested that Flc proteins were not required for PA biosynthesis and that the ER of 1Δ2Δ3tyΔ cells contained normal GPAT and AGPAT activities . Yet , when the same microsomes were assayed under different conditions , with a 2 fold excess of 16:0-CoA over [14C]-G3P plus 0 . 001% ( 19 mol % ) detergent to permeabilize the microsomes , total incorporation of [14C]-G3P into lipids increased as expected , and it appeared that 1Δ2Δ3tyΔ cells grown on Doxy had rather higher GPAT activity than WT ( Fig 5 , lanes 13–18 ) . This was especially apparent after 1 min of incubation . The normal or elevated microsomal GPAT and AGPAT activities seemed to speak against the possibilities 1 and 2 described above . The increased GPAT activity in microsomes from 1Δ2Δ3tyΔ cells was not easy to interpret: GPT2 and SCT1 , encoding the only known GPATs of yeast are not induced during an unfolded protein response [36] but there may be other reasons for enzyme induction or there may be a difference in microsomal membrane lipids allowing detergent to more easily permeabilize the bilayer or to more easily remove some inhibitory component from the GPATs Gpt2 and/or Sct1 . Trying to get more evidence for hypothesis 3 , i . e . the lack of a lipid flippase in flc mutants , we were faced with the dilemma that such lack is anticipated to destabilize the membrane , which is the obligatory support for any biochemical demonstration of flippase activity . It seemed to us that the discrepancy between experiments labeling intact cells and microsomes ( Fig 4A and 4B vs . Fig 5 ) could indicate that 1Δ2Δ3tyΔ cells , because of a flippase defect , produced leaky or even inverted microsomes giving better access of substrates to the active sites of acyltransferases . Therefore , to further explore whether these cells may have a problem with flipping of acyl-CoA or lyso-PA , we opted for the use of intact cells treated with Digitonin to selectively perforate plasma membranes but leaving the ER intact according to a well established method [37 , 38] . To find the lowest concentrations of Digitonin allowing to permeabilize plasma membranes we used the membrane impermeable 5 , 5’-dithiobis-2-nitrobenzoic acid ( DTNB , Elman’s reagent ) , which produces a colored compound upon reaction with thiol groups present on cytosolic proteins and glutathione [39] . In this way it was found that the speed of the Elman reaction was dependent on the concentration of Digitonin added in the range of 0 to 0 . 02% ( S4 Fig ( Permeabilization of cells with Digitonin and detection of cytosolic thiol groups with DTNB ) ) . It should be noted that in this test the Elman's reagent is in large excess over thiols , so that the reaction reaches plateau when all thiols are oxidized . As these plateaus are the same for WT and 1Δ2Δ3tyΔ cells ( S4 Fig ( Permeabilization of cells with Digitonin and detection of cytosolic thiol groups with DTNB ) ) , it appears that although 60% of Doxy-treated 1Δ2Δ3tyΔ cells are unable to resume growth and form a colony , i . e . have ceased to be a CFU , they apparently still maintain a normal redox potential and in this sense are alive . S4 Fig ( Permeabilization of cells with Digitonin and detection of cytosolic thiol groups with DTNB ) shows that in all cell types tested 0 . 005% Digitonin achieved full , but 0 . 001% only partial reduction of the reagent within 20 min ( see the more detailed description of results in S4 Fig ) . We therefore hoped that substrates for acyltransferases added to minimally permeabilized cells would enter the cytosol and be used for lipid biosynthesis in a still preserved ER membrane without skirting the potential need for physiological ER based lipid flippases in this process . Thus , WT and 1Δ2Δ3tyΔ cells grown in sorbitol and Doxy were preincubated for 30 min with increasing concentrations of Digitonin as above ( S4 Fig ( Permeabilization of cells with Digitonin and detection of cytosolic thiol groups with DTNB ) ) whereupon 10 μM C16:0-CoA plus 10 μM [14C]-G3P ( 0 . 5 μCi ) were added ( Fig 6 ) . When no detergent was added , small amounts of [14C]-G3P seemed to enter cells and produce mainly lyso-phosphatidic acid ( LPA ) , whereby this phenomenon was more pronounced with 1Δ2Δ3tyΔ than WT cells ( Fig 6A ) . With 0 . 001% Digitonin , cells started to make significant amounts of phosphatidic acid ( PA ) , which were similar for WT and 1Δ2Δ3tyΔ cells . With 0 . 005% ( 17 mol% ) Digitonin , 1Δ2Δ3tyΔ cells made significantly more PA and neutral lipids than WT ( Fig 6C ) . PA seemed to be chased with time into DAG or TAG , as if PA synthesis was coming to an early plateau and even early halt . Interestingly , in 1Δ2Δ3tyΔ but not WT cells , the further increase of Digitonin to 0 . 02% ( 44 mol% ) led to a further increase of activity ( Fig 6D ) , although already 0 . 005% fully permeabilizes the plasma membrane of all cell types ( S4A and S4B Fig ( Permeabilization of cells with Digitonin and detection of cytosolic thiol groups with DTNB ) ) . This suggests a basic difference between 1Δ2Δ3tyΔ and WT cells in that in the former Digitonin at 0 . 02% may also affect the permeability of the ER membrane or the enzyme activities in the ER . Importantly , higher GPAT/AGPAT activity in 1Δ2Δ3tyΔ cells was only seen after prolonged culture of cells in Doxy , whereas cells grown in sorbitol without Doxy had normal GPAT/AGPAT activity ( S5B Fig ( 1Δ2Δ3tyΔ mutants have normal GPAT and AGPAT activity when not incubated with Doxy ) vs . Fig 6C ) . This finding suggested that it might be the complete depletion of Flc function that induces these acyltransferases or induces a better accessibility of substrates to the enzymes . We repeated the experiment of Fig 6A , but labeling permeabilized cells with [3H]-C16:0-CoA rather than [14C]-G3P . In this setting , in the mere presence of [3H]-C16:0-CoA ( at 5 mol% ) , mainly PI was made , but 1Δ2Δ3tyΔ made significantly more PI than WT ( Fig 7A ) . This reaction most certainly means that [3H]-C16:0-CoA can enter cells and be used for the acylation of lyso-PI . As soon as [3H]-C16:0-CoA was added together with a 100 fold excess of G3P , cells mainly generated PA , again with 1Δ2Δ3tyΔ making much more PA than WT ( Fig 7B , lanes 1–4 ) . ( As PI and PA had similar migration on TLC , selective transformation of PA into DAG by alkaline phosphatase was used to distinguish PA from PI ( Fig 7B , lanes 2B , 4B ) ) . These data indeed argue that the plasma membrane is not an absolute barrier for [3H]-C16:0-CoA , nor for G3P if very high concentrations are used , but that in 1Δ2Δ3tyΔ cells the barrier is significantly weaker , for both [3H]-C16:0-CoA and G3P as had already been suggested by Fig 6A . In summary , flc mutants exhibit a marked deficit in sphingolipid biosynthesis . The higher GPAT and AGPAT activities at 0 . 02% as compared to 0 . 005% Digitonin observed in Doxy treated 1Δ2Δ3tyΔ cells but not WT ( Fig 6C and 6D ) seem to indicate that Digitonin at 0 . 02% has a different effect on ER membranes from Doxy treated1Δ2Δ3tyΔ cells , be it that it creates better access of G3P to the ER-based GPATs and AGPATs or better derepression of these enzymes . In this context it is noteworthy that yeast GPATs as well as AGPATs have been found to have their active sites on the lumenal side of the ER by conventional biochemical methods [16 , 40] , but more data are required to support or discredit this somewhat unorthodox hypothesis . While this heightened detergent susceptibility of ER membranes in Doxy treated 1Δ2Δ3tyΔ cells certainly represents no strong argument for a role of Flc proteins in flopping GPLs or acyl-CoA across the ER membrane , it cannot exclude these possibilities either and data ask for more direct experiments . GPI lipids are built by stepwise addition of sugars to PI: First , N-Acetyl-Glucosamine ( GlcNAc ) is added to PI , the resulting PI-GlcNAc is then N-deacetylated to PI-GlcN , a FA is attached to the inositol ring to form GlcN- ( acyl ) PI , which latter then is mannosylated and further modified . While the formation of PI-GlcN is known to occur on the cytosolic side of the ER membrane , the mannosylation occurs in the ER lumen . As arv1Δ mutants accumulate GlcN- ( acyl ) PI , it has been proposed that Arv1 , having 4 to 6 predicted TMDs , would serve as a GlcN- ( acyl ) PI flippase [41] . However , a more recent report demonstrated that Gwt1 , the acyltransferase transforming PI-GlcN into GlcN- ( acyl ) PI has its putative active site on the lumenal side of the ER [22] . This would imply that Arv1 acts after the GPI substrate ( PI-GlcN ) has already been flipped and suggests that the generation of GlcN- ( acyl ) PI requires a still unknown flippase other than Arv1 , as well as the flipping of acyl-CoA . As flc mutants show a severe cell wall phenotype , characteristic also for mutants with compromised GPI protein biosynthesis , we desired to test whether flc mutants have any difficulty in making GlcN- ( acyl ) PI . We tried to test this by using a microsomal in vitro assay of GPI biosynthesis with UDP-[3H]-GlcNAc added as the substrate [42] . As shown in Fig 8 , microsomes incubated with UDP-[3H]-GlcNAc make GlcN- ( acyl ) PI when they are allowed to make acyl-CoA in presence of added ATP and CoA . Similar results were obtained with permeabilized cells . In all cases Doxy treated 1Δ2Δ3tyΔ cells made significantly more GlcN- ( acyl ) PI than WT , suggesting that the complete depletion of Flc function induces enzymes for GPI biosynthesis . This finding is reminiscent of the increased GPAT and AGPAT activities in these cells ( Figs 5 and 6C and 6D ) . Once more , results are ambiguous and suggest that the flipping of GlcN-PI and acyl-CoA are normal , but they can’t exclude the possibility that an abnormal permeability or orientation of the ER derived microsomes obviates the need for the corresponding PI-GlcN and acyl-CoA flippases . Unrelated to the search of lipid flippases , our attention was caught by the very strong negative interactions of cst26Δ with elo2Δ and elo3Δ deletions in MSP- and MSP/C-E-MAPs ( S scores between– 9 . 5 and -11 . 0 ) . CST26/PSI1 encodes an acyltransferase that transfers stearic acid ( C18:0 ) from C18:0-CoA onto the sn-1 position of lyso-PI [43] . Apart from a further very strong negative interaction with the chitin synthase CHS1 , cst26Δ interacts only with elo2Δ and elo3Δ , but not elo1Δ ( see below ) . ELO1 allows elongating FAs up to C18:0 . ELO2 and ELO3 are partially redundant , cannot be deleted simultaneously and are required to further elongate FAs up to C26:0 . Each of them is required to make C26:0 in sufficient quantity for the ceramide synthases Lag1 and Lac1 , for which C26:0-CoA is the preferred substrate [44] . Elo2Δ and elo3Δ therefore make markedly reduced amounts of ceramide and mature sphingolipids . However , they grow normally whereas cst26Δ elo3Δ cells grow less rapidly ( S6 Fig ( Comparison of growth rates of elo3Δ , cst26Δ and elo3Δ cst26Δ cells ) ) . While sphingolipids are essential , their presence is dispensable if , due to a gain of function suppressor mutation in SLC1 , cells can make PI carrying a very long chain FA in the sn-2 position of the glycerol moiety , even if this form of PI accounts for only a tiny fraction of membrane lipids [45–48] . Moreover , in WT cells there exists a natural variety of PI having C26:0 in the sn-1 position of glycerol [49] , here called PI’ , which apparently cannot compensate for the complete loss of sphingolipids . PI’ accounted for <1% of the total PI in WT cells , whereby no C26:0 was detected in other GPLs [43 , 49–51] . The genetic interaction of ELO2 and ELO3 with CST26 suggested that PI’ helps cells to overcome a relative lack of sphingolipids and that Cst26 may be the still not identified acyltransferase making PI’ . As shown in Fig 9A , blocking sphingolipid biosynthesis pharmacologically with myriocin ( Myr ) and Aureobasidin A ( AbA ) in BY4741 WT cells leads to the intensification of an [3H]-inositol labeled , mild base sensitive band that is less polar than PI , i . e . has a higher mobility on TLC than PI and that we tentatively considered as PI' . The increased synthesis of PI' can be attributed to the accumulation of C26:0-CoA , which in presence of Myr and AbA cannot flow towards its normal destination , ceramides . Biosynthesis of PI’ is significantly weaker in cst26Δ ( Fig 9A ) . Ordinary PI in WT cells most often contains a C16:0 in sn-1 and a C18:1 in sn-2 [43] , so that phospholipase A2 ( PLA2 ) reduces ordinary PI to a lyso-PI with a C16:0 in sn-1 ( Fig 9B , sample 1 ) . When the [PI + PI’] of WT cells treated with Myr and AbA was subjected to PLA2 treatment , an additional lyso-PI with higher TLC mobility was generated ( Fig 9B , sample 2 ) . This additional lyso-PI is presumed to represent lyso-PI’ , retaining its C26:0 in sn-1; this species is also very abundant in lac1Δ lag1Δ cells ( 2Δ . YDC1 ) that lack ceramide synthases [51] . PLA2 treatment of [PI + PI’] from cst26Δ cells treated with Myr/AbA , generated much less lyso-PI' than treatment of [PI + PI’] from Myr/AbA treated WT cells ( Fig 9B , sample 5 vs . 2; Fig 9C ) . We interpret the data as to say that the bulk of PI carrying C26:0 in sn-1 in WT cells is generated by Cst26 , and that in elo2Δ or elo3Δ , Cst26 generates PIs having C18:0 , C22:0 or C24:0 in sn-1 that may take over a function that is normally exerted by mature sphingolipids . Cst26 thus appears to be an acyl-CoA:lyso-PI acyltransferase that can transfer saturated FAs with 18 to 26 C atoms . Furthermore , the strongly negative genetic interaction between gup1Δ and cwh43Δ caught our attention . The lipid moiety of GPI lipids , once attached to proteins , is modified in the ER through so called GPI lipid remodeling reactions . As seen in Fig 10A , the FA on the inositol moiety is removed by Bst1 , then the FA in sn-2 of the glycerol moiety is removed by the PLA2-like Per1 , then a C26:0 is transferred from C26:0-CoA onto the sn-2 of the lyso-GPI anchor by Gup1 , and finally the modified diacylglycerol or phosphatidic acid moiety can be exchanged for a ceramide or a ceramide-phosphate by Cwh43 [52] . All but the last step are prerequisite for an efficient export of GPI proteins out of the ER [15 , 53–55] . As seen in Fig 10B , the only strong negative genetic interaction in this pathway was between cwh43Δ and gup1Δ , which did not make sense in the linear pathway depicted in Fig 10A , since the absence of Gup1 in this scheme ought to make the deletion of CWH43 of no consequence . Indeed , the gup1Δ cwh43Δ double mutant had a strongly negative S score of– 9 . 9 and grew more slowly than the single mutants also in liquid culture ( S7A Fig ( Growth defects of mutants in the right arm of Chromosome II combined with chs1Δ ) ) . The same genetic interaction observed in the W303 background recently led to the proposal that the lyso-GPI anchors accumulating in gup1Δ may represent suitable substrates for Cwh43 [56] as indicated by the red arrow in Fig 10A . To investigate this in detail , we analyzed the anchor lipids of single and double mutants after metabolic labeling with [3H]-myo-inositol . As seen in Fig 10C , bst1Δ and per1Δ cells did not make any base resistant anchor lipids , whereas gup1Δ produced lyso-PI and several mild base resistant anchor lipids , two of which comigrated with IPC/B and IPC/C , the typical base resistant anchor lipids of WT , whereas 3 others were not present in WT . Interestingly , not only IPC/B- and IPC/C-type anchors , but also the 3 abnormal mild base-resistant lipids were no more observed in a gup1Δ cwh43Δ double mutant . The results argue that lyso-GPI anchors indeed are a substrate for Cwh43 , as was also proposed by others [56] . In this report it also was proposed that ceramide anchors can be added to GPI-anchors accumulating in per1Δ mutants ( Fig 10A ) , but mild base resistance of anchor lipids had not been tested and in our genetic background we can’t see any mild base resistant anchors in per1Δ ( Fig 10C , lane 7 ) . Moreover , it seems that in the gup1Δ background , Cwh43 may transfer also ceramides other than the typical phytosphingosine-C26:0 or phytosphingosine-C26:0-OH giving rise to IPC/B and IPC/C , respectively . Addition of a dihydrosphingosine-C26:0 may account for the most hydrophobic lipid ( highest TLC mobility ) , whereas the utilization of ceramides with shorter or more hydroxylated FAs may explain the appearance of the more polar species . The negative S score of the gup1Δ cwh43Δ ( Fig 10B ) argues that the base resistant GPI anchor lipids of gup1Δ increase the amount of functional GPI proteins being integrated into the cell wall . Our E-MAP gene set comprised 99 uncharacterized open reading frames ( ORFs ) . These 99 uncharacterized ORFs however made almost as many significant genetic interactions as the well-characterized genes suggesting that , although still uncharacterized , they are not functionally unimportant or redundant . Some 23 of the 99 non-characterized ORFs were present in 97 gene pairs generating strongly positive correlations ( >0 . 4 ) , whereby in no such pair the partners showed significant genetic interaction with each other ( S2D Table ) . The many high correlations of a deletion in the acyltransferase paralog YDR018c or in the lipase paralog YFL034w with deletions in amino acid permeases suggest that these ORFs may disturb amino acid transport or signaling mediated through such transporters , possibly by disturbing the lipid composition of membranes . Furthermore , in the MSP as well as the MSP/C screen the ENV10-SSH1 pair was highly correlated ( > 0 . 56 ) and showed very negative S scores ( < - 13 ) . ENV10 is a not very well characterized gene somehow involved in secretory protein quality control [57] , whereas SSH1 codes for a non-essential homolog of the essential Sec61 translocon subunit of the ER . The very strong ENV10-SSH1 interaction ( not reported in BIOGRID ) suggests that Env10 , having 4 TMDs and a KXKXX retention signal , may play a role in co-translational protein translocation . The E-MAP set contained a group of 12 MSP proteins all encoded next to each other in the region between 250’000 and 390’000 bp of the right arm of chromosome II ( Chr . II ) that presented similar correlations although they are not functionally related ( Fig 11A , blue color ) . These chromosomally clustered positive correlations may be due , at least in part , to uniformly negative genetic interactions of all these genes with chs1Δ , all genes having S scores < -3 , the genes in the center of the region even <-10 ( Fig 11A ) . Indeed , the colony sizes on the final MSP-E-MAP plates of these pairs on both [query chs1Δ x array B of Chr . II] as well as on reciprocal plates were almost the size of the lethal tda5Δ x tda5Δ control ( Fig 11B ) . The growth rates of the double mutants in liquid and solid media were however normal ( S7A and S7B Fig ( Growth defects of mutants in the right arm of Chromosome II combined with chs1Δ ) ) . To test if negative S-scores appeared also in mutants in that region coding for other proteins than MSPs , we crossed the WT and chs1Δ query strains with an array plate containing all the genes of this chromosomal region . This experiment showed that chs1Δ combined with a deletion in any gene of this region , whether coding for an MSP or not , had a very negative S score . As shown in Fig 11C , plates showed a regional reduction of colony sizes for genes on Chr . II , whereas the crosses involving genes located on other chromosomes ( not boxed ) , showed no difference of colony sizes between the WT query plate and the chs1Δ query plate . In double mutants of chs1Δ combined with chs2-DAmP , fig1Δ , fat1Δ , cst26Δ , or qdr3Δ , we amplified by genomic PCR all genes and intergenic regions starting from CHS2 up to QDR3 and found that there were no rearrangements or deletions present apart from the intended single gene deletion . Interestingly , cst26Δ is one of the gene deletions sitting in the middle of the chromosomal region where deletions show negative S scores with chs1Δ ( Fig 11A ) . As mentioned above , elo2Δ and elo3Δ also have very negative S scores in combination with cst26Δ , similar to chs1Δ ( Fig 11A ) . In this case however only very few genes immediately adjacent to cst26Δ show a negative S score with elo2Δ or elo3Δ . The phenomenon of regionally concentrated negative interactions shown in Fig 11A is not an isolated phenomenon , since several such regions can easily be identified on a heat map of S scores where the genes are ordered according to their chromosomal location as shown in Fig 12 . As the order in this matrix clusters each gene with the genes that sit next to it on the chromosome , all the irrelevant very negative interactions generated by proximity of two deletions on a same chromosome ( < 100 kb ) and hence marked with grey dots are clustering along the diagonal ( Fig 12 ) . Uniform interactions of all deletions in certain chromosomal regions with single deletions on another chromosome or a distant region of the same chromosome appear as short green or red stripes; they are pointed out by numbered arrows , whereby arrow 1 points to the interactions of chs1Δ with genes on the right arm of Chr . II discussed above ( Fig 11A ) . Importantly , these chromosomally clustered interactions do not involve the “hyper-interactors” that show interactions throughout the heat map ( S8A Fig ( Heat maps and main clusters of the MSP-E-MAP ) ) . We believe that these regionally concentrated negative interactions with a deletion at a distant locus ( e . g . chs1Δ ) are caused by non-declared intergenic suppressor mutations that rescue the growth defect caused by the distant deletions . For example , a gain of function suppressor mutation in CHS2 present in the chs1::ura3MX query strain may be present in all crosses of that query except the ones with genes in the vicinity of CHS2 , where the kanMX-marked array gene will be selected for and the suppressor in CHS2 is likely to be lost . Such a suppressor mutation in CHS2 would not exist in elo2Δ and elo3Δ queries and , if it existed , would not have any genetic interaction with elo2Δ and elo3Δ strains , explaining the absence of a regional effect around CST26 in the elo2Δ cst26Δ and elo3Δ cst26Δ mutants ( Fig 11A ) . ( The strong negative S score of fat1Δ elo3Δ may not be a neighboring effect but an independent genetic interaction , since Fat1 is the only one of 6 yeast acyl-CoA synthases that can activate very long chain FAs and hence may prepare the substrate for Cst26 ) . Negative genetic interactions appearing in SGA have been utilized before to localize and identify a suppressor mutation in the SSD1 locus , which suppresses growth effects of mutations in the Cbk1 kinase signaling pathway [58] . In our E-MAP the query strains were generated by swapping the kanMX marker for the ura3MX marker so that suppressors of the array strains had a good chance to be transferred to the query strains ( see S3 Text , Materials and Methods ) . This can explain why the phenomenon for all 8 arrows of Fig 12 was seen symmetrically in both query x array as well as array x query plates . We indeed found that the distant deletions that generated the concerted negative S scores in certain chromosomal regions all had either reduced viability , reduced competitive fitness , a sporulation defect , or reduced respiratory capacity and therefore were susceptible to be overgrown by suppressors . It is conceivable that such undeclared mutations may cause some noise also in other E-MAP studies using the strategies we used . For instance , in another E-MAP study [59] , 6 of the 18 negative interactions of tda5Δ were comprised between YOL108cΔ and YOL27cΔ on the left arm of Chr . XV , the same region as pointed by arrow 4 in Fig 12 , although none of these negatively interacting deletions were present in our MSP deletion set . Moreover , there are high correlations among functionally unrelated but regionally concentrated genes also in previously published E-MAPs from other groups [60–62] . We tried to do a chemogenetic screen in order to identify lipid flippases , the existence of which has been postulated since a long time based on microsomal assays and structural studies showing that certain acyltransferases have their active site in the lumen of the ER . No obvious candidates emerged from this , but , in view of the unusual detergent sensitivity and permeability of the plasma and ER membranes of flc mutants , a flippase activity of Flc proteins remains a definite possibility , which needs to be pursued by trying to reconstitute Flc proteins into large unilamellar vesicles , e . g . by using and adapting the approaches recently established in our lab [63] . LplT is a lyso-PE transporter of the inner membrane of E . coli [64] . Deltablasting ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) shows some highly significant homologies to 13 yeast genes having > 8% identities covering > 90% of lplT sequence ( S2F Table ) . Ten of them were present in our final array set but none of the 10 was involved in any interaction that got severely aggravated ( more negative S score ) on Cerulenin . While such homologs remain candidates for GPL flippases , several are localized at the plasma membrane and have well defined transporter functions and the genetic interactions of the others make it unlikely that they would be ER lipid flippases ( S2F Table ) . Another interesting flippase candidate would be the TMEM16 channel homologue IST2 , which was not present in our deletion library [65] . Our unexpected observation of genetic interactions of certain genes with deletions in an entire chromosomal region may necessitate some additional filtering of the genetic interactions generated in certain E-MAP studies .
All methods have been utilized before and are described in S3 Text and S13 Fig ( Titration of Cerulenin to determine its optimal concentration for the MSP/C-E-MAP ) . | All living cells define their boundaries by lipid-containing membranes , which are impermeable to ions and water-soluble metabolic intermediates , and thus allow maintaining constant conditions inside the cells and stopping metabolic intermediates from diffusing away . Membranes are formed by amphiphilic lipids that have a hydrophilic and a hydrophobic component . Such lipids form flat double-layered sheets ( bilayers ) wherein the hydrophilic components of the constituent lipids are directed towards the aqueous surroundings , the hydrophobic ones populate the center of the bilayer . Membranes grow when enzymes resident in the bilayer synthesize new amphiphilic lipids . These enzymes have their active site on one side of the membrane and insert the newly made lipids in only one of the two layers . To ensure symmetric growth of membranes , cells need flippases catalyzing the transfer of lipids from one into the other layer . To identify unknown flippases we performed a chemogenetic interaction screen able to bring to light functions of unknown proteins through their genetic interaction with genes of known function . The data point to Flc proteins as potential lipid flippases of the endoplasmic reticulum , reveal novel lipid modifying activities of Cst26 and Cwh43 and suggest that undeclared suppressor mutations in certain chromosomal regions can generate false genetic interactions . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"sphingolipids",
"membrane",
"proteins",
"planar",
"chromatography",
"genetic",
"interactions",
"cellular",
"structures",
"and",
"organelles",
"thin-layer",
"chromatography",
"research",
"and",
"analysis",
"methods",
"lipids",
"chromosome",
"biology",
"cell",
"membranes",
"chromatographic",
"techniques",
"biochemistry",
"cell",
"biology",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"chromosomes"
] | 2016 | Chemogenetic E-MAP in Saccharomyces cerevisiae for Identification of Membrane Transporters Operating Lipid Flip Flop |
Model-based epidemiological assessment is useful to support decision-making at the beginning of an emerging Aedes-transmitted outbreak . However , early forecasts are generally unreliable as little information is available in the first few incidence data points . Here , we show how past Aedes-transmitted epidemics help improve these predictions . The approach was applied to the 2015–2017 Zika virus epidemics in three islands of the French West Indies , with historical data including other Aedes-transmitted diseases ( chikungunya and Zika ) in the same and other locations . Hierarchical models were used to build informative a priori distributions on the reproduction ratio and the reporting rates . The accuracy and sharpness of forecasts improved substantially when these a priori distributions were used in models for prediction . For example , early forecasts of final epidemic size obtained without historical information were 3 . 3 times too high on average ( range: 0 . 2 to 5 . 8 ) with respect to the eventual size , but were far closer ( 1 . 1 times the real value on average , range: 0 . 4 to 1 . 5 ) using information on past CHIKV epidemics in the same places . Likewise , the 97 . 5% upper bound for maximal incidence was 15 . 3 times ( range: 2 . 0 to 63 . 1 ) the actual peak incidence , and became much sharper at 2 . 4 times ( range: 1 . 3 to 3 . 9 ) the actual peak incidence with informative a priori distributions . Improvements were more limited for the date of peak incidence and the total duration of the epidemic . The framework can adapt to all forecasting models at the early stages of emerging Aedes-transmitted outbreaks .
Model-based assessments must be done in real time for emerging outbreaks: this was the case in recent years for MERS-CoV in the Middle East [1–3] , Ebola virus in West Africa [4–10] , chikungunya virus ( CHIKV ) [11] and Zika virus ( ZIKV ) [12–14] in the Americas . These analyses often focused on transmissibility and reproduction numbers rather than on forecasting the future impact of the epidemic . Indeed , forecasting is difficult before the epidemic reaches its peak , all the more when information on natural history , transmissibility and under-reporting is limited [15] . Yet , it is precisely at the beginning of an outbreak that forecasts would help public health authorities decide on the best strategies for control or mitigation . Several methods have been used to make epidemic predictions , including exponential growth models [5 , 16] , sigmoid-based extrapolations [17] , SIR-type models [18] and more realistic model accounting for spatial and population structure [19] . But in addition to specifying a model , selecting good parameter values is also essential to obtain good predictions . In models for directly transmitted diseases , this can come from realistic demographic and behavioral characteristics , for example the contact frequency between individuals [20] , mobility patterns [21–24] , and from clinical and epidemiological characteristics like the duration of the serial interval [25] . Such information is less easily available and more limited for mosquito-transmitted diseases [26] . However , outbreaks of the same disease or diseases with similar routes of transmission may have occurred in the same or similar locations , so that relevant information may be recovered from the analysis of past outbreaks . Here , we show that past outbreaks of Aedes-transmitted diseases can substantially improve the epidemiological assessment of diseases transmitted by the same vector from early surveillance data . We introduce a hierarchical statistical model to analyze and extract information from historical data and obtain a priori distributions for required epidemiological parameters [27] . The method is illustrated with ZIKV outbreaks in the French West Indies between December , 2015 and February , 2017 , using historical data regarding CHIKV and ZIKV epidemics in French Polynesia and the French West Indies between 2013 and early 2015 . We assess the improvement in predictability of several operational indicators using different choices of a priori distribution and according to epidemic progress .
Surveillance data on the 2015–2017 ZIKV epidemics in Guadeloupe , Martinique and Saint-Martin was collected by local sentinel networks of general practitioners and reported weekly by the local health authorities ( Fig 1 ) [28] . Cases of ZIKV infection were defined as “a rash with or without fever and at least two signs among conjunctivitis , arthralgia or edema” . We obtained numbers of suspected cases by week for each island , extrapolated from the number of active sentinel sites ( S1 Dataset ( D 1 ) ) . In the West Indies , local health authorities described the situation as “epidemic” when incidence was larger than 1 per 2 , 000 population per week ( i . e . 200 cases in Guadeloupe and Martinique [29] , and 20 cases in Saint-Martin ) . Following this description , we defined the “S” ( -tart ) of the epidemic as the first week above this threshold , the “P” ( -eak ) date when incidence was the highest , and the “E” ( -nd ) of the epidemic as the third consecutive week below the threshold ( to ascertain the downwards trend ) . The time interval from “S” to “E” corresponds to the period of “high epidemic activity” . We then analyzed historical data on the spread of emerging Aedes-transmitted diseases in similar locations . CHIKV epidemics occured in the same three islands during 2013–2015 . Both diseases were transmitted by the same vector ( Aedes aegypti ) , circulated in the same immunologically naive populations within a period of two years , had the same kind of clinical signs ( i . e . , fever , rash and arthralgia ) and were reported by the same surveillance system . Surveillance data on CHIKV epidemics in the French West Indies was available from local health authorities ( S2 Dataset ( D 2 ) ) [30] . Finally , we also selected the ZIKV and CHIKV epidemics that occured in six islands or archipelagoes of French Polynesia between 2013 and 2017 , as they provided information on the differences between the two diseases . Surveillance data regarding the outbreaks in French Polynesia ( S3 Dataset ( D 3 ) ) was collected following similar methods as in the French West Indies [31 , 32] . The ZIKV outbreaks in Guadeloupe , Martinique and Saint-Martin were modelled separately using a dynamic discrete-time SIR model within a Bayesian framework . Briefly , the two main components of the model were: ( i ) a mechanistic reconstruction of the distribution of the serial interval of the disease ( the time interval between disease onset in a primary case and a secondary case ) that allows bypassing vector compartments; ( ii ) a transmission model for the generation of observed secondary cases in the human host . The generation time distribution was reconstructed by estimating the durations of each part of the infection cycle using disease- and mosquito-specific data from the literature , and assuming a fixed local temperature of 28°C , as described in more detail in the supplementary appendix . This led to gamma distributions with mean 2 . 5 weeks ( standard deviation: 0 . 7 ) for ZIKV and with mean 1 . 6 weeks ( sd: 0 . 6 ) for CHIKV . Then , we linked weekly observed incidence Ot , X to past incidence with: O t , X | O 0 , ⋯ , t - 1 , X , R 0 , X , ρ X , ϕ X ∼ NB ( R 0 , X S t , X N ∑ n = 1 5 w X , n O t - n , ϕ X ) ( 1 ) where subscript X refers to disease ( X = C for CHIKV or X = Z for ZIKV ) , R 0 , X is the basic reproduction number , N the population size , and S t , X = N - ∑ k = 0 t - 1 O k , X / ρ X the number of individuals susceptible to infection at time t where ρX is the reporting rate . The term ∑ n = 1 5 w X , n O t - n accounts for exposure to infection at time t , where wX , n is the discretized serial interval distribution . The variance is computed as the mean divided by the overdispersion parameter ϕX . The model was implemented in Stan version 2 . 15 . 1—R version 3 . 4 . 0 [33–35] . More details regarding the epidemic model and the Stan code are available in the supplementary appendix . To analyze Zika epidemics , the reproduction ratio R 0 , Z , the reporting rate ρZ and the overdispersion parameter ϕZ must be estimated . For ϕZ , we used a non-informative prior in all cases [36] . For R 0 , Z and ρZ , we designed three different prior distributions , labelled as “non-informative” ( NI ) , “regional” ( R ) , or “local” ( L ) and described below . The NI prior distributions expressed vague characteristics of the parameters: R 0 , Z will be positive and likely not greater than 20; and ρZ will range between 0 and 1 . The R and L priors were derived from the analysis of D 2 and D 3 datasets in three steps . We first analysed jointly the three CHIKV epidemics in dataset D 2 using model 1 , introducing a two-level hierarchical structure for the island-specific parameters: an island-specific reproduction number R 0 , C , i sampled from a top-level regional distribution N ( μ R 0 , C , σ R 0 , C ) , and similarly logit ( ρ C , i ) ∼ N ( μ ρ C , σ ρ C ) for the logit of the reporting rate . We obtained the posterior distributions of these parameters for the CHIKV outbreaks in each island π ( R 0 , C , i , ρ C , i | D 2 ) , as well as that of the hyperparameters π ( μ R 0 , C , σ R 0 , C , μ ρ , C , σ ρ , C | D 2 ) . We then analysed dataset D 3 using the same hierarchical structure as above and introducing relative transmissibility and reporting of ZIKV with respect to CHIKV as R 0 , Z = β R 0 R 0 , C and ρZ = βρ ρC . We thus used French Polynesia data to estimate the relative transmissibility π ( β R 0 | D 3 ) and the relative reporting rate π ( β ρ | D 3 ) of ZIKV with respect to CHIKV in the French West Indies . In a third step , these posterior distributions were combined to obtain the R and L prior probability distributions as described in Table 1 . The R and L priors differed in how much island-specific information they contained , the R priors being altogether less informative than the L priors . More precisely , the L priors used the bottom level in the hierarchical description , actually combining island-specific distributions for transmission and reporting of CHIKV with relative ratios β R 0 and βρ of ZIKV to CHIKV . The R priors , on the contrary , were based on the top-level distributions in the hierarchical description , and can be interpreted as providing information for a “typical” island of the French West Indies rather than for a specific island . Alternative prior distributions were considered in the sensitivity analysis and results are reported in the supplementary appendix . We fitted model 1 to ZIKV data separately in Martinique , Guadeloupe and Saint-Martin using the K first weeks of data ( varying K from 5 to the number of weeks in the epidemic ) and each a priori distribution ( NI , R or L ) , to obtain posterior distributions for parameters R 0 , Z , ρZ and ϕZ for every combination of island , K and choice of prior . Then , using each set of posterior distributions , the epidemics were simulated forward from week K + 1 for two years using a stochastic version of the model described in Eq 1 . We used 16 , 000 replicates to compute the predictive distribution of the weekly number of future incident cases and a trajectory-wise 95% prediction band [37] . Using these simulated trajectories , we also computed the predictive distributions of four indicators of operational interest , for direct comparison with observed values: The predictive distributions were compared using two measures of forecasting quality: ( i ) accuracy , i . e . the root-mean-square difference between predicted and observed values , and ( ii ) sharpness , i . e . the mean width of the 95% prediction band [39] . In order to improve clarity , these values were multiplied by −1 so that a higher value means a better accuracy or sharpness .
The timecourse of the ZIKV epidemics in Guadeloupe , Martinique and Saint-Martin between December , 2015 and February , 2017 differed markedly: the initial growth was early and sudden in Martinique , while it was delayed in Guadeloupe and Saint-Martin , starting only after four months of low-level transmission ( Fig 1 ) . The epidemic showed a sharp peak in Guadeloupe , reaching a maximal weekly incidence of 6 . 9 cases per 1 , 000 inhabitants 9 weeks after the start of the period of high epidemic activity . In Martinique and Saint-Martin , weekly incidence reached a maximum of 4 . 8 cases per 1 , 000 inhabitants after a period of 10 and 21 weeks , respectively . Conversely , the period of high epidemic activity was longer in Martinique and Saint-Martin ( 37 and 48 weeks , respectively ) , than in Guadeloupe ( 27 weeks ) . In the end , a total of about 37 , 000 cases were observed in Martinique ( 97 cases per 1 , 000 inhabitants ) , more than in Saint-Martin ( 90 cases per 1 , 000 inhabitants ) and Guadeloupe ( 77 cases per 1 , 000 inhabitants ) . The CHIKV epidemics observed in the same three islands of the French West Indies during 2013–2015 are shown in Fig 1B , and the CHIKV and ZIKV epidemics observed in French Polynesia during 2013–2015 in Fig 1C . The a priori distributions on the reproduction ratio and reporting rates for the ZIKV epidemics in the French West Indies defining the NI , R or L approaches are shown in Fig 2 . The R priors were wide , with 95% credible intervals between 0 . 5 and 2 . 5 for R 0 , Z and between 0 and 0 . 30 for ρZ . On the contrary , the more specific L priors on R 0 , Z were highly concentrated around 1 . 5 in Guadeloupe and 1 . 3 in Martinique , and ranged between 1 . 0 and 1 . 8 in Saint-Martin . Likewise , the island-specific priors on ρZ carried more information that their regional counterpart , peaking around 0 . 19 in Guadeloupe , and covering wider intervals in Martinique ( 0 . 20–0 . 40 ) and Saint-Martin ( 0 . 03–0 . 39 ) . Fig 3 shows the future course of the ZIKV epidemics predicted using data available two weeks after date “S” in each island , for the three choices of prior distributions . At this week , predictions with the NI priors were largely off-target , overestimating the future magnitude of the epidemic in all three islands . Using the R priors reduced the gap between forecasts and future observation . Major improvements in both accuracy and sharpness were obtained only with L priors . These results were typical of the initial phase of the epidemics , as shown in Fig 4 . Overall , the quality of the forecasts improved as K increases in all three islands and also as prior distributions brought more specific information . Results of the sensitivity analysis show that L priors defined here outperformed or performed at least equivalently to the alternative priors definition tested ( S1 Appendix ) . The posterior distributions of the parameters built up differently as data accrued for R 0 , Z and ρZ . For all three choices of prior distributions , the posterior distributions of R 0 , Z quickly overlaid after a few points of incidence data were observed ( Fig 5A ) . In sharp contrast , the posterior distributions of ρZ could remain affected by the choice of prior distributions ( Fig 5B ) . In Martinique and Saint-Martin , ρZ remained essentially unidentified with the NI priors for the entire duration of the epidemic , with 95% credible intervals ranging approximately from 20 to 80% , even though the posterior mean was close to the estimates obtained with the more informative priors ( around 25% at the end ) . Informative priors allowed for a more precise estimation of ρZ , and this remained the case over the whole course of the epidemics . In Guadeloupe , all posterior distributions for ρZ were similar after the peak , irrespective of the choice of priors . The forecasts of the four indicators of operational interest produced before peak incidence were contrasted . With NI priors , forecasts of total epidemic size overestimated the final counts by on average 3 . 3 times ( range: 0 . 2 to 5 . 8 ) ( Fig 6A ) , with substantial variations in the forecasts from one week to the next . For instance in Martinique , projections varied from 105 , 000 total observed cases ( 95% prediction interval [95%PI]: 5 , 300–340 , 000 ) on February 7th to 8 , 100 ( 95%PI: 5 , 700–13 , 700 ) on February 14th . For the same indicator , forecasts using the R priors were only 1 . 7 times too high ( range: 0 . 4 to 3 . 3 ) and those produced using L priors were only 1 . 1 times too high ( range: 0 . 4 to 1 . 5 ) . As a comparison , with the L priors on February 7th , the forecast of epidemic size was 47 , 200 ( 95%PI: 20 , 900–71 , 100 ) in Martinique , much closer to the final count of 37 , 400 observed cases at the end of the epidemic in this island . Similarly , forecasts of maximal weekly ( observed ) incidence produced before date “P” were generally too large when using NI priors , on average 15 . 3 times higher than the actual maximal weekly incidence observed thereafter ( range: 2 . 0 to 63 . 1 ) and with large fluctuations ( Fig 6B ) . Using informative prior distributions improved the forecasts , reducing the maximum predicted incidence to 7 . 5 times higher ( range: 3 . 0 to 25 . 5 ) with the R priors and 2 . 4 times higher ( range: 1 . 3 to 3 . 9 ) with the L priors . In all cases , forecasts of maximal incidence were never smaller than actual incidence . Forecasting the dates of interest in the epidemics showed mixed results , with less differences depending on the choice of priors . The forecasts of the date of peak incidence were too late by on average 1 . 0 month ( range: -3 . 4 to +3 . 9 ) using the NI priors , with large variations ( Fig 6C ) . Forecasts were only slightly better with the R priors ( +0 . 7 months , range: −2 . 1 to +2 . 8 ) and the L priors ( +0 . 9 months , range: -1 . 1 to +3 . 2 ) , but sharper . Better forecasts were obtained for Martinique than for the other islands . The forecasts of the total duration of the period of high epidemic activity were overestimated by a factor 1 . 3 ( range: 0 . 1 to 2 . 6 ) with NI priors , again with high variability from week to week ( Fig 6D ) . Informative priors brought a small improvement , in particular regarding the stability of the forecasts over time , overestimating the actual duration by a factor 1 . 2 ( range: 0 . 7 to 2 . 0 ) with R priors and by a factor 1 . 1 ( range: 0 . 9 to 1 . 7 ) with L priors .
Obtaining reliable model-based forecasts in real time at the beginning of an epidemic is a difficult endeavour . It is however precisely during these periods that forecasts may have the most impact to guide interventions . Here , we compared several approaches to provide forecasts for ZIKV epidemics from an early point in a retrospective analysis of the outbreaks that occurred in the French West Indies in 2015–2017 . We found that the accuracy and sharpness of the forecasts before peak incidence were substantially improved when a priori information based on historical data on past epidemics was used . The three ZIKV outbreaks in the French West Indies provided an ideal situation to look for ways to improve prediction of Aedes-transmitted diseases using historical data . Indeed , CHIKV outbreaks had been observed in the same locations about two years before ZIKV , CHIKV is also transmitted by Aedes mosquitoes , and both viruses spread in a region where the populations were immunologically naive at first . Furthermore , all epidemics were observed by the same routinely operating GP-based surveillance networks , and the three locations benefit from a mature public health system , with easy access to medical consultation and individual means of protection . Last , pest control is done in routine , with additional intervention showing limited efficacy in this context [40 , 41] . This motivated our decision to use constant parameters for transmsission and reporting over time in the modelling . Having established the many similarities between ZIKV and CHIKV epidemics regarding transmission and reporting , we assumed that information could be transported ( as defined in [42] ) between diseases and between places . Bayesian approaches allow such transportation using informative a priori distributions on model parameters [27] . Historical data on CHIKV epidemics was therefore used to build informative a priori distributions on the two key parameters R 0 , Z and ρZ . Hierarchical models are particularly adapted to this task , as they naturally pool information among several past epidemics and capture both within- and between-location variability [43] . We used separate hierarchical models to obtain information about ( i ) transmissibility and reporting during CHIKV outbreaks in the French West Indies and ( ii ) relative transmissibility and reporting between ZIKV and CHIKV in French Polynesia , rather than from a global joint model ( as in [44] ) . This choice was made to show that prior information can be combined from separate sources in a modular way . We finally considered two versions of informative priors to capture different degrees of knowledge . The “Local” priors corresponded with estimates for the previous CHIKV epidemic in the same island , therefore including island-specific epidemic drivers , such as population structure and distribution , socio-economic circumstances and environmental conditions . On the other hand , the “Regional” priors encompassed the diversity of past observations within the region , making priors valid for a typical island of the West Indies , especially when no other epidemic has been observed previously . Analyses conducted at the early stage of an epidemic using non-informative a priori distributions—as is often done—led to poor forecasts before the peak of incidence was reached , an observation already made in other studies [45] . Indeed , early forecasts of epidemic size were largely off-target and unstable , varying between 0 . 2 and 5 . 8 times the eventually observed total incidence . Worst case projections on maximal incidence were very imprecise , ranging between 2 and 63 times the eventually reached maximal weekly incidence . Using historical data led to a substantial increase of the quality of these forecasts from the very early stages of the epidemics . Using “local” priors , the ratio between forecasts and reality ranged between 0 . 4 and 1 . 5 for epidemic size and between 1 . 3 and 3 . 9 for maximal incidence . The less specific “regional” priors increased accuracy and sharpness as well , though to a lesser extent . However , the date of peak incidence and the date of the end of the period of high epidemic activity were only slightly improved by integrating historical data . The posterior distributions of all forecasted quantities changed as more data was included ( Fig 4 ) . The posterior estimates of R 0 , Z were quickly similar and in good agreement with prior information . On the contrary , the reporting rate ρZ remained essentially unidentifiable until after the incidence peak in Guadeloupe , and to the end of the outbreak in Martinique and Saint-Martin . This suggests that prior information is essentially required for the reporting rate , a difficult to estimate quantity as already noted [15] . Sensitivity analysis support this result . Indeed , providing informative prior on ρZ only leads to similar results as providing it for both ρZ and R 0 , Z ( S1 Appendix ) . Predicting the future course of epidemics from an early point is increasingly seen as a problem of interest [8 , 9 , 46] , and forecasting challenges have been set up for influenza [47] , for Ebola [45] and for chikungunya [48] . Comparing and systematically evaluating models’ forecasting performances is still at the beginning . As of now , comparisons targeted the merits of different models including exponential growth models , sigmoid models , or mechanistic epidemic models [45] . Our work provides a complementary approach where information from past epidemics is combined using hierarchical models to inform on parameter ranges , thus increasing the reliability of early forecasts . It was applied here to a dynamic discrete-time SIR model that for its parsimony is well-adapted to real-time forecasting . Complex mechanistic models can provide a more realistic description of the epidemic , accounting , for instance , for heterogenous spatial distribution of individuals and mobility coupling—a relevant ingredient for describing epidemics in more extended spatial areas – , or vector population dynamics and its mixing with humans . Our framework could in principle be adapted to these more sophisticated models . Only recently have hierarchical models been used for modelling multiple epidemics , for instance with the joint analysis of six smallpox epidemics [49] , for transmissibility and duration of carriage in the analysis of multistrain pneumococcus carriage [50] , or for forecasting seasonal influenza [51] . Other modelling papers specifically attempted to take advantage of the similarities between different Aedes-transmitted diseases , e . g . , by estimating the risk of acquiring chikungunya from the prevalence of dengue [52] or by assesssing the spatio-temporal coherence of chikungunya , Zika and dengue [53] . Also , using informative priors to make up for the lack of information during the early stages of an epidemic has been done before . For instance , a priori information from the ZIKV epidemics in French Polynesia has been used to support the early forecasts of health-care requirements for the ZIKV epidemic in Martinique [38] . In this case , however , authors concluded that a prior built from an epidemic in a different location resulted in inaccurate predictions at the early stage . We found a similar result in a sensitivity analysis: the direct use of information from ZIKV in French Polynesia , or alternatively the direct use of information from CHIKV in the French West Indies , without adjusting for ZIKV , leads to poor overall forecasting quality compared to the L prior considered here ( S1 Appendix ) . This shows that the choice of appropriate historical data is the cornerstone of any such attempt . Yet little is known regarding the comparative epidemiology of diseases in the same or similar places and on the condition where transportability can be assumed . For influenza , it has been reported that the reproduction ratios in two successive flu pandemics ( 1889 and 1918 ) showed substantial correlation ( r = 0 . 62 ) in the same US cities , even years apart [54] . For Aedes-transmitted diseases , comparisons of ZIKV and dengue virus outbreaks [55] and of ZIKV with CHIKV outbreaks [44] in the same locations have highlighted similarities in the epidemic dynamics . In any case , careful consideration of all the factors that may influence transmission and reporting is needed . For example , contrary to CHIKV , some cases of sexual transmission have been reported for ZIKV [56 , 57] . Yet , no epidemics were seen in locations without enough Aedes mosquitoes , as for example in metropolitan France , despite the introductions of several hundreds ZIKV infected cases [58] . This justifies the use of CHIKV epidemic data to provide prior information on ZIKV in the epidemic context considered here . More generally , documenting , analyzing and comparing more systematically past epidemics [59] is necessary to provide the data required to derive prior information . In particular , informed with epidemiological records of the recent CHIKV and ZIKV epidemics , our approach could be applied to potential future emergences of other Aedes-transmitted diseases such as Mayaro virus [60] , Ross River virus [61] or Usutu virus [62] once transportability is deemed plausible . | In December , 2015 , Aedes mosquito-transmitted Zika outbreaks started in the French West Indies , about two years after chikungunya epidemics , spread by the same mosquito , hit the same region . Building on the similarities between these epidemics—regarding the route of transmission , the surveillance system , the population and the location—we show that prior information available at the time could have improved the forecasting of relevant public health indicators ( i . e . epidemic size , maximal incidence , peak date and epidemic duration ) from a very early point . The method we describe , together with the compilation of past epidemics , improves epidemic forecasting . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"geomorphology",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"chikungunya",
"infection",
"landforms",
"infectious",
"disease",
"epidemiology",
"pathogens",
"topography",
"tropical",
"diseases",
"microbiology",
"geographical",
"locations",
"alphaviruses",
"viruses",
"north",
"america",
"chikungunya",
"virus",
"rna",
"viruses",
"forecasting",
"mathematics",
"statistics",
"(mathematics)",
"neglected",
"tropical",
"diseases",
"caribbean",
"islands",
"research",
"and",
"analysis",
"methods",
"guadeloupe",
"infectious",
"diseases",
"medical",
"microbiology",
"epidemiology",
"mathematical",
"and",
"statistical",
"techniques",
"microbial",
"pathogens",
"people",
"and",
"places",
"flaviviruses",
"viral",
"pathogens",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"physical",
"sciences",
"statistical",
"methods",
"martinique",
"organisms",
"zika",
"virus"
] | 2018 | Improving early epidemiological assessment of emerging Aedes-transmitted epidemics using historical data |
The skin provides an important first line of defence and immunological barrier to invasive pathogens , but immune responses must also be regulated to maintain barrier function and ensure tolerance of skin surface commensal organisms . In schistosomiasis-endemic regions , populations can experience repeated percutaneous exposure to schistosome larvae , however little is known about how repeated exposure to pathogens affects immune regulation in the skin . Here , using a murine model of repeated infection with Schistosoma mansoni larvae , we show that the skin infection site becomes rich in regulatory IL-10 , whilst in its absence , inflammation , neutrophil recruitment , and local lymphocyte proliferation is increased . Whilst CD4+ T cells are the primary cellular source of regulatory IL-10 , they expressed none of the markers conventionally associated with T regulatory ( Treg ) cells ( i . e . FoxP3 , Helios , Nrp1 , CD223 , or CD49b ) . Nevertheless , these IL-10+ CD4+ T cells in the skin from repeatedly infected mice are functionally suppressive as they reduced proliferation of responsive CD4+ T cells from the skin draining lymph node . Moreover , the skin of infected Rag-/- mice had impaired IL-10 production and increased neutrophil recruitment . Finally , we show that the mechanism behind IL-10 production by CD4+ T cells in the skin is due to a combination of an initial ( day 1 ) response specific to skin commensal bacteria , and then over the following days schistosome-specific CD4+ T cell responses , which together contribute towards limiting inflammation and tissue damage following schistosome infection . We propose CD4+ T cells in the skin that do not express markers of conventional T regulatory cell populations have a significant role in immune regulation after repeated pathogen exposure and speculate that these cells may also help to maintain skin barrier function in the context of repeated percutaneous insult by other skin pathogens .
The skin provides an important first line of defence against infectious pathogens which can gain entry via open wounds/abrasions ( e . g . Staphylococcus aureus [1] ) , following injection via insect bites ( e . g . Leishmania sp . protozoa [2] or filarial nematodes [3] ) , or via direct percutaneous penetration ( e . g . soil transmitted hookworms [4] , or the helminth Schistosoma sp [5] ) . As one of the body’s largest tissues , the skin is equipped with a several types of cells with immune function , including myeloid cells [6] , epidermal γδ T cells [7] and innate lymphoid cells [8] , which alongside other types of cell operate in concert against pathogens , but also provide a mechanism of immune regulation to prevent excessive inflammation and ensure tolerance of commensal microorganisms [9–11] . Regulation of the immune response in the skin is particularly important as this organ is host to at least one billion commensal bacteria per square inch [12 , 13] . One type of cell in the skin that has been most often associated with immunomodulation are conventional αβ regulatory CD4+ T cells , such as Tr1 [14] and FoxP3+ cells [15] , which are thought to be important following skin exposure , for example , to Leishmania [16] and Plasmodium [17] . Following percutaneous infection of the skin by schistosome parasites , a balance must be established between providing immune protection for the host whilst preventing excessive tissue damage and promoting wound healing [18–20] . The larval parasite ( cercaria ) gains entry into the skin aided by the release of excretory/secretory ( E/S ) products from the pre-and post-acetabular glands [21] , which have stimulatory , as well as regulatory effects on cells of the host’s innate immune system [22–24] . Indeed , cercarial E/S antigens specifically promote production of regulatory IL-10 by antigen presenting cells [23 , 25] ( Sanin & Mountford , manuscript in preparation ) , and by cultures of whole blood cells obtained from infected individuals from endemic areas for schistosomiasis [26] . IL-10 is often linked to the development of immune regulation following chronic infection with both protozoan , as well as helminth parasites [27] . It has a well-characterized role in limiting liver pathology and mediating resistance to the chronic stage of schistosome disease where eggs released by adult worms act as a major stimulus [28–31] . However , the vast majority of experimental studies of schistosome infection focus upon immune events after a single infection despite the knowledge that for many human residents of schistosome-endemic regions , repeated exposure to cercariae is likely to occur during daily domestic contact with contaminated water sources [32] . A recent study using a murine model of repeated exposures to schistosome cercariae prior to the onset of chronic egg deposition , revealed that the skin infection site becomes rich in Th2-associated IL-4 , but also immune regulatory IL-10 [18] . Moreover , CD4+ T cells in the skin-draining lymph nodes of these repeatedly infected mice became hypo-responsive to schistosome antigens [18] , which was alleviated in the absence of IL-10 [33] . These findings suggest that the suppression of early immune responses to cutaneous infection can be elicited by repeated pathogen exposure and could be mediated by IL-10 at the skin site of infection . In the current study , we show that skin inflammation following repeated schistosome infection is increased in the absence of IL-10 , and that the primary cellular source of IL-10 in the skin was from CD3+ CD4+ T cells . Moreover , IL-10+ dermal CD4+ T cells from repeatedly exposed mice were functionally suppressive as they were able to reduce the proliferation of responsive skin-draining lymph node ( sdLN ) CD4+ T cells from mice exposed to a single dose of cercariae . Finally , the production of IL-10 in the skin , which derived from a combination of schistosome-specific and commensal microbiota-specific CD4+ T cell responses , contributed towards limiting inflammation and tissue damage following infection . Thus , we propose that in the skin , IL-10+ CD4+ T cells that do not express markers of conventional T regulatory cells have an important role in immune regulation after repeated percutaneous exposure to schistosome cercariae . This has relevance for a range of other pathogens which infect their hosts through the skin .
Repeated percutaneous exposure to 4 doses ( 4x ) of infective S . mansoni cercariae causes enhanced production of IL-10 by pinnae skin biopsies compared to those recovered from singly ( 1x ) exposed pinnae ( Fig 1A , p<0 . 001 ) and was accompanied by increased thickening of the skin at the site of infection when compared to skin exposed to a single dose of cercariae ( Fig 1B , p<0 . 001 ) . However , 4x mice that were deficient for IL-10 ( i . e . IL-10-/- ) , had an even greater increase in pinnae thickness compared to 4x wild type ( WT ) animals ( Fig 1B , p<0 . 0001 ) . This was reflected by the recovery of greater numbers of dermal exudate cells ( DEC ) from the biopsies of 4x compared to 1x WT skin , which were even more numerous from 4x IL-10-/- compared to 4x WT pinnae ( Fig 1C , p<0 . 05 ) . Skin biopsies from 4x IL-10-/- mice also released significantly more of the pro-inflammatory cytokine IL-12p40 than 1x IL-10-/- and 4x WT skin ( Fig 1D , both p<0 . 0001 ) . Flow cytometric analysis of myeloid DEC based on F4/80 and MHC-II expression ( defined in S1A Fig ) , combined with mAbs against CD11b and Gr1 , or CD4 and CD3 ( S1B Fig ) , revealed marked differences in the leukocyte populations recovered from the skin of WT and IL-10-/- mice ( Fig 1E and 1F ) . The proportion of DEC that were CD11bhighGr1+ neutrophils increased significantly in 4x IL-10-/- compared to 4x WT skin ( Fig 1E , p<0 . 0001 ) , and when combined with the total numbers of DEC ( Fig 1C ) , the number of neutrophils was 4 . 74-fold greater in 4x compared to 1x IL-10-/- mice . Conversely , fewer SiglecF+F4/80+MHC-II- eosinophils were present in DEC from 4x IL-10-/- compared to 4x WT skin samples ( Fig 1F and S2A Fig , p<0 . 0001 ) . There were no differences between the proportions of MHC-IImid , or MHC-IIhigh cells with putative antigen presenting function between 4x WT and 4x IL-10-/- skin samples ( S2B , S2C and S2D Fig , p>0 . 05 ) , although there were decreased proportions in 4x compared to 1x skin . Thus , the absence of IL-10 in 4x infected mice resulted in increased inflammation evident as exacerbated thickening of the skin , increased recruitment of neutrophils , and greater levels of IL-12p40 . In addition to changes in innate immune cell populations , the proportion of CD3+CD4+ ( F4/80-MHC-II- ) T lymphocytes in the DEC population was increased in 4x WT compared to 1x WT skin samples ( Fig 1G and 1H ) , and they were much more abundant as a proportion of total DEC recovered from 4x IL-10-/- compared to 4x WT skin ( Fig 1H ) . Furthermore , the proliferation of CD3+CD4+ DEC , as measured by the in vivo incorporation of BrdU , was markedly higher in 4x IL-10-/- pinnae compared to all other groups of infected mice ( Fig 1I , p<0 . 001–0 . 0001 ) , illustrating that there is greater proliferation of dermal CD4+ T cells in IL-10 deficient skin . The increase in CD4+ T cell numbers in 4x infected skin compared to a single exposure quantified by flow cytometry was confirmed qualitatively using immunohistochemistry and confocal microscopy . Pinnae from 4x WT mice had increased numbers of CD4+ T cells and CD11b+MHC-II+ cells compared to 1x WT mice ( S3A and S3B Fig ) . Furthermore , increased numbers of CD4+ T cells were apparent in the site of infection of 4x IL-10-/- compared to 4x WT mice ( Fig 1Ji versus 1Jiv , and S3C Fig ) . In addition , whilst CD4+ T cells in 4x WT skin , were distributed fairly evenly throughout the tissue section ( Fig 1Jii and 1Jiii ) , in 4x IL-10-/- pinnae CD4+ T cells were localized in close proximity to CD11b+MHC-II+ cells ( Fig 1Jv and 1Jvi ) . Using IL-10 reporter mice [34] the proportion of IL-10GFP+ DEC recovered from skin exposed to S . mansoni cercariae was found to be significantly increased in 4x compared to 1x mice ( Fig 2A and 2B , p<0 . 0001 ) . Further characterization of IL-10GFP+ DEC ( gating strategy in S1 Fig ) , revealed two IL-10GFP+ populations identified as F4/80+MHC-IIhigh tissue resident macrophages ( R4A gate ) , and CD3+CD4+ T cells ( from F4/80-MHC-II- gate ( R1 ) ) in both 1x and 4x pinnae ( Fig 2C and 2D ) . In naïve skin , IL-10GFP+ DEC were almost exclusively within the F4/80+MHC-IIhigh R4A cell population ( S4A Fig ) . However , one day after infection , the proportion of IL-10 producing cells was equally split between CD4+ T cells and F4/80+MHC-IIhigh macrophages ( S4B Fig ) . By day 4 after infection , CD4+ T cells became the predominant source of IL-10 in both 1x ( Fig 2E ) and 4x ( Fig 2F ) infected pinnae . Importantly , the proportion of IL-10GFP+ cells that were CD4+ T cells was greater in 4x compared to 1x DEC ( Fig 2G , p<0 . 01 ) . Several types of CD4+ T cells can produce IL-10 , including regulatory T cells [35] , but the proportion of FoxP3+ thymic T regulatory cells ( tTreg ) , identified by their co-expression of Helios and Nrp1 in DEC ( Fig 3A left panel ) , was significantly lower in CD3+CD4+ DEC recovered from 4x compared to 1x mice ( Fig 3B , p<0 . 0001 ) , although the absolute number of tTreg remained unchanged between these two groups of mice ( Fig 3C ) . The proportions and numbers of peripheral Treg ( pTreg ) , expressing FoxP3 but not Helios ( Fig 3A middle panel ) , were not significantly different between 1x and 4x DEC ( Fig 3B and 3C , p>0 . 05 ) . Conversely , Type 1 regulatory T ( Tr1 ) cells , which co-express CD223 and CD49b ( Fig 3A , right panel ) , were slightly increased in 4x compared to 1x DEC as a proportion ( Fig 3B , p<0 . 05 ) , but the difference in the total number of Tr1 cells between the infection groups did not reach statistical significance ( Fig 3C , p>0 . 05 ) . Nevertheless , CD4+ T cells that did not fall into any of these three categories of conventional regulatory T cells were the most abundant type of CD4+ cells in both groups of infected skin ( ~70% of all CD4+ DEC , Fig 3B and 3C ) , whilst their proportion and number in 4x pinnae was significantly greater than in 1x skin ( Fig 3B and 3C , p<0 . 0001 ) . Analysis of IL-10GFP+ CD4+ DEC showed that IL-10 producing pTreg were not a prominent source of IL-10 ( <5% of all IL-10GFP+ cells ) in either 1x or 4x infected skin ( Fig 3D ) . Their number also did not differ significantly between the two groups of mice ( Fig 3E ) . Furthermore , tTreg contributed proportionally less IL-10 relative to other CD4+ cell types in 4x skin ( Fig 3D ) , and their number was not significantly different between the two infection groups ( Fig 3E ) . Conversely , the proportion of Tr1 cells in 4x pinnae that were IL-10GFP+ increased ( Fig 3D , p<0 . 05 ) , and the difference in their number between 1x and 4x mice was significantly greater ( Fig 3E ) . However , CD4+ T cells that fell in none of these categories were by far the predominant CD4+ cell population that were IL-10GFP+ in the skin after infection , representing more than 60% of all IL-10GFP+ CD4+ T cells in both 1x and 4x DEC ( Fig 3D ) , as well as increasing in number significantly in 4x DEC ( Fig 3E , p<0 . 0001 ) . To verify the antigen specificity of dermal CD4+ T cells , carboxyfluorescein diacetate succinimidyl ester ( CFDA-SE ) labelled DEC cultures were stimulated with soluble antigen derived from parasite larvae ( i . e . SSAP ) . CD4+ DEC obtained from 1x mice on day 1 after infection exhibited very low levels of proliferation to parasite antigen , which was similar to those in DEC cultured without antigen ( no-antigen controls; Fig 4A left and Fig 4B ) . In contrast , CD4+ DEC from 1x mice recovered on day 4 , as well as those from 4x mice recovered on day 1 and day 4 , all exhibited enhanced proliferation in response to SSAP ( Fig 4A and 4B , p<0 . 0001 ) . The proliferative responses to SSAP of dermal CD4+ T cells recovered on day 4 after infection , from both 1x and 4x mice , were significantly greater than for cells recovered on day 1 ( Fig 4B , p<0 . 0001 ) . Although dermal CD4+ T cells from 1x pinnae recovered on day 1 after infection did not proliferate in response to SSAP , a proportion of them were positive for IL-10 compared to naive mice ( S4B and S5A Figs ) . As it is unlikely that sufficient time will have elapsed for parasite antigen-specific CD4+ T cells to have been primed in the lymph node and recruited to the skin site of infection , it was thought possible that these CD4+ T cells were responding to other foreign antigens . Commensal microbiota resident on the skin surface might gain access to the dermis as the parasite embarks on percutaneous penetration , thus commensals are a possible source of antigen . Indeed , dermal CD4+ T cells from 1x mice recovered on day 1 proliferated in vitro to antigen derived from skin commensal microbes ( SC ) , and the level of proliferation was significantly greater compared to that in response to parasite SSAP antigen ( Fig 4C , p<0 . 0001 ) . In contrast , dermal CD4+ T cells from 1x mice recovered on day 4 after infection proliferated vigorously to parasite SSAP antigen , to a much greater level than in response to SC ( Fig 4C , p<0 . 0001 ) . CD4+ T cell proliferation where no antigen was added is likely to be a result of in situ priming of DEC antigen presenting cells with schistosome antigen already present in the skin of repeatedly exposed mice , as larvae can remain in this tissue for up to 4 days post-exposure . Stimulation of DEC from 1x and 4x infected mice with parasite SSAP antigen resulted in the production of significantly greater quantities of IL-10 compared to unstimulated 1x and 4x DEC , whilst the levels of IL-10 were significantly greater for both groups of mice on day 4 than day 1 ( Fig 4D , p<0 . 0001 ) . High levels of IL-10 were also detected in the supernatants of DEC stimulated with SC antigen , and were at their greatest on day 4 in 1x mice ( Fig 4E , p<0 . 0001 ) . On the other hand , whilst IL-4 was produced by DEC from both 1x and 4x mice in response to SSAP ( Fig 4F ) , detectable levels of IL-4 were only produced by 4x DEC recovered on day 4 in response to SC antigen ( Fig 4G , p<0 . 001 ) . Furthermore , whilst 1x DEC on day 4 produced significant quantities of IFN-γ to SSAP antigen ( Fig 4H ) , only very low levels of this cytokine were detected in DEC supernatants from 1x and 4x mice at both time points in response to SC antigen ( Fig 4I ) . There were significantly more IL-10GFP+ CD4+ T cells in DEC recovered on day 1 after S . mansoni infection than in naïve skin ( S5A Fig ) . However , dermal CD4+ T cells in naïve skin might be exposed to commensal microorganisms during the course of the host’s development [36] . Consequently , although very few CD4+ T cells were recoverable from naïve skin , we show that proliferation of these cells in vitro to SC antigen was comparable to that of dermal CD4+ T cells recovered on day 1 after exposure to schistosome cercariae ( S5B Fig ) . Furthermore , cultures of DEC from naive mice stimulated with SC antigen also produced significantly elevated levels of IL-10 ( S5C Fig ) , but did not produce IL-4 ( S5D Fig ) , and only non-significant quantities of IFN-γ ( S5E Fig ) . DEC from naïve mice did not respond to SSAP . Thus , we demonstrate that a number of CD4+ T cells with specificity to commensal antigens are present in naïve skin and we conclude that infection with schistosome cercariae enhances the exposure of CD4+ T cells to SC antigen resulting in the observed IL-10 production . As the production of IL-10 in our model derived predominantly from dermal CD4+ T cells , the effect of their absence was examined in repeatedly infected Rag2-/- mice . CD3-B220+ B cells and CD3+CD8+ lymphocytes , which are also absent in Rag2-/- mice , were not abundant in S . mansoni infected skin ( S6 Fig ) . As CD3+CD8+ lymphocytes were significantly reduced in number in 4x DEC ( p<0 . 001 ) , and CD3-B220+ B cell numbers remained unchanged , we conclude that differences observed in Rag2-/- mice would be predominantly due to the absence of CD4+ T cells . Indeed , the absence of CD4+ T cells in 4x Rag2-/- mice resulted in an increase in the proportion of CD11b+Gr1+ neutrophils compared to 4x WT controls ( Fig 5A , p<0 . 05 ) , whereas there was a significant reduction in the proportion of SiglecF+F4/80+MHC-II- eosinophils ( Fig 5B , p<0 . 001 ) , reminiscent of the findings for infected IL-10-/- mice ( Fig 1E and 1F ) . The production of IL-10 ( Fig 5C , p<0 . 05 ) and IL-4 ( Fig 5D , p<0 . 05 ) by cultured skin biopsies was also impaired in 4x Rag2-/- mice . Collectively , our data are compatible with the notion that the dermal CD4+ T cell population is the main source of IL-10 in S . mansoni infected skin and that other IL-10+ myeloid cells ( such as those defined in Fig 2 ) are not sufficient to recapitulate the immune response observed in 4x WT mice . Since the presence of IL-10 in the skin is able to regulate the proliferation of CD4+ T cells in situ in the skin site of infection ( Fig 1I ) , we sought to determine whether dermal IL-10 producing cells from 4x mice were able to inhibit in vitro proliferation of CD4+ T cells recovered from sdLN of mice exposed to 1x infection . As expected , sdLN CD4+ cells from 1x mice were responsive and proliferated in vitro to parasite SSAP antigen ( Fig 6A , left ) . The addition of dermal CD4+ T cells from 4x infected IL-10-/- mice did not significantly affect the ability of 1x sdLN CD4+ cells to proliferate ( Fig 6B , p>0 . 05 ) . However , addition of IL-10 competent dermal CD4+ T cells from 4x WT mice significantly inhibited proliferation of 1x sdLN CD4+ cells ( Fig 6B , p<0 . 05 ) . Moreover , the difference between adding dermal CD4+ T cells from 4x WT and 4x IL-10-/- mice was statistically significant ( Fig 6B , p<0 . 001 ) . Furthermore , IL-10GFP+ dermal CD4+ T cells sorted from the DEC of 4x infected mice significantly inhibited proliferation of 1x sdLN CD4+ cells in response to parasite SSAP antigen ( Fig 6C and 6D , p<0 . 05 ) , whereas GFPNeg dermal CD4+ T cells from the same 4x mice did not significantly alter the proliferation of 1x sdLN CD4+ cells ( Fig 6C and 6D , p>0 . 05 ) . Combined , these data demonstrate that IL-10 production from dermal CD4+ T cells is responsible for regulating the responsiveness of sdLN CD4+ T cells .
The role of IL-10 in maintaining immune homeostasis and resolving inflammation is critical for host survival and function [16 , 37–39] , a phenomenon that has been well studied in the pathogenesis of chronic schistosomiasis [28–31] but is not understood , or documented , at the early stage of percutaneous schistosome infection . Here , we show that repeated exposures of the skin to infective S . mansoni cercariae results in an early increase in IL-10 production by CD4+ T cells at the skin site of infection . The production of IL-10 by these cells occurs prior to the late stage production of IL-10 conventionally associated with chronic schistosome infection after the onset of egg deposition [30 , 40] . Single infection with S . mansoni cercariae leads to transient IL-10 production in the skin , which returns to naïve levels after 14 days [41] , whilst repeated infection leads to sustained production of IL-10 between days 1–8 after exposure [18] . In our model , early IL-10 production by dermal CD4+ T cells was important in limiting the extent of inflammation at the site of infection by reducing skin thickening and neutrophil recruitment . Moreover , dermal CD4+ T cells were able to limit local proliferation of CD4+ T cells in the dermis and suppress the proliferation of normally responsive CD4+ T cells from proximal sdLN of singly infected animals . In the current study , CD4+ T cells were the main producers of IL-10 after repeated exposure to S . mansoni cercariae supporting the findings of others , who report that CD4+ T cells are important contributors to the production of IL-10 in a range of disease settings [16 , 42–46] , as well as being an important target of IL-10 [47] . In the mesenteric lymph nodes and the spleen during chronic S . mansoni infection , CD25+CD4+ and CD25-CD4+ T cells were the main producers of IL-10 [48] , and it is well known that IL-10 is important in limiting increased granuloma formation around eggs deposited in the liver , leading to decreased host survival [31 , 49] . However , here we demonstrate that production of IL-10 by dermal CD4+ T cells occurs very early during the initial stages of S . mansoni infection , well in advance of egg deposition , which is a hallmark of chronic/long term infection . We show that early IL-10 was instrumental in limiting inflammation after primary but particularly following repeated exposures to infectious cercariae . Local proliferation of CD4+ T cells , plus the recruitment of neutrophils were regulated by this early production of IL-10 in the skin . This suggests that during natural infection , where repeated exposures to infective cercariae are likely to be common for residents of schistosome-endemic areas [32] , IL-10 could play an important role in limiting leukocyte mediated tissue damage associated with migration of larvae through the skin . Indeed , whole blood cultures of infected individuals in endemic regions produced enhanced levels of IL-10 in response to cercarial E/S antigens [26] , which supports the hypothesis that this cytokine is triggered by the early stage of the parasite . Multiple subsets of T helper cells can make IL-10 [16 , 42–46] , including T helper ( Th ) 2 , thymic Nrp1+ Helios+ Treg ( tTreg ) , peripheral FoxP3+ Helios- Treg ( pTreg ) , and Tr1 cells [27 , 50 , 51] , and previous studies in the skin have highlighted a role of regulatory T cells in the prevention of excessive inflammation during exposure to protozoan parasites [15–17 , 52] . However in our studies , the proportions of both tTreg and pTreg decreased in 4x infected mice , and as they were a scarce source of IL-10 , there was limited evidence for these regulatory CD4+ T cell subtypes being responsible for early IL-10 in the skin of mice repeatedly exposed to larvae of the Schistosoma helminth . Although CD223+ CD49b+ Tr1 cells , which are thought to exert their function through the production of IL-10 [14 , 50] , were slightly increased as a proportion of all CD4+ T cells in 4x mice , and as a proportion and number of IL-10+ CD4+ T cells in the skin , most of the IL-10+ CD4+ T cells in the skin expressed none of the markers associated with conventional ‘regulatory’ phenotypes despite being functionally suppressive . Nevertheless , as several of these functionally suppressive IL-10+ CD4+ T cells co-expressed Nrp1 and CD49b , both of which are up-regulated in activated T cells [53 , 54] , our findings do not rule out the possibility that they arise from conventional regulatory T cells which have lost expression of FoxP3 , Helios , or CD223 ( LAG3 ) , as reported by others [55 , 56] . A major difference that distinguishes our work from those cited above [15–17 , 52] is that they were based upon the study of immune responses to a single pathogen exposure leading to persistent/chronic infection . In contrast , in our study , repeated exposures to cercariae led to regulation being evident early after infection , thus emphasizing the early role of antigen-specific CD4+ T cells in mediating immune regulation in sites of initial infection , such as the skin . The functionally suppressive CD4+ T cells had a putative Th2 bias , as they secreted IL-4 , but not IFN-γ , which is in line with our previous studies showing the production of elevated levels of IL-4 , but not Th1 associated cytokines in skin exposed to multiple doses rather than a single dose of cercariae [18] . CD4+ cells were confirmed as the source of IL-4 since this cytokine was not detected in T cell deficient skin ( i . e . Rag2-/- ) . The role of Th2 cells in limiting skin inflammation has been suggested during disorders such as atopic dermatitis , where innate immune cells are regulated by cytokines from Th2 cells [13 , 57] . However , recent literature highlights the possibility that CD4+ T cells could be tissue resident memory cells ( TRM ) as identified primarily in pulmonary and gastrointestinal mucosal tissue sites [58 , 59] . Whilst CD69 and CD103 expressing CD4+ cells have also been reported in skin [9 , 60–62] , it is not clear whether they are circulating , or are tissue resident , and we did not observe dermal CD69+ CD4+ T , or CD103+ CD4+ T cells in our model . Finally , although CD4+ TRM cells can produce IFNγ/IL-17 [58] , and IL-4/IL-13 following helminth infection [63] , their production of IL-10 has not previously been reported . Consequently , further investigation into the possible definition of IL-10+ CD4+ population in the dermis is warranted . Myeloid cells such as F4/80+MHC-IIhigh tissue resident macrophages were also a source of IL-10 in the skin after S . mansoni infection , most likely in response to cercarial E/S products [25] ( Sanin & Mountford , manuscript in preparation ) . We showed previously that after repeated infection , M2-like dermal APCs conditioned within the skin infection site , which is rich in IL-4 and IL-10 , were associated with decreased CD4+ T cell responsiveness in the skin draining lymph node [18] . However , our current findings show that proliferation of dermal CD4+ T lymphocytes is not impaired in repeatedly infected skin and that IL-10 production is predominantly from CD4+ T cells in the skin rather than myeloid cells . This accords with recent findings that IL-10 is the main driver of CD4+ T cell hypo-responsiveness in the lymph node [33] . In the present study , the role of F4/80+MHC-IIhigh macrophage derived IL-10 might be important for the initial polarization of dermal CD4+ T cells towards their subsequent production of IL-10 , as shown in another recent study of helminth infection [64] . Indeed , IL-10 is able to enhance its own expression through the activation of STAT3 [65] . The clusters of MHC-II+ cells and CD4+ T cells observed in schistosome-infected skin may represent how myeloid cells activate CD4+ T cells after S . mansoni infection , as suggested in other infection models [66 , 67] . IL-10 production in T cells can be regulated by several cytokines ( e . g . IL-4 , IL-6 , IL-12 , and IL-27 ) , as well as multiple transcription factors ( i . e . STAT1 , STAT3 , STAT4 , STAT6 , GATA3 , and c-MAF ) [68] . In particular , IL-27 has been reported to drive IL-10 production by CD4+ T cells in dermal lesions caused by infection with Leishmania parasites [69] , although IL-27 was not detected in the culture supernatants of skin biopsies in our schistosome infection model . However , full exploration of these cytokines and transcription factors was beyond the scope of the current study . In the absence of T lymphocytes ( i . e . Rag2-/- ) , very low levels of IL-10 were noted and the dermal immune response was equivalent to that in IL-10-/- mice , supporting our conclusion that myeloid and/or CD4negative lymphoid cells are not a major source of IL-10 . Other lymphocytes , which are absent in Rag2-/- skin , such as B cells and CD8+ T cells were rare in infected skin ( <1000 cells ) and did not expand in number , or proportion after repeated infection , nor did they produce IL-10 . Therefore , IL-10 derived from dermal CD4+ T cells , rather than myeloid cells [18] , appears to be responsible for inducing CD4+ lymphocyte hypo-responsiveness in the sdLN , as well as conditioning the dermal immune environment . Although functionally suppressive IL-10+ CD4+ T cells in the dermis had specificity for schistosome antigen , a population of dermal CD4+ T cells produced IL-10 in response to antigens from commensal microbiota . Indeed , bacterial commensals in barrier tissues such as the skin can influence the immune response [12 , 13 , 70] , and can exacerbate immune pathology by altering the balance between regulatory and effector T cells by triggering IL-17 and IFN-γ , although a role for IL-10 has not been established [36 , 71 , 72] . Here we show that CD4+ T cells in the skin produce IL-10 in a response that was initially directed against commensal microbiota . We speculate that skin commensals gain access to the dermis as the parasites invade the skin and therefore stimulate resident T cells with specificity for antigens from commensal microbiota as early as day 1 after schistosome infection . These early responding T cells could include Tr1 cells , which are a known source of IL-10 [14 , 50] , especially as their antigen specificity can be to commensal microbiota [52] . The early production of IL-10 triggered by commensals as a result of penetrating S . mansoni cercariae , therefore could be a strategy adopted by the parasite to limit anti-parasite immune responses from the host . This might be particularly relevant in the context of the enhanced neutrophilia observed here in the absence of IL-10 . The use of germ-free animals could provide direct evidence of the role of skin commensals in S . mansoni infection , as topical antibiotic treatment of skin to eliminate skin microbiota is likely to have an adverse effect on cercarial penetration , as seen with the use of soap [73] . However , the impact of commensal microorganisms on the dermal immune response to Schistosoma infection in humans in the face of concurrent antibiotic treatment requires further investigation . In summary , we demonstrate that repeated exposures of the skin to infective S . mansoni cercariae leads to an early increase in IL-10 production by S . mansoni specific CD4+ T cells in the dermis , with a putative Th2 bias , at the site of infection . This is accompanied by a previously unreported bystander response to commensal microbiota that gain access to the dermis as cercariae penetrate the skin . IL-10 production by these dermal CD4+ T cells is important in limiting the extent of skin inflammation , leukocyte proliferation and recruitment . Critically , these dermal CD4+ T cells are able to suppress the proliferation of CD4+ T cells from the draining lymph nodes which could explain the development of hypo-responsiveness should they migrate in vivo to proximal lymph nodes . Collectively our findings demonstrate the importance of early IL-10 production by functionally suppressive CD4+ T cells in the skin in response to S . mansoni cercariae and highlights a possible role of these cells in maintaining host fitness in populations that inhabit areas endemic for schistosomiasis , and other helminth larvae that penetrate via the skin [74 , 75] . Finally , our data show that functionally suppressive CD4+ T helper cells , that are not conventional regulatory CD4+ T cells , are important as modulatory cells in the skin after repeated exposure to pathogens . We suggest that the role of IL-10 in controlling early immune responses to schistosomes , may act as a prelude to the subsequent development of IL-10 mediated immune regulation conventionally associated with chronic helminth infections .
Female mice aged between 6–10 weeks were used for all experiments carried out in accordance with the UK Animal’s Scientific Procedures Act 1986 and with approval of the University of York Ethics Committee ( PPL #60/4340 ) . Wild type C57BL/6 ( WT ) and IL-10 deficient ( IL-10-/- ) mice [76] , as well as transgenic mice lacking RAG2 ( Rag2-/- ) , or IL-10 reporter knockin ( tiger ) animals ( IL-10+/GFP ) [34] were bred and housed at the University of York . Mice were percutaneously exposed via each pinna to four doses ( 4x ) of 150 S . mansoni cercariae at weekly intervals from day 0 to day 21 as previously described ( repeated infection ) [18 , 77] . Age and sex-matched cohorts were exposed in parallel to a single dose ( 1x ) of 150 cercariae on day 21 . Using this infection protocol via the pinna , a 50% penetration rate is observed [77] amounting to ~75 cercariae per pinna at each time-point . Inflammation of pinnae was measured using a dial gauge micrometer ( Mitutoyo , Japan ) . In most experiments , pinnae were harvested 4 days after the last infection , or in selected experiments obtained on day one . Auricular lymph nodes draining the skin site of infection ( sdLN ) were also harvested in specific experiments on day 4 post-final infection . Pinnae from naïve , and infected mice ( on day 1 or 4 post-final exposure to cercariae ) were removed , briefly sterilized with ethanol , air-dried and split along the central cartilage into two halves to obtain the population of dermal exudate cells ( DEC ) as described previously [18 , 78] . Split pinnae were floated overnight at 37°C 5% CO2 in the absence of added antigen on RPMI-1640 media ( Gibco , Paisley , UK ) containing 10% heat inactivated FCS ( Biosera , Uckfield , UK ) , 2mM L-Glutamine solution , 1% Pen/Strep ( both Gibco ) , and 50μM 2-mercaptoethanol ( Sigma-Aldrich , Gillingham , UK ) ( complete RPMI ) in non-adherent 24 well tissue culture plates ( Greiner Labortechnik , Frickenhausen , Germany ) . Floating pinnae tissues were removed and the remaining culture supernatants containing DEC spun at 1000g for 7min at 4°C before being re-suspended in complete RPMI and live cells enumerated using a hemocytometer . Cell-free culture supernatants were recovered and stored at -20°C before being analyzed by cytokine-specific ELISA . In order to stimulate DEC for an antigen-specific recall response , soluble schistosomula antigen preparation ( SSAP ) was prepared from in vitro cultured mechanically-transformed larvae as described previously [79] . Skin commensal antigen ( SC ) was prepared by culturing skin swabs taken from female WT C57BL/6 mice overnight in liquid broth medium at 37°C . Recovered microbes were washed in sterile PBS , sonicated at full power in PBS for 5 min ( 30s on / 30s off ) , and inactivated for 0 . 5h under UV light . DEC were then cultured at 5x105 cells/ml in complete RPMI in the presence , or absence , of 50 μg/ml SSAP , or 5 μg/ml SC , for 96 hours at 37°C . Proliferation was measured after labelling DEC with 3 μM CFDA-SE ( Invitrogen , Paisley , UK ) [18] . DEC were subsequently labelled with specific monoclonal antibodies ( mAb ) detailed below and proliferation determined by flow cytometry according to the decrease in the fluorescence of CFDA-SE . DEC were incubated with 1 μg anti-CD16/32 mAb ( eBioscience , Hatfield , UK ) in goat-serum ( Sigma-Aldrich ) to block non-specific uptake of antibodies and then subsequently labelled with LIVE/DEAD Fixable Aqua Dead Cell Stain ( Life Technologies , Paisley , UK ) according to the manufacturer’s instructions , plus the following mAbs conjugated to various fluorescent labels: anti-CD4 ( clone RM4-5 ) , anti-CD3 ( clone 17A2 ) , anti-MHC-II ( IA-IE ) ( clone M5/114 ) , anti-Nrp1 ( clone 3DS304M ) , anti-CD11b ( clone M1/70 ) , anti-F4/80 ( clone BM8 ) , anti-Gr1 ( clone RB6-8C5 ) , anti-SiglecF ( clone eBio440c ) , anti-CD45 ( clone 2D1 ) ( all eBioscience ) , anti-CD49b ( clone R1-2 ) ( BioLegend , London , UK ) and anti-CD223 ( LAG3 ) ( clone C9B7W ) ( BD Bioscience , Oxford , UK ) . For intracellular staining , cells were washed with and fixed with 2% paraformaldehyde for 1 hour at 4°C before being washed and incubated for 1 hour in 1x permeabilization buffer ( eBioscience ) for anti-FoxP3 ( clone NRRF-30 ) and anti-Helios ( clone 22F6 ) ( eBioscience ) . All flow cytometry was acquired using the Cyan ADP analyser ( DakoCytomation , Stockport , UK ) , or BD LSR Fortessa analyser ( BD Biosciences , Oxford , UK ) . Data was analysed using FlowJo software v7 . 6 . 5 ( Tree Star , Inc , Ashland , Oregon , USA ) . Detection of IL-10 production by different cell types in DEC was achieved using IL-10+/GFP mice . WT and IL-10+/GFP mice were infected and pinnae harvested as described above . Split pinnae were incubated with complete RPMI for 12 hours prior to the addition of 1x Brefeldin A ( eBioscience ) following the manufacturer’s instructions . After a further 8 hours , DEC were prepared for flow cytometric analysis , by washing once with PBS . DEC obtained from 4x infected WT , IL-10+/GFP and IL-10-/- mice were labelled with mAb against CD4 , CD3 , CD45 and MHC-II as described above . Dermal CD45+CD3+CD4+ T cells were recovered following FACS ( MoFlo Astrios , Beckman Coulter , London , UK ) from WT and IL-10-/- mice . For IL-10+/GFP mice , dermal CD45+CD3+CD4+ IL-10GFP+ T cells and CD45+CD3+CD4+ IL-10GFP- T cells were obtained . CFDA-SE labelled single cell suspensions from the sdLN were co-cultured at 2 . 5x105 cell/ml in complete RPMI with , or without , 2x103 sorted dermal T cells in the presence , or absence , of 50 μg/ml SSAP for 96 hours at 37°C . Antigen-specific proliferation of sdLN CD4+ cells was measured by a decrease in levels of CFDA-SE after 72 hours as described above compared to cells stimulated the absence of antigen . Culture supernatants were collected from overnight skin biopsies , or from sdLN cell cultures after 96 hours , for cytokine analysis as previously described [80] . IL-4 , IL-10 and IL-12p40 were quantified by DuoSet ELISA ( R&D Systems , Abingdon , UK ) , whilst IFN-γ was measured using specific capture and detection antibodies ( BD Pharmingen , Oxford , UK ) [77] . Mice received 1 mg BrdU ( Sigma-Aldrich ) i . p . daily for the final 4 days before harvest of pinnae in order to determine in vivo cell proliferation . DEC were then recovered , and blocked with 1 μg anti-CD16/32 mAb in goat-serum . Subsequently , DEC were labelled for surface expression of CD3 , CD4 , MHC-II , F4/80 and CD11b ( all eBioscience ) in PBS supplemented with 1% FCS . Cells were then washed and incubated in 1x Fixation/Permeabilisation buffer ( eBioscience ) for 1 hour at 4°C before being washed and incubated for 1 hour at 37°C in 100 μg DNase ( Sigma-Aldrich ) . Finally , cells were labelled for 45 minutes at room temperature with anti-BrdU APC mAb , or rat IgG1 APC ( eBioscience ) , in 1x Permeabilisation buffer as per the manufacturer’s protocol . Freshly recovered pinnae were fixed in PBS/4% paraformaldehyde on ice for 30 min then transferred to PBS/15% sucrose for a further 1h on ice . Fixed pinnae were then embedded in OCT medium ( Sakura Finetek , Netherlands ) , and frozen at -80°C . Cryosections ( 6μm ) obtained from frozen pinnae were simultaneously blocked and permeabilised in PBS supplemented with 5% goat serum ( Sigma-Aldrich ) and 0 . 05% saponin ( Sigma-Aldrich ) for 30min at room temperature . Sections were incubated in PBS/5% goat serum/0 . 05% saponin for 1h at room temperature with mAbs directly conjugated to various fluorescent labels: anti-CD4 ( clone RM4-5 ) , anti-MHC-II ( IA-IE ) ( clone M5/114 ) and anti-CD11b ( clone M1/70 ) , or suitable isotype controls ( all eBioscience ) and then washed in PBS/0 . 05% saponin ( 3x , 5min each ) . Finally , sections were counter-stained with 2μg/ml DAPI ( Life Technologies ) for 5min and rinsed with distilled water . Slides were mounted in Prolong Gold AntiFade reagent ( Life Technologies ) prior to analysis using a Zeiss 710 inverse confocal microscope ( Carl Zeiss , Cambridge , UK ) . All images were analysed using identical acquisition settings in Zeiss ZEN software . Image handling ( including contrast adjustment ) was performed on ImageJ ( National Institute of Health ) . Analysis of variance ( ANOVA ) and then multiple comparisons tests ( Bonferroni’s , Tukey’s , Sidak’s and Dunnett’s ) were performed to establish significant differences between the groups ( * = p<0 . 05 , ** = p<0 . 01; *** = p<0 . 001 , **** = p<0 . 0001 ) using the software package GraphPad Prism ( GraphPad Software , Inc . , La Jolla , California , USA ) . Similarly , unpaired two-tailed T tests were performed in selected experiments to compare experimental groups . Error bars represent the standard error of the mean ( SEM ) , based on technical replicates for in vitro experiments , or biological replicates for in vivo experiments . | The skin is a major barrier protecting the host from pathogen infection , but is also a site for immune regulation . Using a murine model of repeated percutaneous exposure to infectious Schistosoma mansoni cercariae , we show that , in the skin , CD4+ T cells that do not express markers of conventional regulatory T cells are the main early source of immunoregulatory IL-10 and are functionally suppressive of adaptive immune responses . We demonstrate that the production of regulatory IL-10 in the skin is greatly enhanced after repeated schistosome infection compared to levels present after a single infection and that it limits both neutrophil recruitment and local CD4+ T cell proliferation , thereby preventing excessive inflammation and tissue damage . Initially ( day 1 ) , IL-10 producing CD4+ T cells are reactive towards skin commensal bacteria , although over succeeding days they progressively become specific for schistosome antigens . Consequently , our findings highlight a role for early IL-10 produced by dermal CD4+ T cells to mediate immune regulation in advance of later stage chronic infection conventionally associated with the presence of IL-10 . Our work provides a mechanistic insight into the triggers of early IL-10 production at barrier sites like the skin , and suggests how tolerance and pathogen clearance might be co-regulated early after exposure to infectious agents . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Helminth Infection and Commensal Microbiota Drive Early IL-10 Production in the Skin by CD4+ T Cells That Are Functionally Suppressive |
Mammalian BET proteins comprise a family of bromodomain-containing epigenetic regulators with complex functions in chromatin organization and gene regulation . We identified the sole member of the BET protein family in Drosophila , Fs ( 1 ) h , as an inhibitor of the stress responsive transcription factor CncC , the fly ortholog of Nrf2 . Fs ( 1 ) h physically interacts with CncC in a manner that requires the function of its bromodomains and the acetylation of CncC . Treatment of cultured Drosophila cells or adult flies with fs ( 1 ) h RNAi or with the BET protein inhibitor JQ1 de-represses CncC transcriptional activity and engages protective gene expression programs . The mechanism by which Fs ( 1 ) h inhibits CncC function is distinct from the canonical mechanism that stimulates Nrf2 function by abrogating Keap1-dependent proteasomal degradation . Consistent with the independent modes of CncC regulation by Keap1 and Fs ( 1 ) h , combinations of drugs that can specifically target these pathways cause a strong synergistic and specific activation of protective CncC- dependent gene expression and boosts oxidative stress resistance . This synergism might be exploitable for the design of combinatorial therapies to target diseases associated with oxidative stress or inflammation .
Nrf2 transcription factors are critically important for the health , homeostasis and longevity of multicellular organisms [1–3] . When cells are confronted with oxidative or chemical stress , Nrf2 stimulates the expression of gene products that protect cell integrity including antioxidants , redox regulators , phase II detoxification enzymes , and factors that maintain proteostasis . The Kelch domain protein Keap1 has been identified as a key mediator of the acute activation of Nrf2 in response to oxidative stress [4 , 5] . In unstressed cells Keap1 assembles a Cullin 3-containing ubiquitin ligase complex that targets Nrf2 for proteolysis by associating with its NH2-terminally located NEH2 domain . This Keap1-mediated degradation of Nrf2 is relieved upon stress exposure , so that the transcription factor can accumulate in the nucleus and bind to so-called ARE ( antioxidant response element ) sequences in target gene promoters . Over the last few years Nrf2 has been implicated in a range biological processes in addition to stress responses . Examples include the regulation of energy metabolism [1 , 6] , stem cell maintenance [7] and aging [8 , 9] . These functions are probably regulated by signals other than toxic insults and presumably require a different transcriptional response , in terms of kinetics and target gene profile . In order to explore how this expanded range of Nrf2 functions might be regulated we conducted a large-scale screen for gene products that are involved in Nrf2 target gene activation . The experimental model used in our studies is Drosophila melanogaster . The fruit fly has an Nrf2 signaling system , which resembles that of mammals [3 , 10] . The Drosophila ortholog of Nrf2 , CncC is encoded by a long splice product of the cap’n’collar gene [8] . The conservation of the Nrf2 pathway genes and its powerful genetic tools make Drosophila an excellent model to study this important signaling system . The Drosophila CncC signal transduction pathway can , like its mammalian counterpart , mediate transcriptional responses to various types of chemical or oxidative insults and protect the organism from ensuing damage . CncC can also be activated by dietary dosing with cancer chemo-preventive agents such as oltipraz and sulforaphane [8 , 11] . These drugs cause Nrf2 activation without harmful stress and negligible side effects to exposed cells or organisms . Animal experiments have shown these compounds to protect against chemical carcinogens in an Nrf2-dependent manner [12] . Oltipraz and similar drugs exert their effect by interfering with the inhibitory function of Keap1 [13–15] . In a high throughput RNAi screen using Drosophila S2 cells [16] we found the gene fs ( 1 ) h to encode a negative regulator of the Drosophila Nrf2 homolog , CncC . Multiple independent double stranded RNAs that target fs ( 1 ) h mRNA , caused a significant and specific increase in the activity of an ARE luciferase reporter gene [11] , identifying the fs ( 1 ) h gene as a potential inhibitor of Nrf2 function . The product of the fs ( 1 ) h gene , Fs ( 1 ) h , short for female sterile ( on the first chromosome ) homeotic , is counted as a member of the heterogeneous group of Trithorax proteins which generally function as epigenetic regulators [17 , 18] . Fs ( 1 ) h is the sole member of the BET protein family in Drosophila [19 , 20] . BET proteins are characterized by the presence of two bromodomains , adjacent to a so-called extra terminal , or ET , domain [21] . Through the bromodomains , BET proteins specifically bind to polypeptides carrying acetylated lysine residues , including acetyl-histones [22] . Mammalian BET proteins , notably BRD4 , have been implicated in the regulation of gene expression . They are known to bind to chromatin and to interact with components of the transcriptional machinery such as P-TEF B and RNA polymerase II [23 , 24] . In addition , the activities of specific transcription factors such as NF-κB and Twist can be regulated by direct interaction with BRD4 [25–27] . Similarly , functional experiments and genome-wide ChIP mapping studies suggest that the Drosophila BET protein Fs ( 1 ) h functions in the regulation of gene activity [18] . Fs ( 1 ) h gene products have been found associated with transcription control region and genomic insulator elements [28 , 29] . Our experiments show that Fs ( 1 ) h can physically interact with CncC to inhibit its transcriptional function . This mechanism is independent of Keap1-mediated Nrf2 regulation .
A previously described high throughput RNAi screen in Drosophila S2 cells had identified Fs ( 1 ) h as a possible negative regulator of CncC target gene activity [16] . The function of Fs ( 1 ) h as a CncC inhibitor , as suggested by this screen was confirmed by performing transient transfection assays in Drosophila S2 cells . These experiments demonstrated that knock down of Fs ( 1 ) h caused an increase in ARE reporter gene activity of a similar magnitude as seen upon knock-down of the canonical CncC repressor Keap1 ( Fig 1B ) . When , in addition to Keap1 or Fs ( 1 ) h , CncC was knocked down , the activation of the ARE reporter was significantly reduced ( Fig 1B ) . In addition , over-expression of Keap1 to specifically inhibit Nrf2 signaling by limiting its nuclear accumulation , abrogated the induction of the ARE-luciferase reporter in response to Fs ( 1 ) h knock down ( S5A Fig ) . Taken together , these indicate that the stimulatory effect of Fs ( 1 ) h on ARE reporter is dependent on CncC function . Quantitative mRNA measurements in vivo by RT-qPCR supported this conclusion . Flies in which Fs ( 1 ) h was knocked down by dsRNA expression under the control of the tub-GS-Gal4 driver showed that endogenous CncC target genes ( gcl-C , gstD1 , keap1 ) were up-regulated in fs ( 1 ) h knock down conditions ( Fig 1C ) . Like some mammalian BET protein genes , the Drosophila fs ( 1 ) h locus yields alternatively spliced transcripts that give rise to two different polypeptides , one of 120 and one of 210 kDa molecular mass ( Fig 1A ) [18 , 30] . In the following we will refer to the short and the long isoform as Fs ( 1 ) h-S and Fs ( 1 ) h-L , respectively . Both alternative splice products contain the two bromodomains and the ET domain . The unique peptide sequences that extend the Fs ( 1 ) h-L isoform comprise a C-terminal motif ( CTM ) ( Fig 1A ) . Genome-wide ChIP experiments have shown that the Fs ( 1 ) h-S and L differ substantially in their genomic binding patterns and presumably in their function [20 , 28] . To test whether the two splice forms might also differ in their effect on CncC-regulated transcription , we designed dsRNAs that would specifically target only one or the other splice variant ( see Materials and Methods ) . Selective knock down of the long isoform Fs ( 1 ) h-L induced ARE reporter activity . However , knock down of the short isoform , Fs ( 1 ) h-S failed to do so ( Fig 1D ) . We conclude that the repression of CncC activity is a specific function of Fs ( 1 ) h-L . Western blot experiments confirmed the efficient and selective knock down of individual isoforms by the respective dsRNAs ( S1A Fig ) . The inhibitory function of Fs ( 1 ) h on CncC reporter gene activity can also be observed in vivo . We conducted experiments with Drosophila stocks carrying an ARE-GFP reporter gene in which GFP expression is controlled by four consensus ARE sequences ( ARE-GFP , [11] ) . A UAS fs ( 1 ) h-RNAi construct was expressed under the control of the RU486-inducible tubulin-GS-Gal4 driver to ubiquitously knock down endogenous fs ( 1 ) h transcripts in adult flies . This resulted in robust activation of the ARE-GFP reporter transgene in the animals , recapitulating the effect seen in S2 cells . To rule out that the activation of ARE reporter activity was the consequence of an off-target effect , we conducted this experiments with two independent RNAi expression lines and saw similar results ( Figs 2A , S1B and S2B ) . To examine the in vivo effect of fs ( 1 ) h loss-of-function on the ARE-GFP reporter at the cell level , we conducted knock down experiments in groups of cells using actin-flipout-Gal4 , a driver that can be clonally activated by a short heat-shock induced pulse of flp recombinase expression . Clones of RNAi-expressing cells can be distinguished by the expression of an RFP marker . The resulting RFP-labeled fs ( 1 ) hRNAi-expressing cells showed increased ARE-GFP reporter activity , consistent with the presumed function of Fs ( 1 ) h as a CncC inhibitor . Fig 2B shows such a clone in the crop , a part of the foregut that , together with the anterior midgut , functions as the stomach in Drosophila [31] . We found that the epithelial cells of the crop have low basal and high inducible ARE activity . The stimulatory effect of fs ( 1 ) h knock down was restricted to the cells of the clone , demonstrating that the regulatory function of the protein acts by a cell-autonomous mechanism . The results of the loss-of-function experiments described above suggested that Fs ( 1 ) h can suppress gene activation by CncC . However , an alternative interpretation would be that the observed increase of CncC activity might be an indirect consequence of possible stress or damage in fs ( 1 ) h loss-of-function conditions . To rule out the latter mechanism , and to demonstrate the repressive function of Fs ( 1 ) h directly , we asked whether its over-expression could decrease CncC target gene activity in adult flies . Fs ( 1 ) h over-expression was achieved by two alternative strategies . First , we used an EP-line ( P ( EP ) fs ( 1 ) h[EP439] ) , a fly stock in which a Gal4 responsive enhancer was integrated in the 5’ genomic region of the fs ( 1 ) h gene . If combined with the ubiquitously expressed tub-GS-Gal4 driver , expression of endogenous fs ( 1 ) h mRNA was stimulated when flies were exposed to dietary RU486 ( S1B Fig ) . This resulted in a marked reduction of ARE reporter activity ( Fig 2C ) . Second , we generated a transgenic fly line that expresses a cDNA encoding Fs ( 1 ) h-L , the isoform that showed CncC repressor activity in S2 cell-based studies . Inducible expression of the Fs ( 1 ) -L under the control of tub-GS-Gal4 caused a clear reduction of ARE-GFP reporter activity ( S2A Fig ) . In addition , over-expression of Fs ( 1 ) h-L in RFP-marked hsFlp-induced clones in the ejaculatory bulb of adult Drosophila , a tissue with a high basal level of ARE reporter activity , had a negative effect , confirming that Fs ( 1 ) h represses CncC in a cell-autonomous fashion ( Fig 2D ) . Thus , Fs ( 1 ) h functions as a repressor of CncC’s transcriptional activity both in cell culture and in adult flies . The finding that Fs ( 1 ) h acts as a repressor of CncC’s transcriptional output , predicted that it should modulate the biological functions of the Drosophila Nrf2 ortholog . Nrf2-induced gene expression programs can protect organisms against acute oxidative stress . Experimentally such a stress can be generated by dietary exposure to diethyl maleate ( DEM ) , a glutathione-depleting agent . We have demonstrated previously that genetically increasing CncC activity can boost the resistance of flies against oxidative insults like DEM exposure [8] . To test if Fs ( 1 ) h might similarly affect oxidative stress resistance , we measured the DEM sensitivity of flies , in which the activity of Fs ( 1 ) h was either suppressed or increased . For loss-of-function experiments , fs ( 1 ) h mRNA was knocked down in cohorts of young adults flies for 4 days by RU486-induced expression of a corresponding RNAi construct . Subsequently , the animals were transferred to vials containing filter paper laced with a lethal concentration of DEM and the time course of survival was recorded . Cohorts in which fs ( 1 ) h was depleted displayed a significantly increased survival time as compared to controls ( Figs 2E and S3C ) . For gain-of-function experiments Fs ( 1 ) h was over-expressed by feeding EP-fs ( 1 ) h; tub-GS-Gal4 stocks with RU486 for 4 days . DEM exposure resulted in a more rapid demise of the flies in these cohorts compared to matching controls ( Figs 2F and S3B ) . Similarly , inducible over-expression in the UAS-Fs ( 1 ) h-L fly line also resulted in sensitization to DEM stress ( S3A Fig ) . These DEM exposure experiments support the conclusion that Fs ( 1 ) h is a significant regulator of stress defense mechanisms , presumably through its effect on CncC . Through their double bromodomains BET proteins like Fs ( 1 ) h can bind to polypeptides that carry acetylated lysine residues . To assess whether acetyl lysine binding might contribute to the repressive function of the protein towards CncC’s transcriptional activity , we conducted experiments with JQ1 , a specific inhibitor of BET-domain interactions with acetylated substrates [32] . Treatment of S2 cells with JQ1 caused a strong activation of ARE-luciferase reporter activity that was markedly reduced by knock down of either CncC or its obligate heterodimerization partner MafS ( Fig 3A ) . Consistent with the observation that dsRNA-mediated knock down of either CncC or MafS , individually does not completely eliminate the targeted transcripts , a residual response of the reporter to JQ1 could be observed . To confirm the conclusion that the active form of Drosophila Nrf2 , the MafS-CncC heterodimer , specifically mediates the effect of JQ1 we combined CncC and MafS knock down conditions . Under these conditions , the JQ1-mediated induction of the ARE reporter was almost completely abrogated . This observation confirms the specificity of the JQ1-effect and further supports our conclusion that Fs ( 1 ) h affects ARE activity by interfering with CncC function . Similarly , dietary JQ1 exposure ( Fig 3B ) resulted in a strong enhancement of ARE reporter activity also in flies . This stimulatory effect does not seem to be a stress response , as we did not detect any increased mortality in flies or cells after treatment with JQ1 as they would result from treatment with Nrf2-inducing stressors such as paraquat or DEM . The experiment described above suggests that JQ1 can relieve the inhibition of CncC by Fs ( 1 ) h . Since JQ1 has been shown to block the interaction of BET bromodomains with acetyl lysine residues , we proposed that Fs ( 1 ) h would interact with acetylated CncC and that this interaction would be disrupted by JQ1 , causing an increase in CncC activity . Indeed , specific acetylated forms of mammalian Nrf2 have been described [33 , 34] . To assess whether CncC might also be a substrate for acetylation and could therefore represent a potential binding partner for the Fs ( 1 ) h bromodomains , we performed immune precipitation/western blot experiments . S2 cells expressing a Flag-epitope tagged version of CncC were processed for immuno-precipitation with an anti-Flag antibody . Staining western blots of the resulting immuno-precipitates with a generic anti-acetyl lysine antibody revealed that , like its mammalian counterpart , CncC protein is acetylated ( Fig 3C ) . Treatment of the cells with the broad-spectrum HDAC inhibitor LBH589 increases the acetylation of the immuno-precipitated CncC , confirming that the protein is a substrate for acetylation / deacetylation reactions in vivo . Next , we tested whether Fs ( 1 ) h might physically interact with CncC and , if so , whether Fs ( 1 ) h binding correlates with CncC acetylation status . Confirming this idea , we found that Flag-tagged CncC expressed in Schneider cells could be co-immuno-precipitated with endogenous Fs ( 1 ) h-L . The yield of the recovered Fs ( 1 ) h-L increased in the presence of HDAC inhibitor and decreased in the presence of JQ1 ( Fig 3C ) . These results indicated that Fs ( 1 ) h-L and CncC bind to each other through a BET bromodomain–acetyl lysine interaction . Next we wanted to investigate whether the stimulatory effect of Fs ( 1 ) h inhibition on CncC target gene expression might be mediated by changes in the expression of CncC or some other gene . Therefore , we treated S2 cells with the protein synthesis blocker cycloheximide prior to JQ1 exposure . Under those conditions , protein synthesis , as measured by luciferase expression ( Fig 4A ) , was completely inhibited , but the JQ1-mediated induction of the CncC target genes gstD1 and keap1 was unaffected ( Fig 4B and 4C ) . We conclude that Fs ( 1 ) h regulates CncC at the post-translational level to affect the expression of its target genes . Such a model fits our hypothesis that Fs ( 1 ) h regulates CncC protein function by physically interacting with it . The cnc gene generates several different protein products by alternative splicing [37 , 38] . The three initially described Cnc splice forms CncA , CncB and CncC all share a common C-terminal region that comprises the bZIP dimerization and DNA-binding domain . Moving from the A to the C isoform the polypeptides include progressively more N-terminal sequences . Of these three gene products only CncC contains the NEH2 domain , which mediates binding to and repression by Keap1 [8] . To assess which sequences are required for binding to Fs ( 1 ) h we performed co-immuno-precipitation experiments with different Cnc protein isoforms . Epitope-tagged versions of CncA , B , and C were expressed in S2 cells and immuno-precipitated with an anti-Flag antibody . The immuno-precipitates were analyzed by western blot probed with an anti Fs ( 1 ) h antibody that recognizes both the short and the long isoforms of Fs ( 1 ) h . Endogenous Fs ( 1 ) h-L protein was only detected in the immuno-precipitated material from S2 cells that expressed the longest isoform , CncC . However , it was not co-precipitated with CncA or CncB ( S4 Fig ) . Notably this behavior is shared by Keap1 , which also binds to CncC only but not to the two shorter isoforms . No interaction between CncC and Fs ( 1 ) h-S was identified in this experiment . This is consistent with our finding that only Fs ( 1 ) h-L is involved in the negative regulation of CncC . We conclude that sequences within the region of amino acids 1–578 which are unique to CncC , the only isoform with recognized functions in stress response , are required for both Fs ( 1 ) h-L and Keap1 binding . Further experiments are required to test if the binding of the two inhibitory proteins , Keap1 and Fs ( 1 ) h can happen simultaneously or is mutually exclusive [25–27] . Interestingly , we did detect the short isoform Fs ( 1 ) h-S as in the co-precipitate with CncB ( S4 Fig ) . CncB is not involved in stress responses , but functions in embryonic patterning and development . It has been suggested to act as a negative regulator of transcription [10 , 37] . The functional relevance of a potential interaction between CncB and Fs ( 1 ) h-S awaits experimental analysis . The discovery that Fs ( 1 ) h physically interacts with and inhibits CncC raised the question of how this activity relates to the regulatory function of Keap1 , the canonical inhibitor of Nrf2 proteins . Specifically , we wanted to determine whether Fs ( 1 ) h acts via Keap1-mediated CncC degradation , or through an independent mechanism to control target gene expression . To address this question we conducted transfection experiments in S2 cells with drugs and RNAi’s that specifically interfere with Keap1 or Fs ( 1 ) h functions ( Fig 5A ) . The experiments described above indicated that JQ1 prevents the inhibitory effect of Fs ( 1 ) h on CncC , thereby stimulating ARE-dependent gene expression . Consistent with this interpretation , fs ( 1 ) h knock down and JQ1 exposure stimulate ARE-luciferase reporter activity to a similar degree ( Fig 5A , compare lane 1 with 2 and 3 ) , and combining the fs ( 1 ) h RNAi with JQ1 treatment does not further enhance this activity ( Fig 5A , lane 4 ) . Likewise , oltipraz exposure and Keap1 knock down activated the reporter to comparable levels , yet the effect of the two treatments is non-additive , confirming that oltipraz acts specifically through inhibition of the Keap1-CncC interaction ( lanes 5–7 ) . Interestingly , however , eliminating the function of both Fs ( 1 ) h and Keap1 simultaneously by combined RNAi and drug treatments caused a synergistic activation of the reporter ( lanes 8 and 9 ) . Similarly , combining JQ1 and Oltipraz treatment resulted in a synergistic activation of the ARE reporter in S2 cells over a range of concentrations ( Fig 5B ) . Confirming the results from S2 cells , the combined treatment of adult flies with oltipraz and JQ1 lead to stronger induction of the gstD-GFP reporter relative to the effects of either oltipraz or JQ1 alone ( Fig 5C ) . gstD1 is a target gene of CncC and a gstD-GFP reporter had previously been generated by placing a GFP transgene under the control of a gstD1 promoter fragment [8] . The induction of gstD-GFP reporter activity in flies in response to JQ1 treatment also indicated that natural CncC regulated enhancers , and not only synthetic ARE reporters are responsive to Fs ( 1 ) h inhibition in adult Drosophila . Mirroring the measurements of luciferase reporter constructs , the expression of four conserved Nrf2 target genes ( gstD1 , gstE9 , keap1 , gclc ) was inducible by treatment with JQ1 or oltipraz and combined treatment had a more than additive effect relative to the response to the individual drugs ( Fig 5D ) . In further experiments we found that fs ( 1 ) h knock down also led to a synergistic activation of ARE luciferase reporter activity when combined with either Keap1 knock down or CncC over-expression . Presumably , CncC over expression overwhelms the capacity of Keap1 to target the protein for degradation . Thus , both in conditions of CncC over-expression and Keap1 knock down , CncC can accumulate in the nucleus . Gene activation by this increased nuclear pool of CncC protein would then be under negative control by Fs ( 1 ) h ( S5A Fig ) . Taken together , these results indicate that Fs ( 1 ) h and Keap1 regulate Nrf2 function through independent mechanisms . The effective and specific pharmacological activation of the Nrf2 response by combined JQ1 and oltipraz treatment might be therapeutically beneficial , provided that the mechanisms that we have characterized in Drosophila are conserved and operational in mammals . Interestingly , it has been shown in recent publications [39 , 40] that mammalian Nrf2 can be stimulated by JQ1 . We therefore tested if mammalian cells displayed cooperative activation of Nrf2 target genes in response to the combinatorial treatment that we had found effective in Drosophila . Confirming this idea , combined treatment of human HEK293 cells with sulforaphane , a Keap1 inhibitor functionally similar to oltipraz , and JQ1 results in a more than additive activation when compared to individual treatments ( Fig 5E ) . In addition , JQ1 but not sulforaphane caused synergistic activation of NQO1-luciferase reporter when combined with over-expressed Nrf2 , indicating that the Keap1-independent regulation of Nrf2 by BET protein is conserved in mammals ( S5B Fig ) . Combined dosing of these two drugs might therefore open new avenues for treatment of diseases where increased Nrf2 activity could be beneficial . If a combinatorial treatment with the Keap1 inhibitor oltipraz , and the BET bromodomain inhibitor JQ1 activates Nrf2-dependent stress defense and antioxidant gene expression programs more strongly than oltipraz or JQ1 alone , we would expect a further increase in the resistance of flies to acute oxidative challenges . To confirm this prediction we exposed adults to a lethal dose of DEM after they had been kept on food containing oltipraz , JQ1 or both drugs for 4 days . Monitoring the time course of mortality showed that pretreatment with oltipraz and JQ1 alone enhances the oxidative stress tolerance when compared to the control group . However , pretreatment with a combination of oltipraz and JQ1 increased survival times even further ( Figs 6A and S3D ) . Thus , combinatorial treatment with oltipraz and JQ1 can yield effective protective effects against oxidative challenges and may also be useful for therapeutic applications in which Nrf2 function is expected to be beneficial .
BET proteins including Drosophila Fs ( 1 ) h and the members of the mammalian Brd family function as epigenetic readers and transcriptional regulators [28 , 29 , 41 , 42] . They can affect the function of specific transcription factors such as NF-kB , FosL and Twist , but also interact with components of the general transcription machinery such as P-TEFb [23–27 , 43] . Mammalian BET proteins , most notably Brd4 , have raised broad biomedical interest as well , because they can support several types of cancer , including NUT midline carcinoma , multiple myeloma and acute leukemia [44–47] . The implication of BET proteins in these malignancies could have clinical impact as the availability of specific inhibitors makes them potentially promising drug targets . In order to evaluate and exploit these opportunities , it is critical to gain a better understanding of the complex functions and molecular targets of BET proteins . We have identified the Drosophila BET protein Fs ( 1 ) h as a novel inhibitor of the Drosophila Nrf2 homolog CncC . The fs ( 1 ) h gene encodes two different isoforms , Fs ( 1 ) h-S and Fs ( 1 ) h-L . Fs ( 1 ) h-S has previously been reported to transcriptionally activate the ultrabithorax ( ubx ) gene , a function that is not shared by the larger Fs ( 1 ) h-L isoform [18] . Conversely , our study showed that only Fs ( 1 ) h-L , but not Fs ( 1 ) h-S , has the potential to inhibit CncC target gene activation . The divergence of function between Fs ( 1 ) h-S and L is also reflected in their respective genomic binding patterns: A recent genome-wide ChIP-seq study showed that Fs ( 1 ) h-S and Fs ( 1 ) h-L are present at different locations throughout the Drosophila genome [28] . Whereas Fs ( 1 ) h-S was enriched in promoters and enhancers , Fs ( 1 ) h-L was predominantly localized to the insulators . Interestingly , our analysis of the same ChIP-seq data revealed that Fs ( 1 ) h-L also localizes to the promoter region of many CncC target genes such as gstD1 , gclC , and keap1 . It is possible that Fs ( 1 ) h-L interacts with promoter-bound CncC to prevent the induction of its target genes . Nrf2 is an attractive drug target . Compounds such as oltipraz and sulforaphane which interfere with the Keap1-Nrf2 interaction to activate target genes without causing cell stress , are effective as cancer chemo-preventive agents in animal experiments [48–51] . Nrf2-activating drugs can also be beneficial in the treatment of other diseases that are associated with oxidative stress , including neurodegenerative and inflammatory conditions [52 , 53] . Indeed , an Nrf2 inducer , dimethyl fumarate , marketed under the brand name Tecfidera , has recently gained FDA approval for treatment of multiple sclerosis [54 , 55] . The identification of JQ1 as a CncC activator and , the discovery that JQ1 and oltipraz synergistically activate Nrf2 dependent gene expression programs in Drosophila and mammalian cells suggest innovative therapeutic options . Conventional Nrf2 activating drugs like sulforaphane and oltipraz , though effective in some cases , show moderate induction of Nrf2 and often have off-target effects [49 , 52] . Combinations of JQ1 and oltipraz-like drugs might stimulate Nrf2 dependent protective gene expression more efficiently than either compound alone , or could achieve beneficial effects at lower doses , thereby decreasing the risk of unwanted side effects . Based on the known properties of Fs ( 1 ) h and Keap1 and our cell culture and biochemical experiments we suggest a mechanisms by which Fs ( 1 ) h represses CncC function . As opposed to the effect of Keap1 , which targets CncC for cullin 3-mediated proteasomal degradation in the cytoplasm , the inhibitory , acetylation-dependent interaction between Fs ( 1 ) h and CncC presumably occurs in the nucleus where the BET protein resides ( Fig 6B ) . The independent mechanisms of CncC inhibition by Keap1 and Fs ( 1 ) h are consistent with the synergistic effect of combined JQ1 and oltipraz treatment on Nrf2 target gene expression . fs ( 1 ) h is the only BET protein encoding gene in Drosophila and Brd4 is the closest ortholog of Fs ( 1 ) h in mammals [28] . Recent publications by Michaeloudes et al and Hussong et al have implicated Brd4 in the regulation of mammalian Nrf2 [39 , 40] . In agreement with our conclusions , Michaeloudes and colleagues have found that Brd4 and other mammalian BET proteins namely Brd2 and Brd3 , physically interact with Nrf2 . Hussong and colleagues , on the other hand , suggested a somewhat different model of Nrf2 regulation by Brd4 . They reported that JQ1-treatment of human cells activates Nrf2 target genes , but suppresses Keap1 expression . Accordingly , they reason that the activation of Nrf2 target genes by JQ1 might be mediated by a loss of Keap1 repression [40] . The data we generated in Drosophila S2 cells , on the other hand , show that both Fs ( 1 ) h knock down and JQ1 treatment leads to increased keap1 transcription , which is consistent with the previous identification of the Keap1 gene as a target of CncC mediated feedback regulation . It is possible that species- or cell type-specific differences in JQ1-Keap1 cross talk exist and account for these divergent results . In any case , our finding that JQ1 and sulforaphane synergistically activate an ARE reporter in human cells supports the notion that combinatorial treatment with these two drug types might be therapeutically beneficial . Our data show that the suppression of CncC activity by Fs ( 1 ) h relies on a bromodomain–acetyl lysine interaction . This finding implies a repressive function of acetylation on CncC’s transactivation potential . Such a conclusion appears to be at odds with reports showing acetylation of Nrf2 by CBP to increase its binding to target DNA and to enhance target gene transcription [33 , 34] . However , Mercado et al . have reported that inhibition of HDAC2 and the resulting increase in Nrf2 acetylation can suppress Nrf2-mediated target gene induction and antioxidant defense in chronic obstructive pulmonary disease ( COPD ) [56] . It seems therefore that Nrf2 acetylation can have positive as well as negative effects on the transcriptional function of Nrf2 proteins . Further studies are needed to uncover whether specific acetylation marks on Nrf2 lead to either activation or suppression of its transcriptional activity and whether BET proteins can selectively interact with the inhibitory acetylation marks on Nrf2 . The identification of BET proteins as Nrf2 repressors adds another facet to an increasingly complex picture of Nrf2 signaling and biology . Over the last few years several other Keap1-independent pathways of Nrf2 regulation have been described based on experiments in mammalian cell culture: for example it was shown that PKC-δ mediated phosphorylation of Ser40 of mammalian Nrf2 promotes its stabilization and nuclear translocation [57–59] . The Src-family tyrosine kinase Fyn can phosphorylate Tyr568 of Nrf2 causing its nuclear export and degradation [60] . Glycogen synthase kinase 3 beta ( GSK-3β ) can phosphorylate Fyn and increase its nuclear accumulation and thereby promotes nuclear export of Nrf2 and inhibition of Nrf2 signaling [61] . In addition , GSK-3β can phosphorylate the serine residues at the β-TRCP-binding motif ( DSGIS338 ) in the Neh6 domain of Nrf2 and promote Cullin 1 dependent proteasomal degradation by β-TRCP [62] . Future experiments will help to evaluate whether the Fs ( 1 ) h and CncC interaction cooperates with any of these more recently described mechanisms of Nrf2 regulation .
All plasmids were generated by standard recombinant DNA and PCR methods . The pUAS-HA-attB plasmid was generated by cloning attB sequence amplified with PattB-F and PattB-R primers into pUAS-HA plasmid [8] . UAS-Fs ( 1 ) h-L plasmid was generated in two steps . First the Fs ( 1 ) h short isoform cDNA was amplified from LD26482 cDNA clone with PFs ( 1 ) hS-UAST-HA-F and PFs ( 1 ) hS-UAS-HA-R primers and was cloned into pUAS-HA-attB plasmid to generate pUAS-Fs ( 1 ) h-S-HA-attB plasmid . Then the cDNA for the C-terminal motif of Fs ( 1 ) h-L was amplified from total Drosophila cDNA with PUAST-Fs ( 1 ) h-L-int-F and PUAST-Fs ( 1 ) hL-int-R primers and was cloned into pUAS-Fs ( 1 ) h-S-HA-attB plasmid to generate pUAS-Fs ( 1 ) hL-HA-attB plasmid . The primer sequences are provided in Table A in S1 Text . UAS-Fs ( 1 ) h-L transgenic fly strain was generated by Genetic Services Inc , MA using ΦC31 recombinase-mediated site-directed transformation [63] . EP-Fs ( 1 ) h fly line #10097 ( Bloomington Stock Center ) was used to over-express Fs ( 1 ) h whereas v51227 and v108662 fly lines ( Vienna Stock Center ) were used to knock down Fs ( 1 ) h . Drosophila S2 cells and HEK293 cells were transiently transfected with plasmid constructs using the calcium phosphate method [11] . To study the effect of JQ1 on reporter gene expression in vivo , 5-day-old flies that were mated for one day and then separated into males and females , were fed food supplemented with 0 . 25 mM JQ1 ( APExBIO ) for 48 hours . 15–20 flies were used in each group for these experiments and 3–5 representative flies were chosen randomly for imaging . To assess the effect of JQ1 on the cell-based reporters , S2 cells were transiently transfected with the reporter plasmids by the calcium phosphate method . 8 hours after the PBS wash and medium change , the transfected cells were transferred to 96-well plates and treated with 1 μM JQ1 and were incubated at 25°C for 24 hrs . Treatment with oltipraz was done following the same protocol [11] . In order to study the effect of different chemicals on the ARE reporter in mammalian cells , transfected 293T cells were treated with 0 . 5μM JQ1 and 10μM sulforaphane ( Enzo Life Sciences ) and were incubated at 37°C for 18hrs . DMSO was used as the solvent control in all the drug treatments . dsRNAs ( 200–700 bp ) were synthesized and purified following the protocol provided as described [64] using ‘T7 RiboMAX Express RNAi System’ kit ( Promega ) and ‘RNeasy Kit’ ( QIAGEN ) . Briefly , 1X106 S2 cells were bathed with 8μg dsRNAs and these cells were transfected with luciferase reporter plasmids using the calcium phosphate method 3 days after dsRNA treatment . Predesigned dsRNAs obtained through the E-RNAi webservice [65] were used to knock down Fs ( 1 ) h , CncC , Keap1 whereas Fs ( 1 ) h-L was knocked down using the dsRNA described by Kockmann et al . [29] . dsRNA targeting the unique 3’ UTR region of Fs ( 1 ) h-S was designed by the ‘SnapDragon’ webservice . The sequences of primers used to generate amplicons for dsRNA synthesis are provided in Table B in S1 Text . The ‘Dual Glo Luciferase Assay System’ kit ( Promega ) was used to measure the activities of cell-based firefly and renilla luciferase reporters . Oxidative stress resistance of adult flies of different genotypes was assessed as previously described [8] . Newly emerged flies of the specified genotypes were mated for one day , separated into females and males , and at 5 days of age were transferred to RU486-containing ( 300μM ) or control food containing the solvent ( ethanol ) . After 4 days 4 groups of 20 flies from each condition were starved for 3 hours in empty vials , and then fed a solution of 5% sucrose ± a semi-lethal dose of DEM ( 20mM ) . Survivors were scored after 36 hours and Log-rank tests were performed on the survivorship data using ‘GraphPad Prism’ software . In order to study the synergistic effect of oltipraz and JQ1 on stress sensitivity , flies raised on food supplemented with 0 . 4mM oltipraz and/or 0 . 1mM JQ1 for four days were exposed to similar DEM treatment and the survivorship data was collected and analyzed as before . To study the effect of Fs ( 1 ) h knock down and Fs ( 1 ) h-L over-expression on ARE activity in clones of tissue hsFlp; Act>Y>Gal4 , UAS-RFP , ARE-GFP; UAS-Fs ( 1 ) h-RNAi and hsFlp; Act>Y>Gal4 , UAS-RFP , ARE-GFP; UAS-Fs ( 1 ) h-L flies were used . The embryos and larvae were incubated at 18°C before L2 larvae were heat treated at 37°C for 30 minutes and then returned to 18°C . Adult flies were dissected in PBS and crops and ejaculatory bulbs were fixed at room temperature for 30 minutes in 100mM glutamic acid , 25mM KCl , 20mM MgSO4 , 4mM sodium phosphate , 1mM MgCl2 , and 4% formaldehyde ( pH 7 . 5 ) . DNA was stained with Hoechst dye . Confocal images were collected using a Leica TCS SP5 system and were processed using Adobe Photoshop . S2 cells transfected with pAct-Gal4 and pUAS-CncC-FLAG plasmids were harvested and washed with cold 1XPBS . The cells were then re-suspended in lysis buffer ( 50mM Tris-HCl , 200mM NaCl , 5mM EDTA , 5% Glycerol , 0 . 2% NP-40 ) that contained protease inhibitor complex ( Roche ) and kept on a nutator at 4°C for 1 hour . The cell debris was removed by centrifugation and the lysate was passed through hypodermic syringe to shear DNA . The protein concentration was estimated using Bradford’s reagent and 20–30 μg of protein from each sample was set aside as input . The lysate was pre-cleared with Protein G beads ( GE Biosciences ) for 1 hour at 4°C . The beads were separated by centrifugation and the cleared lysate was transferred into the fresh chilled tube . Anti-FLAG ( Sigma-Aldrich Co ) antibody was added at 2 μg per mg lysate protein concentration and was incubated at 4°C for 2 hours . Then 20 μl of 50% Protein G bead slurry was added to the lysate and incubated overnight on rotating disc at 4°C . The beads were spun down and were washed 5 times with 750 ul of cold lysis buffer . After the final wash , 7 . 5 μl of sample buffer and 7 . 5 μl of lysis buffer were added to the beads and incubated at 95°C along with the input samples for 10 minutes . The beads were spun down at the top speed for 5 minutes and the supernatant was loaded for electrophoresis . For LBH and JQ1 treatment , the cells were treated with 2μM of LBH ( APExBIO ) and 10 μM JQ1 for 6 hours prior to the cell lysis . JQ1 was also added to the lysate from cells treated with JQ1 at 10 μM concentration after the cell lysis . Primary antibody specific to Fs ( 1 ) h-L isoform ( 1:2000 dilution ) ( Kindly provided by Dr . Victor Corces ) [28] and secondary anti-rabbit antibody ( 1:5000 dilution ) ( BioRad ) were used to probe the western blot membrane to examine the binding of Fs ( 1 ) h-L protein to CncC-FLAG protein . Antibody that recognizes both isoforms of Fs ( 1 ) h ( Kindly provided by Dr . Igor Dawid ) was used in 1:2000 dilution in western blot experiments to validate the selective knock down of Fs ( 1 ) h-S and Fs ( 1 ) h-L with isoform-specific dsRNAs . Antibody against acetylated lysine ( Cell Signaling Technology ) was used in ( 1:1000 ) concentration . mRNA was prepared from whole flies or from S2 cells using Trizol reagent ( Invitrogen ) . DNaseI ( NEB ) was used to remove contaminating DNA before phenol: chloroform: isoamyl alcohol ( 25:24:1 , Amresco ) extraction . Maxima reverse transcriptase ( Fermentas ) and oligo dT primers were used to generate cDNA . cDNA was diluted 1:50 to serve as template for quantitative real time PCR ( qPCR ) . qPCR reactions were done in triplicates using qPCR super-mix ( BioRad ) on a BioRad MyIQ thermal cycler . ‘Delta-delta Ct’ method was used for normalization to actin5c transcript levels . Data shown are averages and standard deviations from at least three biological replicates . The sequences of primers used to qPCR are provided in Table C in S1 Text . | Nrf2-related transcription factors regulate gene expression programs that protect organisms against chemical or oxidative stress . Nrf2-activating drugs hold promise for the treatment of diseases that are connected to oxidative stress or inflammation . We identified Fs ( 1 ) h , a bromodomain-containing BET protein , as a negative regulator of Nrf2 function in Drosophila . BET proteins are involved in transcription regulation and chromatin organization and have been implicated in several diseases , including cancer . Fs ( 1 ) h interacts with acetylated lysines on CncC , the homolog of Nrf2 in Drosophila , and thereby prevents target gene activation . Nrf2 can be released from this inhibitory effect by small molecules that specifically interfere with the binding of BET proteins to acetylated targets . Fs ( 1 ) h regulates Nrf2 independently of Keap1 , a well-studied Nrf2 regulator . Consequently , chemical inhibitors of Keap1 and of Fs ( 1 ) h can be combined to achieve synergistic activation of Nrf2 target genes and strongly boost oxidative stress tolerance in Drosophila . The Keap1-independent mechanism of Nrf2 regulation is conserved in mammals . We suggest that the synergistic effect of combinatorial Nrf2 targeting drugs may be effective for the treatment of different oxidative stress and inflammation-related diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"oxidative",
"stress",
"gene",
"regulation",
"regulatory",
"proteins",
"dna-binding",
"proteins",
"animals",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"pharmaceutics",
"transcription",
"factors",
"epigenetics",
"drosophila",
"research",
"and",
"analysis",
"methods",
"genetic",
"interference",
"proteins",
"gene",
"expression",
"chemistry",
"insects",
"arthropoda",
"biochemistry",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"post-translational",
"modification",
"acetylation",
"genetics",
"biology",
"and",
"life",
"sciences",
"chemical",
"reactions",
"physical",
"sciences",
"drug",
"therapy",
"organisms"
] | 2016 | Keap1-Independent Regulation of Nrf2 Activity by Protein Acetylation and a BET Bromodomain Protein |
Human tuberculosis caused by M . bovis is a zoonosis presently considered sporadic in developed countries , but remains a poorly studied problem in low and middle resource countries . The disease in humans is mainly attributed to unpasteurized dairy products consumption . However , transmission due to exposure of humans to infected animals has been also recognized . The prevalence of tuberculosis infection and associated risk factors have been insufficiently characterized among dairy farm workers ( DFW ) exposed in settings with poor control of bovine tuberculosis . Tuberculin skin test ( TST ) and Interferon-gamma release assay ( IGRA ) were administered to 311 dairy farm and abattoir workers and their household contacts linked to a dairy production and livestock facility in Mexico . Sputa of individuals with respiratory symptoms and samples from routine cattle necropsies were cultured for M . bovis and resulting spoligotypes were compared . The overall prevalence of latent tuberculosis infection ( LTBI ) was 76 . 2% ( 95% CI , 71 . 4–80 . 9% ) by TST and 58 . 5% ( 95% CI , 53 . 0–64 . 0% ) by IGRA . Occupational exposure was associated to TST ( OR 2 . 72; 95% CI , 1 . 31–5 . 64 ) and IGRA ( OR 2 . 38; 95% CI , 1 . 31–4 . 30 ) adjusting for relevant variables . Two subjects were diagnosed with pulmonary tuberculosis , both caused by M . bovis . In one case , the spoligotype was identical to a strain isolated from bovines . We documented a high prevalence of latent and pulmonary TB among workers exposed to cattle infected with M . bovis , and increased risk among those occupationally exposed in non-ventilated spaces . Interspecies transmission is frequent and represents an occupational hazard in this setting .
The most frequent etiologic agent of tuberculosis ( TB ) in humans is Mycobacterium tuberculosis , followed by M . bovis , that causes bovine tuberculosis [1] . The zoonosis caused by M . bovis in developed countries is now considered a sporadic disease since control of the disease in cattle was achieved and pasteurization of dairy products became extensively practiced [2] , [3] . Nevertheless , an increase in the number cases of TB caused by M . bovis in some regions of the United States of America has recently been reported and it has been attributed to migrant populations from low and middle resource countries ( specially from Mexico ) , and in most cases , associated to consumption of unpasteurized dairy products [4] , [5] , [6] . Moreover , we have recently noted an increase in TB cases due M . bovis in a tertiary-care hospital in Mexico , accounting for the 22 . 3% of the TB isolates from 2000–2007 ( Franco-Cendejas R , et al . Mycobacterium bovis infections in a tertiary-care centre in Mexico: A case-control study . 20th ECCMID . Vienna , Austria . April 10–13 , 2010 . Poster 108 ) . Most of the human cases of contagion by close contact with infected animals , occurring among veterinarians , zoo workers , hunters , abattoir workers and dairy farm workers ( DFW ) have been reported in developed countries [7] , [8] , [9] , [10] , [11] , but the situation in low and middle resource countries , where the control of the disease in cattle is poor and the human-cattle interface is more likely to occur , remains poorly documented [12] . Because of insufficient information , the World Health Organization recommended the collection of epidemiological data about human tuberculosis due to M . bovis specially in these regions [13] . Previous medical literature examining this zoonosis shows important limitations . Most studies have focused on active disease , and only a few have reported the prevalence of latent tuberculosis [10] , [11] , [14] . Until the past decade , the only diagnostic test available to identify LTBI was the tuberculin skin test ( TST ) . Presently , interferon-γ release assays ( IGRA ) are considered an interchangeable or complementary test to TST by international guidelines [15] , [16] . To our knowledge , these tests have never been used to detect LTBI in DFW exposed to infected cattle . The main objective of our study was to determine the prevalence of active tuberculosis and LTBI among DFW with different degrees of exposure to cattle , the risk associated to occupational exposure and the occurrence of interspecies transmission .
Dairy production facility: The study was performed in a dairy production and livestock facility located in a municipality of the state of Hidalgo , Mexico . This facility was composed by 126 cowsheds in dairy production , with an average cattle population of 29 , 000 . Inside most of the cowsheds , improvised housing was provided to workers and their families in small rooms near to the areas where farming activities occurred . Abattoir house: We also included workers from an abattoir house in the same municipality that processed cattle from the studied dairy farm facility . Design and study period: A cross-sectional , comparative study was conducted during 2009–2011 . We requested authorization from owners of the 126 cowsheds within the facility to invite their employees to the study . On those facilities where authorization was granted , all the abattoir and dairy production workers and their household contacts living inside the improvised housing in cowsheds , older than 15 years were invited to participate . A standardized questionnaire ( investigating exposure to TB cases , consumption of unpasteurized dairy products , BCG vaccination status as well as other socio-demographic and health conditions ) , full medical exam , TST and IGRA were administered to all consenting participants . Active TB was investigated on individuals with respiratory or systemic symptoms by AFB smear and culture of appropriate sputum samples and chest X-ray . Fasting glucose , complete blood count , urine exam and chest X-ray were also performed . TST was performed by the Mantoux method administering 5 IU ( 0 . 1 ml ) of PPD ( Tubersol Sanofi-Pasteur , Toronto , CA ) in the volar surface of the forearm . After 48–72 hr , the reading was performed by previously trained personal using a caliper to measure the induration and the result was registered in millimeters and determined positive using a 10 mm cut-off . In those cases with a negative result a booster test was administered after three weeks . Venous blood was drawn to determine interferon gamma response to ESAT-6 and CFP-10 antigens by ELISPOT ( Lionex Diagnostics & Therapeutics GmbH , Braunschweig , Germany ) . The test was determined positive if more than 6 spots were observed . To determine the bovine tuberculosis burden and circulating strains in the cattle , a parallel surveillance was conducted in animals during the same period . As a regular practice in the facility , the cowshed owners are requested to perform a necropsy study to cattle that dies within the facility . Using standardized procedures , a veterinarian visually inspected lymph nodes , lungs and other organs for lesions suggestive of tuberculosis . Data from all necropsy reports were obtained during the study period and tuberculosis macroscopic diagnosis were recorded . Tissue samples were obtained for M . bovis culture and spoligotyping whenever possible . Slaughtered animals were not formally inspected and no records as to the macroscopic findings were available or informed to us . In those subjects who reported respiratory symptoms for more than 2 weeks , 3 sputa samples were collected and Ziehl-Neelsen stains were examined microscopically for the presence of acid-fast bacilli ( AFB ) . The samples from humans and animals were decontaminated and digested to increase the positivity yield , and were inoculated in solid culture media ( Lowestein-Jensen and Stonebrink ) and MGIT tubes ( Becton-Dickinson , Sparks , MA , USA ) according to the manufacturer's specifications . Positive cultures were identified as M . bovis by conventional biochemical tests , and DNA probing ( Accuprobe , GEN-PROBE , San Diego , CA ) . DR region was amplified by polymerase chain reaction ( PCR ) and then analyzed according to the spoligotyping protocol , as described elsewhere [17] , M . tuberculosis H37Rv and M . bovis BCGP3 were used as controls . Data from spoligotyping of the strains was introduced in an international database ( www . mbovis . org ) . The following exposure groups were formed according to type of activity , duration and conditions of exposure to cattle; high exposure: individuals in direct contact with livestock working in closed spaces ( abattoir workers; veterinary personnel performing cattle necropsies , foremen and milkers ) , medium exposure: participants working in open spaces in direct contact with cattle ( tractor operators , breeders , those in charge of providing food to cattle , veterinary personal , and maintenance technicians ) and household contacts living in the cowshed; and low exposure: workers without direct contact with livestock ( owners of the cowshed , administrative clerks , and people involved in commercial activities ) ( Table 1 ) . Using bivariate and multivariate analyses ( unconditional logistic regression ) we tested the association between exposure groups and other relevant variables with LTBI ( as determined by TST or IGRA ) . Three combinations were additionally analyzed: IGRA and TST positive compared with any other combination of test results , IGRA or TST positive compared with IGRA and TST negative results and IGRA and TST positive compared with IGRA and TST negative results . Variables with p<0 . 20 in the bivariate analysis and biological plausibility were included in multivariate models . We estimated the odds ratios ( OR ) and 95 percent confidence intervals ( CI ) , and identified the covariates that were independently associated with each outcome . Statistical analysis was performed using STATA 11 . 0 software ( StataCorp , College Station , Tx ) . The aims of the study were communicated to the participants and a written informed consent form was signed before the inclusion to the study . Whenever the subjects were minors the informed consent was given by the parents or mentors . Respecting animal population the owners of the cattle were informed of the purpose of the study and their permission was given for data use and culture recollection during routine necropsy studies . The protocol was reviewed and approved by the Salvador Zubiran National Institute of Medical Sciences and Nutrition ethics committee ( approval reference 234 ) . All participants with abnormal results were referred for appropriate treatment . All contacts of pulmonary TB cases were studied as specified in the national regulations , although excluded from analysis if they lived outside the facility .
During the study period 1561/29 , 000 cows died within the facility . Necropsy was performed on 718/1561 of dead cattle . Macroscopic diagnosis of TB was made in 247/718 necropsies; of these , mycobacterial cultures were performed on 163/247 and M . bovis was identified on 154 instances ( Table 2 ) . Estimated eligible population was comprised of 1 , 411 persons . Four hundred and forty two subjects from 56 different cowsheds were invited to participate , 389 subjects signed informed consent; 68 subjects were excluded of the study because of incomplete study procedures ( TST or IGRA not performed ) or residence outside the facility ( 10 household contacts ) . Data on 311 subjects were analyzed . ( Fig . 1 ) Median age of the subjects was 36 years ( IQR 27–45 ) , 78 . 0% were male . The median time living or working in the dairy facility was 132 months ( IQR 72–240 ) . One hundred and eighty eight subjects ( 60 . 4% ) lived inside the cowshed at the moment of the study . The most frequently reported activities were milker ( 22 . 3% ) , followed by veterinarians ( 12 . 3% ) , diverse activities with no contact with cattle ( 12 . 3% ) , and feeders ( 11 . 0% ) . Household contacts represented 11 . 0% of study population . Abattoir workers accounted for 2 . 5% . The median time of daily exposure to cattle was 6 h ( IQR 0–8 ) . According to the level of exposure , 33 . 9% ( 105/309 ) , 46 . 9% ( 145/309 ) and 19% ( 59/309 ) of the subjects were assigned to the high , medium and low exposure groups , respectively . ( Table 3 ) Eighty-five percent ( 265/311 ) of subjects had a BCG scar , 9 . 9% ( 31/311 ) informed previous close contact with a TB case , and 30 . 0% ( 93/310 ) informed consumption of unpasteurized dairy products . Fourteen per cent ( 37/251 ) of the chest X rays showed TB scars . Four subjects reported having been previously diagnosed with TB ( 1 . 29% ) ( Table 3 ) . The median induration of the TST in the population was 14 mm ( IQR 10–20 ) , the prevalence of LTBI diagnosed by TST was 76 . 2% ( 95% CI , 71 . 4–80 . 9% ) . Reactivity was 86 . 6% , 70 . 3% and 71 . 1% among the high , medium and low exposure groups , respectively . Statistical difference was found between reactivity of the high exposure group when compared with medium and low exposure groups . For the rest of the analysis , activities comprised in the high exposure group were considered high exposure activities . Thirty nine percent of the TST positive ( 91/235 ) subjects performed high exposure activities while 19% ( 14/74 ) of the TST negative subjects ( p = 0 . 002 ) belonged to this group . Twenty seven per cent ( 64/236 ) of the TST positive subjects had history of unpasteurized products consumption compared with 39% ( 29/74 ) of the TST negative ( p = 0 . 04 ) . Sixty one percent ( 145/236 ) of the TST positive subjects reported more than 4 hours of daily contact with cattle while 49% ( 36/74 ) of the TST negative subjects reported that intensity of contact ( p = 0 . 051 ) . No other differences were found between other TB related characteristics and the TST positivity ( Table 4 ) . Multivariate analysis revealed association between the high exposure group and TST reactivity ( OR 2 . 72; 95% CI 1 . 31–5 . 64 , p = 0 . 003 ) adjusting by age , gender and other TB related co-variables ( Table 5 ) . The prevalence of LTBI diagnosed by IGRA was 58 . 5% ( 95% CI , 53 . 0–64 . 0% ) . A positive test was found in 70 . 4% , 48 . 9% and 61 . 2% of individuals assigned to high , medium and low exposure groups respectively . As with TST results , statistical difference was found between positivity of the high exposure group when compared with medium and low exposure groups . For the rest of the analysis , activities comprised in the high exposure group were considered high exposure activities . Ninety four percent ( 171/181 ) of the IGRA positive and 88 . 0% ( 112/128 ) of IGRA negative subjects had a length of stay at the facility longer than 1 year ( p = 0 . 03 ) . Forty one percent ( 74/181 ) of the IGRA positive and 24 . 0% ( 31/128 ) of the IGRA negative subjects performed high exposure activities ( p = 0 . 002 ) . No differences were found between other TB related characteristics and the IGRA positivity . ( Table 4 ) The high exposure group was associated with a positive result in the IGRA test ( OR 2 . 38; 95% CI 1 . 31–4 . 30 , p = 0 . 003 ) after adjusting by age , gender and other TB related co-variables . ( Table 5 ) Forty four percent ( 65/148 ) of the TST+ & IGRA+ subjects and 25% ( 40/161 ) of the subjects with any other combination performed high exposure activities ( p<0 . 001 ) . No other characteristic was found statistically different ( Table 6 ) . Multivariate analyses revealed association of this result with high exposure activities ( OR 2 . 49 95% CI 1 . 40–4 . 42 ) ( Table 5 ) . Thirty seven percent ( 100/268 ) of the TST+ and/or IGRA+ and 12% ( 5/41 ) of the subjects with both negative tests results performed high exposure activities ( p = 0 . 002 ) ( Table 6 ) . Multivariate analyses revealed association of this result with high exposure activities ( OR 4 . 67 95% CI 1 . 62–13 . 47 ) ( Table 5 ) . Forty four percent ( 65/148 ) of the TST+ & IGRA+ subjects and 12% ( 5/41 ) of the subjects with both negative tests results performed high exposure activities ( p<0 . 001 ) . No other characteristic was found statistically different ( Table 6 ) . Multivariate analyses revealed association of this result with high exposure activities ( OR 6 . 09 95% CI 2 . 04–18 . 23 ) . ( Table 5 ) At the time of evaluation , 9 . 6% ( 30/311 ) of the subjects had had respiratory symptoms for more than two weeks . Of these , only two cases were bacteriologically confirmed with pulmonary TB . The prevalence of active disease among the study population was estimated to be of 643 cases/100 , 000 inhabitants . The first case was identified in August 2010; the patient was a 48 year-old male , with type-2 diabetes mellitus , poorly controlled . When diagnosed , he worked as a salesman in a grocery store inside the facility , but he had worked as a milker in one of the cowsheds of this facility 12 years before . M . bovis was identified from sputa and showed a spoligotype ( 676 741 077 777 600 ) that was not present among the circulating strains in the cattle , and its lineage was different from those reported in the international electronic database . Following the guidelines of Mexico's National TB Control Program , ( Modificación a la Norma Oficial Mexicana NOM-006-SSA2-1993 , para la prevención y control de la tuberculosis en la atención primaria a la salud: Secretaria de Salud , Diario Oficial de la Federación , Mexico , 2000 . ) , the patient received directly observed daily doses of two months isoniazid ( H ) , rifampin ( R ) pyrazinamide ( Z ) and ethambutol IE ) followed by a three times weekly six-month continuation phase of HR [2HRZE/6 ( HR ) 3] with a favorable clinical outcome . The second case was identified in April 2011; the patient was a 37 year-old woman , with rheumatoid arthritis under treatment with prednisone and methotrexate; she worked as a milker at the moment of diagnosis . M . bovis was isolated in her sputa and the spoligotype of the strain ( SB0121 ) belonged to the BOVIS-1 linage and was identical to the isolates cultured from two animals from the same cowshed where she worked . The patient was treated same as above [2HRZE/6 ( HR ) 3] with a favorable clinical outcome .
We recognize several limitations in the study . First , we did not sample all the facility personnel , because of the reluctance of some of the cowshed owners to allow access to their facilities . However , since we studied 46 . 6% of the total cowsheds and the cowsheds were similar in farming practices , we consider that our results may be representative of the problem in the study area . Secondly , since the tests we used are not specific for M . bovis , we may have detected infection caused by other species of M . tuberculosis complex [27] . It may be hypothesized that at least 30% to 35% of the cases could be attributed to M . bovis infection , since the prevalence of TST positivity in Mexico is around 40% in other settings [18] , [19] , [20] . Thirdly , given the fact that BCG vaccine is administered to all newborns in Mexico , we were unable to determine the proportion of TST reactivity due to BCG vaccination or to exposure to infected cattle . The interpretation of TST results in BCG vaccinated populations has been controversial . However , we consider that TST was useful in our study to identify individuals who had been exposed to infected cattle . TST reactivity associated with neonatal BCG vaccination has been shown to wane after several years [28] . Furthermore , we have previously documented that the TST used together with a standardized questionnaire eliciting information about risk factors for exposure to infectious tuberculosis can identify individuals who have been exposed to an active pulmonary tuberculosis case in an area with high BCG vaccination coverage [26] . Therefore , we believe that the difference in the positivity rates between TST and IGRA cannot be completely attributed to BCG vaccination , but may also reflect the inaccurate performance of IGRA in high tuberculosis burden settings [29] . In conclusion , our findings provide evidence of occupational risk in a dairy production and livestock facility in Mexico and may be representative of settings in which close contact of humans and infected animals occurs . Based on this evidence , it is recommended to establish more stringent prevention and control measures of bovine tuberculosis including protection to workers , especially those who are exposed in closed environments . | Mycobacterium tuberculosis complex causes tuberculosis in humans and other mammals . The complex includes M . bovis , which causes bovine tuberculosis . The main route of transmission of this zoonosis is the consumption of unpasteurized dairy products . Nevertheless , exposure to infected cattle while performing husbandry and farm activities may cause disease as well . In this study we were able to demonstrate: 1 ) A high prevalence of tuberculosis asymptomatic infection ( latent tuberculosis ) among workers exposed to infected cattle; 2 ) A higher probability of infection among individuals who are occupationally exposed in closed spaces; and 3 ) Cattle to human transmission confirmed by molecular methods ( spoligotyping ) . We conclude that occupational exposure is frequent , and therefore strict prevention and control measures are required in these settings . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"bovine",
"tuberculosis",
"in",
"humans",
"neglected",
"tropical",
"diseases"
] | 2013 | Prevalence of Latent and Active Tuberculosis among Dairy Farm Workers Exposed to Cattle Infected by Mycobacterium bovis |
Post-synaptic potential ( PSP ) variability is typically attributed to mechanisms inside synapses , yet recent advances in experimental methods and biophysical understanding have led us to reconsider the role of axons as highly reliable transmission channels . We show that in many thin axons of our brain , the action potential ( AP ) waveform and thus the Ca++ signal controlling vesicle release at synapses will be significantly affected by the inherent variability of ion channel gating . We investigate how and to what extent fluctuations in the AP waveform explain observed PSP variability . Using both biophysical theory and stochastic simulations of central and peripheral nervous system axons from vertebrates and invertebrates , we show that channel noise in thin axons ( <1 µm diameter ) causes random fluctuations in AP waveforms . AP height and width , both experimentally characterised parameters of post-synaptic response amplitude , vary e . g . by up to 20 mV and 0 . 5 ms while a single AP propagates in C-fibre axons . We show how AP height and width variabilities increase with a ¾ power-law as diameter decreases and translate these fluctuations into post-synaptic response variability using biophysical data and models of synaptic transmission . We find for example that for mammalian unmyelinated axons with 0 . 2 µm diameter ( matching cerebellar parallel fibres ) axonal noise alone can explain half of the PSP variability in cerebellar synapses . We conclude that axonal variability may have considerable impact on synaptic response variability . Thus , in many experimental frameworks investigating synaptic transmission through paired-cell recordings or extracellular stimulation of presynaptic neurons , causes of variability may have been confounded . We thereby show how bottom-up aggregation of molecular noise sources contributes to our understanding of variability observed at higher levels of biological organisation .
The great majority of axons use action potentials ( APs ) to transmit information reliably to synapses . Once the AP arrives at the synapse the characteristics of its waveform are fundamental in determining the strength and reliability of information transmission , as was extensively shown in the central and peripheral nervous system of both vertebrates and invertebrates [1]–[14] . Although the nervous system exhibits stochastic variability ( noise ) at all levels ( see [15] for a review ) , it is generally assumed that little random variability affects the AP waveform as it travels from the soma along the axon to the synapse . However , recent understanding of biophysics and experimental methods prompt us to reconsider this common assumption . The AP is mediated by voltage-gated ion channels , which control the flow of ionic currents through the membrane . Thermodynamic fluctuations in voltage-gated ion channels result in probabilistic gating , producing random electrical currents called channel noise [16] . In thin axons , the behaviour of individual ion channels can have significant effects on the membrane potential dynamics due to the higher input resistance of those axons [17]–[19] . Fewer channels sustain AP conduction and fluctuations in individual ion channels have a larger impact on the membrane potential in thinner axons . Faisal et al . [20] have shown that channel noise sets a lower limit to reliable axonal communication at 0 . 08–0 . 1 µm diameter , a general limit matched by anatomical data across species . Above this limit , in axons of 0 . 1–0 . 5 µm diameter , channel noise causes variability in the rising phase of the AP and the resting input resistance of axons . Therefore APs are jittered , shifted , added and deleted in a history-dependent way along the axon [18] . Thus , noise in axons affects the timing of APs and therefore reduces the information capacity of the neural code . Here , we are going to investigate how noise in axons affects the waveform of APs , and produces random variability in the responses of synapses , with implications for information transmission and learning . Attempts at investigating the impact of axonal noise on the synapse have so far been limited to rather large diameter axons ( ≥1 µm diameter ) [21] , [22] . However , many unmyelinated axons are very thin ( 0 . 1–0 . 3 µm diameter [23] ) . Examples include cerebellar parallel fibres ( average diameter 0 . 2 µm [24] ) , C-fibres implicated in sensory and pain transmission ( diameter range 0 . 1–0 . 2 µm [25] ) and cortical pyramidal cell axon collaterals ( average diameter 0 . 3 µm [26] , making up most of the local cortical connectivity [26] ) . The variability of the AP waveform in all these axons is unknown . Basic biophysical considerations suggest that axonal noise sources are bound to introduce fluctuations [20] , [27] in the shape of the travelling AP waveform in thin axons with immediate consequences for synaptic transmission and reliability [28] . Intracellular recordings from such thin axons are difficult to obtain . Extracellular stimulation offers only limited signal resolution and stimulus control , and tiny intracellular volumes limit the application of imaging methods to quantify AP waveforms accurately . This motivated the study presented here which uses biophysically detailed stochastic simulations of travelling APs in thin axons and basic biophysical theory . Our goal is to investigate the mechanisms behind the observed synaptic variability; specifically how much variability can be explained by channel noise in axons . We quantify waveform fluctuations of single propagating APs in terms of standard synaptic efficacy measures , namely AP width and height . We explain how channel noise causes AP waveform fluctuations and show that these fluctuations scale with axon diameter according to an inverse power law , i . e . the finer an axon the bigger the impact . We then investigate the AP waveform fluctuations for propagating spike trains and predict the post-synaptic response variability axonal noise would cause in two ways: 1 . by using models of synaptic dynamics and vesicle release and 2 . by using direct experimental data linking AP waveform to post-synaptic response . Thus , we will be able to estimate the influence of axonal channel noise on synaptic variability .
We find that single APs propagating in central and peripheral nervous systems ( CNS and PNS ) , mammalian and invertebrate axons of up to 1 µm diameter display large random variability in their waveform as they propagate . We visualize this by measuring the AP waveform ( membrane potential versus time ) at various positions along the axon ( Figure 1 . A , B ) and then align the waveforms at the instant of half-peak crossing ( Figure 1 . D ) . As a control , we simulated a deterministic axon , i . e . one that had the same set of biophysical parameters and received the same stimuli but where we modelled the ion channels using deterministic kinetics instead of the corresponding stochastic kinetics [19] , [29] . APs in all our deterministic simulations , starting from the same initial condition and receiving the same trigger input , exhibit no waveform variability across repeated trials . Since the stochastic kinetics of ion channels are the only source of variability given that all other parameters and stimuli are controlled by our simulation , the variability of the travelling AP waveform observed must be due to channel noise and thus entirely random in nature . Crucially , the AP waveform is not only variable across repeated trials with identical stimulus , but also varies as the same AP propagates along the axon . Comparing the variability of the AP waveform in axons with identical biophysical parameters and ion channels but varying axon diameter from 0 . 1 µm to 1 µm , shows that the waveform fluctuations become larger as the axon becomes thinner . This is true for both models of squid giant axons , rat hippocampal interneuron and C-fibre axons . The general structure of the variability profile remains preserved across diameters . The width of the propagating AP varies as it travels down a thin axon in the order of a tenth of a millisecond ( Figure 2 . A , C , E , G ) . Similarly , AP height varies in the order of 1 to 10 millivolts ( Figure 2 . B , D , F , H ) . The variability is more pronounced the thinner the axon is ( Figure 3 . A , B ) . The variations between proximal and distal AP shape are , as expected uncorrelated ( R2<<0 . 2 across all diameters and axons for both AP heights and AP widths ) . This implies that both AP width and height become decorrelated with themselves ( autocorrelation decreases ) and between each other ( cross-correlation decreases ) the further the AP propagates down an axon ( Supplementary Figure S1 ) . To measure how channel noise affects the propagating waveform , one has to track the relevant quantities at corresponding points of the moving AP . To this end , the time series recorded at closely spaced axonal positions ( here , corresponding to a cylindrical membrane compartment of the axon model ) are superimposed after having been aligned at the instant when the membrane potential crosses its half AP peak value . Thus , the individual quantities and their variability at corresponding points of the travelling AP are displayed at corresponding points ( Figure 6 ) . The variability of the waveform has a characteristic structure that is conserved across different axons types and diameters as it is caused by the basic mechanism of the AP itself . The first maximum in waveform variability is reached in the late rising phase of the AP ( between half peak and peak depolarisation , see Figure 4 . A , B ) . The location of this peak is not an artefact of our aligning of APs ( at 50% AP height , c . f . alignment at 20% AP height in Supplementary Figure S2 ) . This first peak is due to fluctuations in the number of opening Na+ channels and Na+ current ( red curves in Figure 6 . B , C ) , as the first peak of Na+ current variability is reached at half-peak membrane depolarisation ( Figure 6 . C , shortly after 0 ms ) . The variability of the depolarising Na+ current accounts for the variations in AP height because the number of Na+ open channels and their inactivation prior to reaching Na+ reversal potential ( the upper limit to AP peak ) determine how much driving current is depolarising the membrane capacitance . K+ channels begin to open later , and thus Na+ channels carry most of the net membrane current in this initial phase of the AP and are responsible for the initial variability ( Standard deviation ( SD ) profile of Na+ current , red curve , and K+ current , blue curve , with net membrane current , green curve , in Figure 6 . E ) . The second , broad peak in waveform variability is reached in the repolarizing phase ( Figure 6 . A and Figure 4 . A , B beginning at 1 ms and increasing up to 2 . 5 ms ) . As the rate of repolarization ( here , <50 mV/ms ) is much slower than that of depolarization ( here , >200 mV/ms ) , variability in the height of the AP waveform translates into much larger changes of AP width . Note , AP width is measured between the up and down crossings of any given membrane potential level , here chosen to be half-peak depolarization . Thus , AP width variability is mainly generated in the repolarizing phase of the AP and caused by a long period of large fluctuations in net membrane current ( Figure 6 . D , between 0 . 75 and 2 . 25 ms ) . Variability is generated initially by K+ current noise and then by Na+ current noise ( Na+ , red , and K+ , blue , in Figure 6 . B , C ) . After K+ channels begin to open in the early repolarizing phase , K+ current fluctuations peak as K+ channel opening probabilities increase and the variance of the number of open channels becomes larger . The increase in variance can be understood , if one considers that a population of ion channels with open probability follows a binomial distribution for the number of open channels . The variance in the number of open channels is given by and thus has a maximum as the open probability approaches 0 . 5 from all channels closed ( ) or all channels open ( ) . By the time the maximum K+ channel open probability is reached ( which is not necessarily for many voltage-gated ion channels ) , electro-motive forces are lower than near AP peak and , membrane potential fluctuations due to K+ currents have consequently lower amplitudes . An equally large and broad maximum in the fluctuations is due to Na+ channel inactivation in the late repolarizing phase for analogous reasons , following a similar binomial argument , when Na+ electro-motive forces are large . Thus , AP height variability ( Figure 5 . A , C , E , G ) is mainly caused by the fluctuating number of open and inactivating Na+ channels during the upstroke of the AP . AP width variability ( Figure 5 . B , D , F , H ) is predominantly caused by the noisy repolarizing phase of the AP , where both K+ and Na+ channels contribute to large fluctuations in the rate of repolarization . Having described how channel noise affects a single AP's waveform , the question arises whether AP waveforms are more variable in spike trains , as APs may influence each other . Using a naturalistic white noise current stimulus protocol [32] ( 1 kHz cut-off frequency , see methods ) , we elicited spike trains for a period of 10 minutes in a 0 . 2 µm diameter axon ( average cerebellar parallel fibre diameter ) using the rat hippocampal interneuron model . Note that interspike intervals and AP triggering currents therefore varied as successive APs were triggered ( f = 40 . 8 Hz±42 . 8 Hz , mean ± SD ) . At the axon's distal end ( measured at approx . 95% of the axon's total length to exclude boundary effects ) waveforms showed considerable variation in the AP shape ( Figure 7 . A ) . Plotting pairs of an AP's width measured at the mid and distal position revealed an uncorrelated structure ( correlation coefficients 0 . 04 and 0 . 03 for AP width and height ) , as in the case of single APs . AP widths measured at half-peak had a coefficient of variation ( CV = SD/average ) of 6% ( 0 . 7 ms±0 . 04 ms ) and AP amplitude ( resting potential to peak ) had a CV of 3% ( 93 . 7 mV±2 . 5 mV ) . AP waveform variability in spike trains was larger than in the case of individual spikes propagating . The standard deviation of change in AP height after propagating for 1 mm in the axon was 3 . 5 mV ( 0 . 05 ms for the width , N = 2000 ) , compared to 1 . 6 mV ( 0 . 04 ms for the width , N = 250 ) for the single spike protocol . The profile of waveform variability ( Figure 7 . A ) peaks close to AP threshold and , at a higher level , in the late repolarising phase . Random waveform variability has matching profiles in the spike train and the single AP protocol as it is caused in both cases by the AP mechanism itself . This illustrates that AP waveform variability is a constantly acting random process , occurring independent of AP initiation or stimulus . Stochastic waveform variability is non-existent in identical simulations where we replace stochastic ion channel models by the equivalent deterministic Hodgkin-Huxley type conductance models . Thus , all axonal variability observed here must result from the effects of the only source of noise modelled – channel noise in Na+ and K+ channels . We have previously shown that the memory of voltage-gated ion channels causes an increased effect on membrane potential noise , affecting the speed of propagation [18] . The same mechanism is also acting on the waveform . We have thus quantified the impact of axonal channel noise on AP waveform variability and explained the biophysical mechanisms acting in thin axons . This previously overlooked effect will only become relevant at the neural circuit and behavioural level if it can influence synaptic transmission . Therefore , we modelled next the synaptic transmission process from arrival of the AP to the post-synaptic response . Synaptic transmission follows a general sequence of events leading to a post-synaptic response . An AP propagates down the axon and causes the opening of voltage-gated Ca++ channels resulting in the influx of Ca++ at the pre-synaptic terminal . Ca++-sensitive proteins trigger the fusion of vesicles , which release neurotransmitters into the synaptic cleft . These transmitters diffuse and trigger the opening of ion channels in the post-synaptic cell , producing a voltage response . Thus , AP waveform variability could perturb post-synaptic responses [33] . We estimate the synaptic impact of waveform variability for spike trains propagating down a 0 . 2 µm diameter axon using two distinct approaches: first , we model the individual stages of synaptic transmission in a synapse driven by our thin axons using biophysical models . Second , we use experimental data relating AP width and height to post-synaptic response amplitude to estimate directly how the variability of the AP would transform into response variability .
The AP that drives the synapse has to travel along an axon , yet the impact of axonal noise sources on the AP waveform in thin axons was , complete propagation failures set aside [44] , not considered in previous studies . Thus , synaptic response reliability and variability [45]–[48] have been in general attributed to mechanisms inside the synapse alone [44] . The results presented here show that in thin unmyelinated axons below 1 µm diameter , commonly found in the CNS and PNS , the travelling waveform of an AP undergoes considerable random variability . This random variability is caused by axonal Na+ and K+ channel noise , which continuously acts during propagation and thus accumulates with distance [18] . The variability of AP width and amplitude , key parameters linked to synaptic efficacy , dramatically increased ( the CV increasing by a factor of approx . 4 , see Figure 5 ) as diameter decreased from 1 µm to 0 . 2 µm . We predict this change by deriving a scaling relationship which is the direct result of the geometry and general biophysics of axons and thus independent of specific channel kinetics or other biophysical parameters [20] . Invariably , channel noise is bound to increase as diameter decreases to the point that it affects the waveform of the AP . Therefore , we can observe the effects of this variability in CNS and PNS axons , in both vertebrates and invertebrates . The range of the waveform fluctuations is about 4 to 6 times the SD , thus we found that AP widths vary by 0 . 1–1 ms in axons between 0 . 2 and 1 µm diameter . AP width fluctuations result mainly from K+ channel noise and inactivating Na+ channels during the repolarising phase of the AP . While Na+ channel noise principally effects AP propagation speed and thus spike timing reliability [18] , K+ channel noise has more impact on waveform variability ( Although variability in Na+ and K+ channels partially compensate each other [20] ) . This fits well with genetic knock-out studies where one type of K+ channels was removed from the central nervous system , and which showed increased temporal response jitter [49] . Activity-dependent modulation mechanisms specific to the pre-synaptic terminal are well-known and provide neurons with means for positive or negative feedback regulation of pre-synaptic Ca++ influx through regulation of the AP width at the synapse [50]–[53] . One example of such modulatory mechanism , the broadening of APs during spike trains due to slow deactivation of A-type K+ channels in mossy fibres has been observed at the level of the synapse [54] , and postulated in the axon [41] . This mechanism can be disrupted by random opening of Na+ channels in the repolarising phase ( which broadens the AP ) or random opening of K+ channels ( which shortens the AP ) , independently of the spiking history . In general , the observable variability in synaptic responses could be due to two sources: ( 1 ) noise and/or ( 2 ) very complex mechanisms that appear random . We can distinguish to which extent these two sources of variability are present at the cellular level , by aggregating the effect of random variability generated by identified molecular stochastic processes ( such as thermodynamic fluctuations in molecular conformations , reviewed in [15] ) . Here , we considered axonal noise as a source of synaptic variability due to channel noise in axons . We modelled a Calyx-of-Held synapse and used data on the Cerebellar Granule-to-Purkinje synapse to estimate the effects of AP waveform noise on synaptic responses in the absence of detailed models for small synapses . Quantitative measurements and models of the mechanistic level of synaptic transmission are limited in small synapses by the technical difficulties to record from thin axon terminals ( <1 µm diameter [27] ) and the need to look at very short range connections ( <500 µm ) . Therefore , we ignored pre- and post-synaptic activity dependent effects – which may reduce the effects of waveform variability – and used simplified synaptic transmission models . Care has to be taken when extrapolating results from these synapses to small CNS synapses [55] , [56] , and extrapolating from any type of synapse to another – even synapses from the same parent axon – may be difficult when details are considered [57] . Bearing that in mind , individual active zones in the Calyx are known to be ultra-structurally similar to those found in small , bouton-like CNS synapses [58]–[60] and the Calyx's functional organization corresponds to a parallel arrangement of several hundred conventional active zones in a single – bouton-like – terminal [61] . Mapping our AP waveform variability data for parallel-fibre like axons onto the empirical relationship between AP width and EPSC amplitude [11] showed a CV of approx . 30% for EPSC amplitude . Other synapses also display this common power-law relationship between AP width and synaptic response , suggesting that axonally induced random variability of the waveform would scale accordingly [5] , [6] , [11] . The detailed allosteric model of vesicle release rate for Calyx-type synaptic transmission produced comparable amplification of the AP waveform noise ( CV increased from 6% to 25% ) . Empirical synaptic response CV is typically between 20 and 60% . In all cases modelled here the extrapolated post-synaptic variability is considerable ( CV 10 to 30% ) and suggests that the observed synaptic variability could be partially explained by axonal noise . Axonal variability will show more impact in synapses placed 1 mm and more down the axon; yet , due to the technical difficulties of finding cell pairs at these distances , their variability is little studied . The Hodgkin-Huxley axon model and related deterministic axon models allow information about the stimulus to be retained in the AP waveform , e . g . stimulus strength is correlated with AP height [62] . It has been shown in vitro that APs triggered and measured at the soma of the same cell can indeed encode information about the stimulus [63] , [64] . Changes in the width of APs , whether due to a depolarized soma [65] or application of glutamate [27] , have been shown to influence post-synaptic potentials ( PSPs ) . Moradmand et al . [31] studied the deterministic transformation of propagating AP waveforms in a paired-pulse framework and showed that the second AP waveform in the pair becomes increasingly stereotyped due to refractory interaction with the first AP . Thus , even in bursts , only the first spike would be the likely candidate to carry stimulus information in the waveform over long distances . Kole et al . [40] have shown that changes in the waveform of APs operated at the AIS are conserved along the axonal arbour for relatively large diameter axons . Here , we show that for both single APs and spike trains , channel noise decorrelates the waveform ( within the limits of the AP's regenerative dynamics ) as the AP propagates even over short distances of less than 0 . 5 mm . Thus , any en-passant synapses along the path of an AP will be driven by randomly differing waveforms and produce different responses ( even if the synapses were identical [66] ) . Synapses innervated by thin axons may have developed mechanisms to circumvent the problem of axonal variability . A simple solution would be to treat an incoming noisy AP waveform as a unitary event . Taking this view , small synapses on thin axons should treat APs as unitary signals and adjust their transduction mechanisms accordingly to be robust to axonal noise effects . This is , in addition to spontaneous APs , another way in which noise constraints affect neural coding [67] . Both axonal and synaptic noise will affect post-synaptic response amplitude variability and vesicle release probability . The above results suggest that the amount of random variability depends on axonal and synaptic morphology . Impedance matching and volume conservation in densely packed CNS tissues suggest that the diameter of axons should be closely related to the size of the synapse ( If the synapse was much larger , the impedence mismatch would cause a drop in the membrane potential , preventing incoming APs from triggering vesicle release . ) Although there exists little data relating axon length and diameter to synaptic morphology and function , one can see that the input resistance of synapses is proportional to their surface area . Similarly , the voltage-sensitivity to an incoming AP increases with the inverse of the squared synaptic diameter . Thus , a smaller synaptic geometry supra-linearly amplifies variations in Ca++ current and influx , and due to the smaller volume , variations in Ca++ concentration as well . Thus , if the properties and kinetics of signal transduction were invariant to synaptic size , an increase in synaptic response variability due to noise in pre-synaptic signal transduction would be expected for smaller synapses . Variability in the waveform of action potentials is known to also affect synaptic latency [68] . However , a careful investigation of this phenomenon requires stochastic simulations of both Ca++ channels and the 5-state vesicle release model . Axons play an important active role for information processing that may be comparable to that of dendritic computation [41] . However , axonal variability has traditionally not been considered as a source of neuronal variability [69] because the AP mechanism was considered highly reliable by extrapolating from classic studies in large ( 3 orders of magnitude larger diameters ) fibres such as squid giant axons [70] . Yet , in densely connected central neural circuits the APs become sensitive to channel noise [18] . The effects of channel noise will inescapably increase non-linearly as diameter decreases due to the very nature of the AP mechanism [20] . Axonal channel noise will affect the reliability ( <0 . 1 µm diameter ) and cause considerable variability to both timing ( <0 . 5 µm diameter ) [18] and , as shown here , the shape of the AP in axons below 1 µm diameter . Thin unmyelinated axons typically innervate large numbers of small CNS synapses [71] and are associated with , and required for , the high level and density of circuit miniaturisation encountered in the cortex and the cerebellum [20] , [72] . In sensory and motor nervous systems , reliability is typically achieved by averaging over many release sites and high release rates . The corresponding large synapses are associated with large axons . However , in the cerebral cortex , hippocampus and cerebellum the dense connectivity within a restricted space limits the diameter of axons , the number of redundant axonal connections and the size of the synaptic contact areas . This makes synaptic transmission prone to the effects of axonal channel noise in thin axons innervating small synapses . The results presented here prompt careful experimental consideration , because paired-cell measurements and optical methods do not offer the control and resolution necessary to determine the source of PSP variability . More generally , we show how molecular noise sources can explain observed variability at higher levels of biological organisation .
We used two sets of simulation protocols . In the first protocol , 0 . 1 µm , 0 . 2 µm , 0 . 3 µm , 0 . 5 µm diameter ( 1 cm long ) and 1 µm diameter ( 2 cm long ) axons were stimulated in a single spike per trial framework ( N = 250 trials per diameter ) to allow for fast parameter exploration . In the second protocol , we simulated 10 minutes long spike trains . To this end , a 0 . 2 µm diameter ( 2 mm long ) axon was stimulated with a zero-mean white noise current ( SD = 0 . 01 nA , 1 kHz corner frequency ) injected at the proximal end . Membrane properties were set to Ra = 70 Ωcm , Rm = 20000 Ωcm2 , typical for cortical cells [81] . All axons had a resting potential of −65 mV . After visual inspection of the data , we used a threshold discriminator detecting AP height and aligned their waveforms at the rising half-peak potential crossing time . We measured voltage-traces of the AP waveforms at regular intervals ( typical distance 10% of total axon length , see Figure 1 . A ) between 5% and 95% of the axon's length ( 0% being the axon's proximal end ) to avoid measuring stimulus artefacts or boundary effects and to measure the evolution of the AP shape along the axon . The height and width of APs were defined according to Figure 1 . C . We estimated the impact of action potential waveform variability on synaptic transmission using two approaches . First , synaptic transmission was modelled using data and deterministic models from the Calyx of Held synapse ( reviewed in [34] ) . We drive the Calyx- of-Held synapse with noisy spike train waveforms directly ( voltage clamping the synapse ) to circumvent any potential issues of impedance mismatch . We then compute Ca++ currents evoked by the AP waveform by integrating the dynamics of a Hodgkin-Huxley type conductance-based Ca++ channel model of Calyx synapses [82] . We describe the Ca++ channel behaviour using a conductance-based Hodgkin-Huxley type model with two identical gating particles ( denoted m ) with channel opening probability . The corresponding rate functions are and , with dynamics [83] . Here we modelled the voltage-gated Ca++ channel deterministically , and calculated the waveform of the incoming Ca++ current using a reversal potential of . This Ca++ current model simplifies the heterogeneity in both the biophysical and the pharmacological properties of Ca++ currents found in Calyx-type synapses [84] , [85] , but was shown to capture sufficient detail for quantitative modelling of synaptic transmission [82] . The transient encountered by vesicles in the proximity of Ca++ channels is shown to follow a time course similar to that of the Ca++ current [86] , [87] , i . e . the Ca++ concentration rapidly declines due to effects such as buffering [88] . We use the dynamics of Ca++ channels and modify them slightly so that the rise time is conserved , but the width at half-height becomes approx . 100 µs longer [86] . We then scaled the resulting waveform to have a peak of approx . 12 µm . This value was chosen based on an approximate reading of figure 6 . B in [89] . The resting was 50 nm . Finally , we modelled the impact of on transmitter release using an allosteric “5 state” model [90] , which allows us to derive the instantaneous vesicle release rate ( Figure 9 ) . Synaptic transmission dependence on the AP waveform was also modelled using experimental data for the much smaller rodent cerebellar Granule cell-to-Purkinje cell synapse . In this synapse the width of the pre-synaptic AP waveform , pre-synaptic Ca++ entry and the resulting post-synaptic currents were directly measured [11] . To obtain an estimate of how AP waveform variability would affect this synapse , we passed the simulated APs' widths through the experimentally characterised relationship between postsynaptic response and AP width . We then computed the variability of the post-synaptic response over all APs . | The fundamental signal of the nervous system is the action potential: an electrical spike propagated along neurons and transmitted between them via synapses . Once triggered , action potentials are generally assumed to be robust to noise , and the variability observed at all levels of the nervous system is primarily attributed to synapses . However , this view is based on data from classically studied axons , which are very large compared to the average diameter of axons in the mammalian nervous system , and even larger when compared to the thinnest axons . As the effects of thermodynamic noise affecting the proteins responsible for the initiation and propagation of action potentials are much bigger in thin axons , the assumption does not necessarily hold for very thin axons . We show that the action potentials waveform in thin axons is subject to random variability . Fluctuations in this waveform result in fluctuations in synaptic ionic currents , and account for a significant portion of the variability observed at the synapse . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"computational",
"neuroscience",
"neuroscience",
"biology",
"and",
"life",
"sciences",
"computational",
"biology"
] | 2014 | Axonal Noise as a Source of Synaptic Variability |
Phleboviruses transmitted by sandflies are endemic in the Mediterranean area . The last decade has witnessed the description of an accumulating number of novel viruses . Although , the risk of exposure of vertebrates is globally assessed , detailed geographic knowledge is poor even in Greece and Cyprus where sandfly fever has been recognized for a long time and repeatedly . A total of 1 , 250 dogs from mainland Greece and Greek archipelago on one hand and 422 dogs from Cyprus on the other hand have been sampled and tested for neutralising antibodies against Toscana virus ( TOSV ) , Sandfly fever Sicilian virus ( SFSV ) , Arbia virus , and Adana virus i . e . four viruses belonging to the 3 sandfly-borne serocomplexes known to circulate actively in the Mediterranean area . Our results showed that ( i ) SFSV is highly prevalent with 71 . 9% ( 50 . 7–84 . 9% depending on the region ) in Greece and 60 . 2% ( 40 . 0–72 . 6% ) in Cyprus; ( ii ) TOSV ranked second with 4 . 4% ( 0–15 . 4% ) in Greece and 8 . 4% ( 0–11 . 4% ) in Cyprus; ( iii ) Salehabad viruses ( Arbia and Adana ) displayed also substantial prevalence rates in both countries with values ranging from 0–22 . 6% depending on the region and on the virus strain used in the test . These results demonstrate that circulation of viruses transmitted by sand flies can be estimated qualitatively using dog sera . As reported in other regions of the Mediterranean , these results indicate that it is time to shift these viruses from the "neglected" status to the "priority" status in order to stimulate studies aiming at defining and quantifying their medical and veterinary importance and possible public health impact . Specifically , viruses belonging to the Sandfly fever Sicilian complex should be given careful consideration . This calls for implementation of direct and indirect diagnosis in National reference centers and in hospital microbiology laboratories and systematic testing of unelucidated febrile illness and central and peripheral nervous system febrile manifestations .
In the Old world , phleboviruses ( Bunyaviridae family , Phlebovirus genus ) transmitted by phlebotomines consist of three species or antigenic groups , namely Sandfly fever Naples , Salehabad , and Sandfly fever Sicilian serocomplexes . Each species contains several viruses among which Naples , Sicilian and Toscana virus cause 3-day fever , commonly called sandfly fever in humans; Toscana virus ( TOSV ) causes of neuroinvasive human infections such as meningitis and encephalitis [1] . In Greece , outbreaks of sandfly fever were reported in Athens among the local population , and among American , British and German troops during World War II [2] . Sandfly fever has been described in Cyprus and Greece with both sporadic cases and epidemics [3–7] . In both countries , the high rates of antibodies observed in seroprevalence studies indicate that viruses belonging to Sandfly fever Naples and Sandfly fever Sicilian serocomplexes are transmitted by local sand flies to human populations [7–10] . Sandfly fever Cyprus virus ( SFCV ) , closely related to Sandfly fever Sicilian virus ( SFSV ) , was isolated during a large outbreak of sandfly fever in Swedish United Nations troops stationed in Cyprus; few cases were also caused by TOSV [5] . In Greece , in recent sporadic cases of meningitis , ( i ) TOSV RNA was detected in the CSF of a patient [11] , and ( ii ) viral RNA corresponding to Adria virus , a novel virus belonging to the Salehabad species , was also identified in the CSF [12 , 13] . To date , SFSV or another SFS-like virus have been neither isolated nor detected by molecular techniques in Greece . During the last decade , field-to-laboratory integrated studies associating virologists , parasitologists and entomologists have discovered several new phlebotomine-borne phleboviruses; thus there is an increased diversity in each of the three aforementioned species or serocomplexes [14] . Although the pathogenicity of most of these newly discovered viruses remains unknown , they are sympatric with recognized pathogenic phleboviruses [15–17] . Because several viruses of the same serocomplex co-circulate in various regions , interpretation of seroprevalence studies requires using techniques that hold the capacity to discriminate between these antigenically-related viruses . To the best of our knowledge , all studies performed in Greece and Cyprus used either ELISA or IFA tests , which are notoriously prone to cross-reactivity between viruses belonging to the same serocomplex [7–10] . To conduct our nation-wide ( mainland Greece , Greek islands , Cyprus ) seroprevalence study in dogs , we selected neutralisation tests which is the most discriminant assay as previously reported in Algeria , Tunisia , Turkey and Portugal [15 , 18–20] . Although virus exposure to viruses may be quantitatively different in humans and dogs , because of different feeding preferences of phlebotomines , recent studies suggest that virus circulation can be estimated using either human or dog sera since dogs live in close proximity to humans and are readily infected by these viruses [15 , 18 , 20 , 21] . In our study , dog sera were tested for the presence of neutralising antibodies against TOSV , SFSV , and two viruses belonging to the Salehabad complex ( Arbia virus isolated in Italy and Adana virus isolated in Turkey ) .
From 2005 to 2010 , a total of 422 and 1 , 250 dog sera were collected in Cyprus and Greece , respectively . These sera originated from the five districts of Cyprus and 32 prefectures belonging to 12 regions of Greece ( Table 1 ) . Veterinarians were asked to provide dog samples from animals visiting their clinic for any reason: vaccination , hair cut , nail cut , deworming , general check up , treatments , and other purposes , without discrimination . The animals were examined clinically and peripheral blood samples ( without EDTA ) was collected , after the written consent of the owner , and questionnaires with personal , epidemiological , and clinical data for each dog were completed . Only domestic dogs that were raised in the area were considered for the study . The domestic dogs were included after owners’ informed consent . Information regarding age , sex , was obtained after interviewing dog owners ( Table 1 ) . Each dog was examined clinically by the veterinarian and blood samples were collected . Whole blood samples were collected ( 1–2 mL ) by cephalic or jugular venipuncture and serum was separated by centrifugation and stored at −20°C . Data on the region , gender , and age ( distributed according to 3 classes: young 6–11 months , adult 12–83 months , senior ≥ 84 months ) were recorded . This study was ethically approved by the Institutional Animal Care and Use Committee of the University of Crete Medical School and conform with the European Union Directive 2010/63/EU regarding use of animals and biological specimens in research , as well as the relevant Hellenic legislation ( Presidential Decree 160/91 , under the Code Numbers 31 EE 05 , 31 EPR 04 and 31EP 020 ) . Written informed consent was obtained from the dog owners , according to the aforementioned national legislations . Sera were tested by the virus microneutralisation assay ( MN ) , described for phleboviruses [19] in parallel for 3 distinct sandfly-borne phleboviruses: ( i ) TOSV strain MRS2010-4319501 ( TOSV belongs to the Sandfly fever Naples virus species or complex ) [22] , ( ii ) SFSV strain Sabin [23] , ( iii ) Arbia-like virus strain T131 ( Salehabad species or complex ) , and ( iv ) Adana virus strain 195 ( Salehabad species or complex ) [15] . Briefly , two-fold serial dilutions from 1:10 to 1:80 were prepared for each serum and a volume of 50μL was pipeted into 96-well plate . Viruses were titrated in Vero cells ( ATCC CCL81 ) . A volume of 50 μL containing 1000 TCID50 was added into each well except for the controls that consisted of PBS . A volume of 50 μL of EMEM medium enriched with 5% fetal bovine serum , 1% Penicilin Streptomycin , 1% L-Glutamine 200 mM , 1% Kanamycin , 3% Fungizone , was added to each well of the controls . The plates were incubated at 37C° for one hour . Then , a 100μL suspension of Vero cells containing approximately 2 x105 cells/mL of EMEM medium ( as previously described ) was added to each well , and incubated at 37C° in presence of 5% CO2 . The first row of each plate contained control sera diluted 1:10 and Vero cells without virus . After 5 days ( Toscana and Arbia virus ) and 7 day ( Sicilian and Adana virus ) , the microplates were read under an inverted microscope , and the presence ( neutralization titer at 20 , 40 , 80 and 160 ) or absence ( no neutralization ) of cytopathic effect was noted . Cut-off value for positivity was set at titre ≥ 20 [15 , 18 , 20 , 21] . Due to insufficient volume in Greek samples , ADAV was used for testing Cyprus specimens , only . Dog seroprevalence for each virus was estimated for each prefecture and mapped using the geographical information system software ( GIS , Redlands , CA; ArcGIS 10 ) . The chi-square or Fisher’s exact tests were used to compare percentages of positivity among categories of the same independent variables and also the total prevalence of each virus . A p value < 0 . 05 was considered as statistically significant . Analyses were performed with StatLib and SPSS® 21 software for Windows .
In Greece , a total of 1 , 250 sera ( 540 male and 710 female , sex ratio 0 . 76 ) were collected . The median age was 36 months ( range: 3–216 ) . The sera were collected from 32 prefectures , but owing to the variability in the number of collected sera from each prefecture ( range: 1–410 ) , the sera were grouped into 12 regions . Of these 12 regions , Thessaly was not included in the analyses because it consisted of 1 serum only . For the other 11 regions , the number of sera ranged from 14 to 410 . In Cyprus , a total of 442 sera ( 202 male and 240 female , sex ratio 0 . 84 ) were collected . The median age was 36 ( range: 3–144 ) . They consisted of 67 , 27 , 97 , 74 , and 177 sera collected from the districts of Ammochostos , Larnaca , Limassol , Nicosia , and Paphos , respectively . The two dog populations had the same median age ( 36 months ) and a similar sex ratio ( 0 . 76 vs 0 . 84 ) . Characteristics of the dogs and their geographic origin are presented in Table 1 . Results for domesticated dogs living in Greece and in Cyprus are presented in Table 2 and Table 3 , respectively . Toxic activity in the serum was detected in 65 and 73 sera from Greece and Cyprus , respectively; therefore calculations were done on the basis of 1 , 185 and 369 sera of Greece and Cyprus , respectively . As previously shown [26 , 27] , a cut-off titre ≥ 20-when used for 1000TCID50 inoculum , is equivalent to a cut-off titre ≥ 40 when a 100 TCID50 is used [28] . In Greece , a much higher rate of SFSV-NT-Ab was observed ( 71 . 9% ) compared with 4 . 4% and 2 . 6% for TOSV and ARBV , respectively ( Table 1 and Fig 1 ) . Similar results were observed in Cyprus where SFSV-NT-Ab was present in 60 . 2% of the dog sera , whereas 16 . 3% , 8 . 4% and 5 . 4% of sera were positive for ADAV , TOSV and ARBV , respectively ( Table 2 and Fig 2 ) . The distribution of TOSV positive sera is quite homogenous within the studied regions ( p = 0 . 248 , p = 0 . 094 ) . There is no significant difference according to the sex of the dogs . In contrast , it appears that the prevalence increases with the age , although it is not statistically significant even when the results of dogs from Greece and Cyprus are merged ( 3 . 7% / 5 . 3% / 7 . 7% , p = 0 . 3 ) . In both countries , a statistic association was found between SFSV prevalence and geographic area . In Cyprus , none of the sera containing ARBV-NT-Abs were also positive for ADAV-NT-Abs and vice versa; this demonstrated that there is no cross-reactivity through MN assay between these two viruses despite the fact that they belong to the same serocomplex . Exposure to ARBV and ADAV is significantly associated with the age with higher rates observed in older dogs . In contrast , the high rates of SFSV-NT-Abs were observed in the 6-11-month age class in Cyprus and Greece .
At the outset of our study , the following data were available for Greece: ( i ) there was no serological data in domestic animals ( cattle , goats , sheep , dogs or cats ) for any phlebovirus transmitted by sand flies , ( ii ) in the 1970's , 13 . 1% of 38 adults living in Crete had NT-Abs against Naples virus , and 24 . 7% and 8 . 5% of 632 human sera from Athens inhabitants had NT-Abs against Naples virus and Sicilian virus , respectively [29]; ( iii ) more recent studies reported various rates of TOSV IgG using ELISA and/or IIF tests in continental Greece as well as in the Ionian and Aegean islands [8–10]; ( iv ) TOSV RNA ( belonging to the lineage C ) was detected in the CSF of a patient with meningitis [11] , and ( v ) viral RNA corresponding to Adria virus , a novel virus belonging to the Salehabad species , was also identified in the CSF of a patient with meningitis [12 , 13] . To date , SFSV or another SFS-like virus have been neither isolated nor detected by molecular techniques in Greece . At the outset of our study , the following data were available for Cyprus: ( i ) there was no serological data in domestic animals ( cattle , goats , sheep , dogs or cats ) for any phlebovirus transmitted by sand flies; ( ii ) first evidence of the presence of TOSV , Naples and Sicilian viruses were observed in Swedish soldiers of the United Nations force [3] through detection of NT-Abs and isolation of strains of Naples and Sicilian viruses [30]; ( iii ) NT-based seroprevalence results showed that Naples , Sicilian , and TOSV were endemic with respective rates of 57% , 32% and 20% [7]; ( iv ) investigation of a second outbreak in Greek troops stationed in Nicosia of which almost 50% developed febrile syndrome had resulted in isolation of Cyprus virus ( SFCV ) , closely related but distinct from Sicilian virus although belonging to the SFSV serocomplex [4 , 5] . The recent discovery of several new viruses belonging to the three species associated with phlebotomines in the Old World has raised questions about the viral strain currently circulating in the two regions . Since broadly cross-reactive techniques such as ELISA and IIF are not capable to distinguish between viruses belonging to the same serocomplex , we decided to use microneutralisation assay using viral strains or surrogates which presence had been assessed in Greece and Cyprus . Indeed , we consider it valid to use SFSV as a surrogate for SFCV ( isolated in Cyprus ) and other SFSV-related viruses because amino acid distances observed between the proteins that elicit neutralizing antibodies ( Gn and Gc ) are well within the acceptable range , ie <5% different for SFSV and SFSV-related viruses[25 , 31] . Thus , neutralising antibodies are unlikely to discriminate between closely-related SFSV isolates . Since collecting human sera displaying a large geographic distribution was challenging , we decided to use dog sera; indeed , dog sera can be good surrogates for the following reasons: ( i ) dogs are readily infected with phlebotomine-borne phleboviruses which are human pathogens [18 , 20 , 31 , 32]; ( ii ) domestic dogs live in close contact with humans and therefore are exposed to sandfly bites , although different feeding preferences of sand fly species have to be considered [31] . The largest amount of data available on dogs , at the outset of this study , concerned TOSV , which observed rates ( 4 . 4% in Greece and 8 . 4% in Cyprus ) are in the same order of magnitude as those recently reported in dogs in Tunisia ( 6 . 8% , [20] ) , in Algeria ( 4 . 3% , [18] ) , in France ( 3 . 9% , [21] ) and in Portugal ( 6 . 8% [31] . Because all these studies measured neutralising antibodies against TOSV , their results are comparable and they reflect local circulation of TOSV only , not other viruses belonging to the SFNV complex . Together these results demonstrate that TOSV can readily infect dogs . Exposure level of dogs and humans may be drastically different in the same area as previously shown in Tunisia where MN-based seroprevalence rates were respectively at 6 . 8% in dogs compared with 41% in humans [20] . In the present study , dogs living in the Ionian island of Corfu showed a much lower seroprevalence compared to the human population living in the same island ( 3 . 9% vs 51 . 7% ) [8]; however , in this case the techniques used were different; in the human study , ELISA/IIF detected not only TOSV IgG but also IgG raised after infection with other viruses belonging to the Sandfly fever Naples serocomplex in which 6 new viruses were described during the last decade ( Arrabida , Fermo , Granada , Massilia , Punique , Zerdali ) [16 , 33–37] in addition to Naples virus ( a proven human pathogen ) and Tehran virus . Although none of these viruses were detected or isolated in Greece or Cyprus , the presence of one of these 6 recently discovered viruses or of a yet to be discovered virus may account for these apparently discrepant results . Last , these techniques do not hold the same sensitivity [38] . The same explanation applies for discrepancies observed between high rates of ELISA/IFA TOSV IgG reported in Aegean islands ( 17 . 6% , 11 . 5% , 20% , 22% and 34 . 7% for Lesbos , Rodos , Siros , Crete and Evia , respectively ) [9] compared with our findings: 5 . 3% in north Aegean islands ( Chios and Lesbos ) , 0% in south Aegean islands ( Rodos , Siros and Santorini ) , 1% in Crete and 4 . 4% in Evia ( Stere Hellas ) ( Table 2 ) . In Central Macedonia , 7 . 3% of dog sera contained TOSV-NT-Abs , which is in agreement with reported cases of human infections [38 , 39] and a recent study showing that TOSV and/or antigenically related viruses are circulating extensively in the area [10] . It is worth underlining that , despite using the same technique , discrepant prevalence rates were also described between dogs and humans in Tunisia [19 , 20] . Therefore , it is difficult to compare results of serological studies performed with different techniques . When using the same technique , results observed in humans and in dogs consistently detected TOSV although they varied considerably quantitatively; therefore dogs can serve as sentinel for humans and vice versa for assessing the presence of TOSV although quantitative results must be interpreted carefully . The absence of cross-protection between ARBV-NT-Abs and ADAV-NT-Abs confirm previous data from Turkey [15] . Accordingly , cumulative percentage of viruses belonging to the Salehabad species is 21 . 7% . This is congruent with the results observed in Adana , southern Anatolia , Turkey where domestic animals were presenting high rates of NT-Ab against viruses belonging to the Salehabad serocomplex [15] . Tesh et al [29] did not detect NT-Abs against Salehabad virus ( SALV ) in human populations suggesting that SALV was not infecting humans . In contrast , NT-Abs against Medjerda Valley virus were described in 1 . 35% ( 14/1260 ) of human sera collected from the general population living in Northern Tunisia [24] . This suggests that at least some viruses belonging to the Salehabad complex can infect humans and other vertebrates . Interestingly , Adria virus RNA has been detected in the CSF of a Greek patient presenting with meningitis [13] but was never isolated precluding serological studies aiming at defining the possible impact of this virus in the region and beyond . However , molecular detection of Adria virus in Albania ( in sand flies ) and in Greece ( in human ) suggests that its distribution might cover a large geographic area . This constituted the first direct evidence supporting human pathogenicity of a virus belonging to the Salehabad virus complex . Isolation of Adria virus is now a priority in order to pursue the studies using neutralization-based serological studies in humans and animals . Very high rates of SFSV-NT-Abs were observed in Cyprus and Greece . In the latter , rates were consistently above 50% ( range 50 . 7–84 . 9% ) ; in Cyprus , rates were above 40% ( range 40 . 0–72 . 6% ) except in Ammochostos ( 26 . 3% ) . The extremely high prevalence rates observed with SFSV in young dogs show that this virus continues to circulate very actively in these regions , and beyond as recently described in dogs from Tunisia ( 50 . 8% , [20] ) and in Portugal ( 38 . 1% , [31] ) . In both countries , a statistic association was found between SFSV prevalence and geographic area . The differences of prevalence depending upon the region may be due to the geographic and climatic characteristics of these regions which affect the distribution , proliferation and abundance of phlebotomine vectors of SFSV . Analysis of the questionnaires did not identify any clinical manifestations such a fever and/or neurological signs during the past weeks and months in the SFSV-positive dogs . This tends to suggest that SFSV is not or mildly affecting dogs during the viremic period . Whether or not dogs can play a role as reservoir in the natural cycle remains to be studied . To do so , experimental studies to understand the virus kinetics are necessary . Also , studies aiming at the identification of viremic domestic dogs should be planned in high prevalence areas . The massive prevalence of SFSV-NT-Ab observed in our study is not unexpected and is congruent with entomological and human data in the literature: ( i ) isolation of Corfu virus on the eponymous island from Phlebotomus neglectus [17]; ( ii ) SFSV IgG detected by IFA in human sera in Northern Greece ( Macedonia ) , Central Greece ( Evritania and Larisa ) , North–Western Greece ( Epirus ) , and Corfu Island; ( iii ) detection of Chios virus , SFSV-like , in Chios island; ( iv ) sandfly fever epidemics were reported in Swedish UN soldiers and Greek soldiers in 1984 and 2002 , respectively [3–5]; ( v ) a high attack rate ( 63% ) in tourists hosted in Cyprus for a short period [6]; ( vi ) a 32% prevalence rate of SFSV IgG in Cyprus native population [7] . In contrast with the two other serocomplexes which display an important range of genetic distance between their respective members , Sicilian virus strains are genetically and antigenically much more closely related [14 , 16]; therefore , exposure to different SFSV strains ( Italy , Turkey , Cyprus , Greece , Ethiopia ) can be measured by using the prototypic Italian strain . Despite high rates of antibodies in humans and other vertebrates and successive outbreaks in Italy , Cyprus , Greece and Ethiopia [3 , 5 , 23 , 40] , SFSV remains a neglected pathogen , almost never included in diagnostic algorithms despite repeated and accumulating evidence of its involvement in febrile syndromes and in neuroinvasive infections . In conclusion , this study indicates that ( i ) sandfly-borne phleboviruses belonging to 3 distinct genetic and antigenic groups are widely spread and co-circulate; ( ii ) dogs represent excellent qualitative sentinels for virus transmitted by sandflies and further studies must be done to estimate the role of dogs in the dynamics of transmission , and whether they play a role as reservoir hosts in the natural cycle of these viruses . Since several of these viruses are proven human pathogens , our results plead for performing similar studies using human sera to identify geographic hot spots . The increasing number of sequence data for these phlebotomine-borne phleboviruses now enables to design and develop real-time molecular assays . The improved diagnostic toolbox will allow to investgate the medical impact of these viruses in patients presenting unexplained febrile illness and neuroinvasive infections . | Phleboviruses transmitted by sandflies are endemic in the Mediterranean basin . An increased number of new viruses was described during the last decade . However , levels of exposure of human and animal populations are poorly known . A total of 1 , 250 dogs from Greece and 422 dogs from Cyprus were tested for the presence of neutralising antibodies signing previous infection with selected phleboviruses representing the 3 serological complexes known to be present in the Old World: Toscana virus ( TOSV ) , Sandfly fever Sicilian virus ( SFSV ) and Salehabad viruses ( Arbia and Adana viruses ) . Our data showed that ( i ) SFSV is largely predominant with infection rates higher than 50% , ( ii ) TOSV is widely distributed with 4 . 4% and 8 . 4% in Greece and Cyprus , respectively , and ( iii ) that viruses belonging to the Salehabad serocomplex should be further studied for their capacity to cause human disease in view of prevalence rates in dogs up to 22 . 6% . These findings confirm that dogs can be considered as excellent sentinels for sandfly-borne phleboviruses . The results also underline the importance to study the role of SFSV in humans and may lead to the set-up diagnostic tests for patients presenting unexplained febrile illness and neuroinvasive infections . Further studies are also needed to define whether these viruses cause diseases in dogs . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"animal",
"types",
"greece",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"geographical",
"locations",
"microbiology",
"vertebrates",
"sand",
"flies",
"pets",
"and",
"companion",
"animals",
"dogs",
"animals",
"mammals",
"viruses",
"signs",
"and",
"symptoms",
"antibodies",
"insect",
"vectors",
"zoology",
"immune",
"system",
"proteins",
"proteins",
"medical",
"microbiology",
"epidemiology",
"cyprus",
"microbial",
"pathogens",
"disease",
"vectors",
"people",
"and",
"places",
"biochemistry",
"diagnostic",
"medicine",
"asia",
"fevers",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"europe",
"amniotes",
"organisms"
] | 2016 | Seroprevalence of Sandfly‐Borne Phleboviruses Belonging to Three Serocomplexes (Sandfly fever Naples, Sandfly fever Sicilian and Salehabad) in Dogs from Greece and Cyprus Using Neutralization Test |
Rabies is a neglected zoonotic disease . Given the low incidence , apart from the existing reporting syst , there is a need to look for other means of case detection strategies for rabies . Contact tracing is one such method to efficiently capture information . To find out the rabid status of biting animal through contact tracing and to determine health seeking behavior of the bite victims . An exploratory study using contact tracing was conducted during the first quarter of 2017 in villages coming under three Public Health Centers . The households of the bite victims were visited and details of rabies exposure obtained from the bite victim/ adult responsible respondent using a standardized questionnaire . A total of 69 dog/cat bite cases were identified . 69 . 5% of bites were by stray dogs . 97 . 1% bite victims had Category III bites . Only 4 . 5% bite victims had taken PEP . 70 . 1% of animal bite cases were administered ARV . Only 7 . 2% bite victims had exposure to probable rabid animals . All dog bite victims were alive after 3 months of follow up . Contact tracing was successful in case detection of probable rabid animal exposures and suitable for a period of one year .
Rabies is a 100% fatal viral zoonotic disease . [1] However , Rabies can be prevented if , health care providers are able to identify the type of exposure , categorize the wounds and provide post exposure prophylaxis ( PEP ) as early as possible . [2 , 3] South East Asia region accounts approximately for 60% of human rabies deaths in the world . An estimated 20 , 000 rabies deaths ( approximately 2/100 , 000 population ) and 17 . 4 million exposures to animal bite occur every year in India . [4] The data on incidence of human and canine rabies is currently not known and needs to be updated . In reality , the burden of rabies is usually not captured by the health system due to varied reasons . Moreover , Rabies is not a notifiable disease in India and acts as a surveillance barrier for measuring the burden of disease . Also , the laboratory-based animal rabies surveillance program to measure the burden is non-existent in many regions of the country . Human and animal rabies cases are under reported in most of the developing countries due to ineffective rabies surveillance methods . [5] To understand the national burden of rabies , estimation methods must be periodically conducted . [6]In India , the populations most at risk from rabies exposures are living in rural areas , in poor , marginalized communities where surveillance and reporting systems are weak . The need of the hour is measurement of bite victims bitten by a rabid dog and having access to PEP . Contact tracing provides an efficient method of documenting such information . Contact tracing consists of sequential steps of data collection and investigation , revealing different aspects of health seeking behavior , treatment cost and health outcomes . Contact tracing is finding everyone who comes in direct contact with rabies exposure usually occurring through dogs and helps in targeting at risk population [7] In India , A number of changes have taken place over the years including introduction of intra dermal rabies vaccine ( IDRV ) , abolition of nerve tissue vaccine , availability and accessibility to rabies immunobiologicals and better awareness about rabies in the population . [8]Hence it is assumed that the rabies burden would have come down . In this background , an attempt has been made to find out the rabid status of the biting animal through contact tracing in a rural area , to determine the health seeking behavior of the bite victims and to recommend measures for rabies prevention .
The local government primary care provider ( Anganwadi Centre ) in each village was the starting point of contact tracing . The female staff was the first informant , as she is the front line primary care worker and would have had basic information on the morbidities reported . Through her cooperation , the initial process of identifying the index bite victims was started , which would have otherwise been difficult for the investigator in terms of acceptability and feasibility . The bite victims were people who had been exposed to animal bites in the last one year from the date of survey . Ideally , the bite victims details should have been got from the health center . However , the address and village would have been difficult to get as they are usually not recorded in the health center . Hence , the index case was the starting point of contact tracing in the village . Majority of the bite victims were interviewed within 6 months of the exposure , the minimum and maximum duration of exposure was one week and 11 months from the date of survey . The bite victims were followed up for a period of three months to know the outcome i . e alive/dead . The household of each bite victim was visited by the investigators and detailed information regarding biting animal , availability of animal , rabies exposure , PEP , rabies-related deaths following animal bites and clinical signs of rabies in the biting animal , etc was obtained using a standardized semi-structured questionnaire . The bite victims were the respondents in majority of the interviews and when they were not available , the head of family /adult responsible respondent were interviewed . For contact tracing of animal bite , the respondents were asked about the index biting animal having bitten other animals or humans and if they had come across people who had been bitten by a dog/cat . [10]Subsequently , the households of other bite victims were visited and detailed information on exposure was obtained . This procedure of contact tracing ( snowball sampling technique ) was repeated until all probable exposures were identified . To maximize the efforts to trace all bite victims , information was also obtained from villagers , Accredited Social Health Activist ( ASHA ) workers , formal and informal leaders . Additionally , the investigators searched for the availability of the biting dogs/cat during the survey . The bite victim must have been a resident of the village for a minimum of six months . Subjects bitten by animals not known to cause rabies in humans were excluded from the study . Finally , a total of 69 dog/cat bite victims were selected through contact tracing during first quarter of 2017 covering 17 villages .
A total of 69 bite victims were followed up . The age range of bite victims was 3 to 89 years respectively . The median age of bite victim with Inter Quartile Range ( IQR ) was 17 ( 9 , 18 ) years . 52% of the bite victims were males . 33 animals ( 32 dogs & 1 cat ) were responsible for the exposure among 69 bite victims . 69 . 5% of the bites were by stray dogs . There was one ( 1 . 4% ) event of a dog having bitten a cat . There were no wild animal bites . Information regarding outcome ( alive , dead , sick , unknown ) of all the 33 dogs/cat was available and given by the bite victims . Out of the 69 , 5 ( 7 . 2% ) bite victims had exposure to 3 probable rabid dogs and 1 probable rabid cat . The remaining bite victims were exposed to non cases . All 4 ( 100 . 0% ) probable rabid dogs/cat had died within 10 days . The clinical signs observed in the probable rabid dogs/cat were hypersalivation in 3 ( 75 . 0% ) and aggressive behavior in 3 ( 75 . 0% ) . Laboratory confirmation of rabies diagnosis was not done in any of the dogs/cat that had died . All the 69 ( 100% ) bite victims were alive at the end of three months of follow up ( Fig 2 ) . There was no relationship between any two dog bites within the village and between adjacent villages . Each of the bites had occurred in different time periods as sporadic event . In majority of the villages , one dog was responsible for each exposure . A maximum of 7 persons were bitten by a single dog . Only in one village , 4 dogs were responsible for 4 different exposures . An example of contact tracing in village X is given in Fig 3 . 67 ( 97 . 1% ) bite victims had category-III exposures and 2 ( 2 . 9% ) had category-II exposures . 3 ( 4 . 4% ) category-III exposures had received PEP ( Rabies Immunogobulin & Anti Rabies Vaccination ) , 46 ( 68 . 7% ) had received Anti rabies Vaccination ( ARV ) alone and 18 ( 26 . 9% ) had not sought any treatment . 1 ( 50 . 0% ) category-II bite victims had received ARV and 1 ( 50 . 0% ) had not sought any treatment . Among those who had received ARV ( PEP included ) , 31 ( 62% ) bite victims had completed full course of ARV ( 5 doses ) and 19 ( 38% ) had incomplete Anti Rabies Vaccination . 14 ( 73 . 6% ) bite victims informed that doctor /health worker did not advice and 5 ( 26 . 4% ) said busy as reason for not completing ARV . Among the 5 bite victims exposed to probable rabid dogs/cat , the age of the victims were 14 , 16 , 17 , 32 , and 60 years . 3 ( 60 . 0% ) were males and 2 ( 40 . 0% ) were females , all the exposures were unprovoked in nature , 2 ( 40 . 0% ) bites were over the hand , 1 ( 20 . 0% ) on the back , 1 ( 20 . 0% ) forearm and 1 ( 20 . 0% ) was on the leg . 2 ( 40% ) bite victims had washed the wound with soap and water , 1 ( 20% ) washed only with water , 1 ( 20% ) had applied antiseptic and 1 ( 20% ) did not do anything . Two ( 40% ) probable rabid bite victims had completed ARV ( 5 doses ) , 1 ( 20% ) had incomplete Anti Rabies Vaccination , while the other 2 ( 20% ) had not sought PEP and reason being ignorant of ARV . An average of one dose of ARV was taken by non cases and average of three doses of ARV by probable rabid animal bite victims . The average number of people bitten by probable rabid dogs was one and average number of people bitten by non rabid dog was two . None of the bite victims had received first aid by the village primary care provider and all of them were referred to higher centers . 26 ( 52% ) bite victims had visited the private health care provider for availing treatment and 10 ( 20% ) had visited both government and private health care centers . The average cost of PEP incurred per person was Rs . 1049 . 70 ( 16$ ) and average cost of transport Rs . 165 . 24 ( 3$ ) . However , assessment of costs is additional information gathered and was not looked as barrier for PEP . Table 1 describe the ratio of rabies exposure and PEP .
Contact tracing conducted immediately with little or no delay between the bite and the interview , will give detailed and accurate information . Contact tracing can also be carried out retrospectively to maximum of one year for more reliable data . Data collected beyond one year can result in recall error . [10] From contact tracing it was observed that , less than 10% of the exposures were by probable rabid animals and no brain sample was examined for confirmation of rabies . The interview of index case , family members and stake holders to trace bite victims and animal cases in the villages was similar to the studies done in Bali and Bohol . [12 , 13] Different studies on contact tracing among community members and healthcare workers , were either to trace bite events , or to look for clinical signs of rabies in the biting dogs , or to see if they could find out strategies for rabies control or to identify contacts for post-exposure prophylaxis to prevent the disease . [14 , 15 , 16] Majority of the bites were category-III exposures . Thirteen stray dogs were responsible for more than half of the bite victims in concordance to the observations from other studies . [17 , 18] Majority of bite victims had first visited a clinic/hospital in the town for treatment as the primary care provider in the village was not aware about rabies PEP . Two ( 40 . 0% ) probable rabies exposure victims did not seek treatment in the present study compared to contact tracing survey in Tanzania , which had showed that 15% and 24% of suspect rabies exposure did not seek medical attention . [6] In a hospital based study in Bhutan , 32% of the subjects mentioned that the biting dog looked normal and 9% mentioned that the biting dogs looked like suspect rabies contrary to the observation of the present study . [19] Studies have shown that hypersalivation and aggression are the common clinical signs observed for diagnosis of rabies in dogs similar to the finding in the present study . [12 , 14] The World Health Organization reported that , almost all rabies death victims had not sought rabies PEP and there were no facilities or health personnel available to provide PEP in many areas where the disease is prevalent and suggested strengthening availability of Rabies Immunobiologicals in these places . [20 , 7] It is concerning that 60% victims of probable rabid animal exposures were either not aware or , did not seek care or did not complete PEP . The study is too small to make any determinations about the rate or risk of human rabies in the study area . The standard surveillance practices applied to many human and animal diseases consist of case identification , contact tracing , epidemiologic investigation , and laboratory confirmation . [21] The outcome of animal ( alive , dead , sick , unknown ) was mainly based on information provided by the bite victim/people in the village . The surveillance system for dog rabies diagnosis was non existent in the villages surveyed . The methods of rabies surveillance practiced in many countries suffer from fundamental problems including a lack of trained professionals and lack of diagnostic laboratory capacity . This results in a lack of awareness of case burden , reduced funding for control , and poor community engagement around prevention . [8 , 11] Having a functioning surveillance system in villages will go a long way in achieving the WHO , OIE and FAO goal to educate , vaccinate and eliminate dog-mediated human rabies deaths in the world by 2030 . [22]Laboratory diagnosis is critical to confirm the status of a suspect case , in part , to justify prophylaxis in exposed persons or animals . [23] Human rabies is underreported and the disease is not a priority in endemic countries . [24] One Health emphasizes that the rabies control and elimination should be a joint effort of veterinary and medical field . [25] The Indian rabies survey had estimated that , for every 870 bites , there will be one rabies case . However , it was not possible to elicit , which bite was responsible for the rabies death . [4] The present study through contact tracing was able to identify bite victims who were exposed to probable rabid animal exposures . India is a vast country with limited resources . Data on availability , accessibility and affordability of PEP is needed to plan for better intervention strategies . Nearly 28% of the subjects did not receive any rabies prophylaxis in the present study . In such a situation , a rabies risk score card based on the knowledge of local rabies transmission , category of bite and dog rabid status , etc can be developed . The rabies score card/check list would be able to identify the bite victims who need rabies PEP . This can be made available to the primary care providers in the village along with campaigns for strengthening of rabies IEC in the community . Yes , there are chances of missing out genuine cases , however these can be overcome if the scale/check list has very high sensitivity , treating physicians are asked to administer PEP in case of doubt and individuals with high risk are targeted . Snowball methods are typically performed to find rare events , such as human rabies deaths . They are less accurate at describing rates of more common events , like bites or healthcare seeking behaviors . Some of the bite victims may have been missed because of the contact tracing methodology followed ( selection bias ) . Information on details of animal bites , PEP seeking behavior and follow up vaccination of the bite victims was based on the history revealed by the bite victim ( Information bias ) . Categorization of bites was according to bite victims observation ( observer bias ) , money spent and treatment taken are as revealed by bite victims ( recall bias ) . The classification of biting animal as non cases and probable rabid is based on facts given by the bite victims/people ( information bias ) . Clinical and laboratory confirmation of rabies in the biting animal was not possible . A study covering a wider geographical area was not possible due to feasibility issues . Contact tracing was successful in case detection of probable rabid animal exposures and suitable for a period of one year . In addition , contact tracing identified that only one tenth of the bite victims required PEP . Implementation of contact tracing technique for identification of rabid status of the biting animal . To conduct a prospective study in a larger geographic area . Development of a rabies risk score card for the health care provider . | In India , Rabies is a neglected zoonotic disease and the burden of rabies is usually not captured by the health system due to varied reasons . Hence , an exploratory study was attempted to find out the rabid status of the biting animal through contact tracing during the first quarter of 2017 in villages coming under three Public Health Centers . A total of 69 dog/cat bite cases were identified . Majority of bites were by stray dogs , Category-III , exposure to non cases and few received PEP . Percentage of bite victims who actually required PEP was calculated . These findings support the fact that contact tracing can be used as an additional tool by the health care provider for measurement of rabid animal bites , where resources are scarce , reporting systems are weak and priorities for disease vary . A rabies check list/score card to be made available at Public Health Centers , which shall ensure better utilization of limited but lifesaving Rabies Immunobiologicals . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"post-exposure",
"prophylaxis",
"tropical",
"diseases",
"microbiology",
"vertebrates",
"dogs",
"animals",
"mammals",
"health",
"care",
"viruses",
"preventive",
"medicine",
"rabies",
"health",
"care",
"providers",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"veterinary",
"science",
"rabies",
"virus",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"veterinary",
"diseases",
"zoonoses",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"lyssavirus",
"eukaryota",
"prophylaxis",
"viral",
"pathogens",
"disease",
"surveillance",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"amniotes",
"organisms"
] | 2018 | An exploratory study on rabies exposure through contact tracing in a rural area near Bengaluru, Karnataka, India |
A review of 400 clinical records of paracoccidioidomycosis ( PCM ) patients , 93 with the acute/subacute ( AF ) and 307 with the chronic form ( CF ) , attended from 1977 to 2011 , selected as to the schedule of release for study by the Office of Medical Records at the University Hospital of the Faculdade de Medicina de Botucatu – São Paulo State University – UNESP , was performed to detect cases in relapse . The control of cure was performed by clinical and serological evaluation using the double agar gel immunodiffusion test ( DID ) . In the diagnosis of relapse , DID , enzyme-linked immunosorbent assay ( ELISA ) and immunoblotting assay ( IBgp70 and IBgp43 ) were evaluated . Out of 400 patients , 21 ( 5 . 2% ) went through relapse , 18 of them were male and 3 were female , 6∶1 male/female ratio . Out of the 21 patients in relapse , 15 ( 4 . 8% ) showed the CF , and 6 ( 6 . 4% ) the AF ( p>0 . 05 ) . The sensitivity of DID and ELISA before treatment was the same ( 76 . 1% ) . DID presented higher sensitivity in pre-treatment ( 80% ) than at relapse ( 45%; p = 0 . 017 ) , while ELISA showed the same sensitivity ( 80% vs 65%; p = 0 . 125 ) . The serological methods for identifying PCM patients in relapse showed low rates of sensitivity , from 12 . 5% in IBgp70 to 65 . 0% in IBgp43 identification and 68 . 8% in ELISA . The sensitivity of ELISA in diagnosing PCM relapse showed a strong tendency to be higher than DID ( p = 0 . 06 ) and is equal to IBgp43 ( p = 0 . 11 ) . In sum , prevalence of relapse was not high in PCM patients whose treatment duration was based on immunological parameters . However , the used methods for serological diagnosis present low sensitivity . While more accurate serological methods are not available , we pay special attention to the mycological and histopathological diagnosis of PCM relapse . Hence , direct mycological , cytopathological , and histopathological examinations and isolation in culture for P . brasiliensis must be appropriately and routinely performed when the hypothesis of relapse is considered .
Paracoccidioidomycosis ( PCM ) is a systemic mycosis caused by thermo-dimorphic fungi from the Paracoccidioides brasiliensis complex and the Paracoccidioides lutzii complex [1] . Confined to Latin America , PCM is endemic to the area extending from Mexico to Argentina [2] . Although incomplete , the available data indicate a higher incidence of such mycosis in Brazil , where they are frequently diagnosed in the State of São Paulo [3] . PCM is known to be able to reactivate despite effective treatment because of quiescent fungi remain and disease relapse is possible . However , a few studies have investigated the relapse of paracoccidioidomycosis . A study that was conducted with 58 patients who were infected with paracoccidioidomycosis and treated with itraconazole indicated that there was a relapse in 8 ( 13 . 8% ) of the cases , where 50 . 0% of them occurred after 36 months of discontinued treatment [4] . The gold standard for diagnosing PCM is either the direct visualization of characteristic multiple-budding cells in biological fluids and tissue sections , or fungus isolation from clinical specimens [5] . Serological tests are useful for diagnosis , severity assessment and follow-up , especially the double agar gel immunodiffusion test ( DID ) [6] . The ELISA ( Enzyme-Linked Immunosorbent Assay ) test has been the subject of a number of publications in regard to the detection of circulating antibodies PCM patients [7] , although it has not been included in most clinical laboratories . This paper is aimed at evaluating the ELISA test and its ability to detect antibodies against Paracoccidioides brasiliensis in PCM patients in relapse .
This study was approved by the Research Ethics Committee of FMB-UNESP . Written informed consent was obtained from all participants . In this study IRB was signed by all the adult patients and by one of the parents of the children . We had no IRB signed by the closest relative or the legal representative . The criteria for the inclusion of the patients in the study were as follows: ( A ) Upon diagnosis: ( a ) PCM confirmed cases , characterized by the presence of a suggestive clinical condition and the identification of typical forms of P . brasiliensis yeast phase in one or more clinical materials; ( b ) PCM probable cases , characterized by the presence of suggestive clinical conditions and specific serum antibodies detected by the results of the DID test . ( B ) After treatment: patients showing PCM relapse , characterized by the recurrence of signs and symptoms indicative of PCM , with or without the identification of the typical forms of P . brasiliensis yeast phase in any clinical specimen , and/or the serology reaction to the DID test . Those in this group have also been given the appropriate therapy and have shown clinical cure , experienced a normalization of their erythrocyte sedimentation rate ( ESR ) and a serology regression to negative values , and have continued antifungal treatment for at least one year after serological cure . The exclusion criteria were as follows: presence of other systemic diseases of infectious , inflammatory or neoplastic source , such as co-morbidity; pregnancy; lactation; history of hypersensitivity or severe side effects in response to azole or cotrimoxazole; concomitant use of medicines that can interact with such antifungals or change their serum levels or concomitant use of other antifungals . Eight patients with mycologically confirmed and PCM who did not relapse constituted the control group; 122 serum samples from these patients were serologically evaluated by the double agar gel immunodiffusion test ( DID ) and the enzyme-linked immunosorbent assay ( ELISA ) . The categorization of patients and the assessment of disease severity in each patient were carried out according to Mendes [8] and the Paracoccidioidomycosis Surveillance and Control Guideline [6] by the infectious disease MD responsible for attending the patients . A review of 400 clinical records of PCM patients , 93 of whom with the acute/subacute ( AF ) and 307 with the chronic form ( CF ) , attended from 1977 to 2011 selected as to the order of release for study by the Office of Medical Records in the University Hospital of the Faculdade de Medicina de Botucatu – São Paulo State University – UNESP , was performed to detected cases patients in relapse . A standard form was filled in per patient containing their name , sex , age , enrollment code at the Hospital , date of admission to the service , presence of previous treatment , clinical form , treatment start date , and results of diagnostic and serological tests . Likewise , a review of the serological records of PCM patients was conducted by the Tropical Diseases & Mycology Research Laboratory – Department of Infectious and Parasitic Diseases , in Botucatu Medical School – UNESP . Treatment was deemed appropriate if the signs and symptoms of the disease were gone , there was a normalization of the erythrocyte sedimentation rate ( ESR ) and a negative reaction to DID could be observed during one year of antifungal medicine administration and for at least one year without such therapy . The categorization of cases of relapse was characterized by the reappearance of signs and symptoms that were compatible with PCM , with or without the identification of the typical P . brasiliensis yeast forms in any clinical specimen , or the reaction to DID after the appropriate treatment in patients with confirmed or probable PCM . Five types of relapse have been taken into account: ( a ) clinical , serological and mycological , characterized by the reappearance of signs and symptoms compatible with PCM , with or without identification of the typical P . brasiliensis yeast forms in any clinical specimen , and a positive DID serology test; ( b ) clinical and serological , characterized by the reappearance of signs and symptoms compatible with PCM and a positive DID serology test; ( c ) clinical and mycological , characterized by the reappearance of signs and symptoms compatible with PCM , with the identification of the typical P . brasiliensis yeast s in their common forms in any clinical specimen; ( d ) clinical , characterized by the reappearance of the clinical signs of PCM , responsiveness to sulfamethoxazole-trimethoprim combination ( TMP-SMX ) ; and ( e ) serological , characterized only by a positive DID response without clinical signs . The P . brasiliensis 113 ( Pb-113 ) yeast phase culture filtrate , prepared at the Clinical Mycology Laboratory of Araraquara Pharmaceutical Sciences College – UNESP , was used to run DID , ELISA and immunoblotting ( IB ) tests . The serum levels of anti-Pb antibodies were determined through a DID test which was run according to the specifications of Restrepo [9] at the Tropical Diseases Experimental Laboratory in Botucatu Medical School – UNESP . Non-diluted sera were tested first and then diluted by ½ , with serial dilutions ( ×2 ) . For each test , positive and negative control sera were included . The serum levels of anti-Pb antibodies were determined through an ELISA test [10]–[11] , which was run at the Tropical Diseases Experimental Laboratory in Botucatu Medical School – UNESP , according to the standard protocol set forth by the Mycosis Immunodiagnosis Laboratory of Adolfo Lutz Institute ( São Paulo , Brazil ) . The cutoff point was determined by drawing the ROC ( receiver operator characteristic ) curve for 200 PCM patients and 100 healthy subjects ( control subjects ) from Botucatu Blood Bank . The final end point was statistically defined according to the specifications made by Frey et al . [12] for a 95% confidence interval , and was equal to 0 . 710 optical density . Such serological techniques were conducted at the Mycosis Immunodiagnosis Laboratory of Adolfo Lutz Institute , in São Paulo , Brazil . Electrophoresis that involved the transfer of proteins contained in polyacrylamide gels to nitrocellulose membranes was carried out according to a methodology described by Towbin et al . [13] , as amended by Passos [14] . A 2×2 comparison of the sensitivity of the serological methods that were used was carried out through a binomial test , according to the specifications made by Siegel [15] . A comparison of the curves representing the time to achieve serological response in different groups of patients and serological techniques was conducted by using the ranks for the two methods , and regression was calculated for each of them , which were then compared to both regressions , through F-test [16] . A kappa coefficient was ascertained by means of SAS – statistical analysis system , Version 6 . 12 , SAS Institute Inc . , USA . The kappa coefficient was interpreted as follows: ( a ) mild , when lower than 0 . 4; ( b ) moderate , from 0 . 4 to 0 . 79; ( c ) strong , from 0 . 8 to 0 . 99; and ( d ) perfect , when equal to 1 . 0 . For each such statistic test , differences were set up at p≤0 . 05 .
Out of 400 patients , 21 ( 5 . 2% ) went through relapse , 3 of which ( 14 . 3% ) were female and 18 ( 85 . 7% ) were male , resulting in a male/female ratio of 6∶1 . The prevalence of male patients was 85 . 7% among the patients in relapse , 88 . 1% among the patients who did not relapse and 88 . 0% among the 400 studied patients , with no difference between patients in relapse or not ( p = 0 . 72 ) . Out of the 21 patients who experienced a relapse , 15 ( 71 . 4% ) showed a chronic form and 6 ( 28 . 6% ) showed an acute/subacute form . The incidence of relapse as to clinical form was the same: 15/307 ( 4 . 9% ) in the CF and 6/93 ( 6 . 5% ) in the AF ( p = 0 . 60 ) . Out of the 15 patients in relapse with chronic PCM , 5 of them showed the severe chronic form ( SCF ) , 9 of them showed the moderate chronic form ( MCF ) , and 1 of them showed the mild chronic form ( MiCF ) , while the 6 patients with acute/sub-acute PCM showed the severe acute/subacute form ( SAF ) . Fourteen out of 21 patients in relapse were treated with TMP-SMX and 7 with ITC . The duration of the treatment with TMP-SMX lasted 33 months , ranging from 28 to 91 , and 48 ( 21–135 ) with ITC . These 21 patients showed five diagnosis patterns of relapse: a ) positive clinical , serological and mycological – 3; b ) positive clinical and mycological – 7; c ) positive clinical and serological – 4; d ) only positive serological – 2; e ) only positive clinical – 5 . Regarding as to four of these patterns , in addition to clinical signs , a positive mycological and/or serological test was noted . The five patients who presented only clinical signs were diagnosed as relapse patients based on the exclusion of the other possible etiologies and progress to cure after resumption the treatment with TMP-SMX . In contrast , 2 patients only had a serological relapse , which progressed to a negative response after resuming of the antifungal treatment . The case concerning 9 ( 47 . 4% ) patients with clinical and/or mycological relapse should be highlighted because the DID tested negative . Relapses occurred from 46 to 296 months ( 96 on the average ) after treatment introduction and from 4 to 267 months ( 60 on the average ) after treatment discontinuation . The time of relapse did not show any differences compared to the clinical forms: ( a ) due to treatment introduction: AF = 77 . 5 months ( 38–296 ) ; CF = 112 . 0 ( 46–234 ) ; p>0 . 05; ( b ) due to treatment discontinuation: AF = 38 . 5 months ( 4–267 ) ; CF = 75 . 0 ( 14–168 ) ; p>0 . 05 .
Thus , relapse is defined in this paper as the reappearance of symptomatology suggesting PCM ( the etiology of which has been confirmed by mycological and/or serological tests by DID ) , specific antibodies serum level regression to negative ( non-reactive ) , and persistence of such conditions for at least one year after antifungal therapy discontinuation in patients with prior etiological diagnoses of and successful treatment for PCM , which included clinical cure and the normalization of erythrocyte sedimentation rate . This approach is common in clinical practice , and this paper is aimed at addressing the problem using other methods of serological diagnosis , rather than trying to identifying the source of new infectious processes; otherwise , the research design would have to have been completely different . The prevalence of relapse out of the 400 patients was equal to 5 . 2% , i . e . , lower than the 13 . 8% that was observed by Marques [4] , who followed up with 58 patients . The prevalence of relapse did not vary according to clinical form , a finding that confirms those of Marques [4] . The diagnosis of relapse was late , i . e . , within an average of 60 months after antifungal therapy discontinuation . Marques [4] findings revealed that 50 . 0% of the reactivations could be observed 36 months after antifungal therapy discontinuation , while our findings presented that 66 . 6% of the reactivations could be observed after the same period . Few authors address the subject of PCM relapse , an event that is not rare which can bring serious repercussions to patients , since preserved organs may be impaired and the injured can worsen . In addition , a low prevalence of serological relapse can lead to complications in future diagnoses because it can suggest another manifesting disease , which may lead the MD to opt for unjustified treatments , thus delaying the diagnosis and treatment of the PCM relapse . This study appraised different serological methods for PCM relapse diagnoses , from the DID reaction , which has been the method of choice due to its specificity , positive predictive value , repeatability and simplicity [6] , [17] , to immunoblotting with gp43 identification , which to the best of our knowledge and belief , is routinely available only at the Adolfo Lutz Institute ( São Paulo ) . A comparison of both moments , the initial pretreatment and relapse , disclosed that the ELISA test had a higher sensitivity than the DID reaction , which provided several patients with a diagnosis that was not discovered by the gel precipitation test . These findings evince that the ELISA test should be the method of choice for diagnosing PCM relapse . The sensitivity of EIA is still only 65% so a negative result does not rule out relapse infection . Several mechanisms might explain negative serological results in some relapse patient . One is that the fungus causing the relapse may not be P . brasiliensis , a second species could induce the production of DID-undetectable antibodies when using the same antigen as that in the initial reaction . Another explanation would be the modification of the antigenic composition of quiescent fungi , which would initiate the production of modified antibodies that are undetectable by methods designed on the basis of the typical antigens of P . brasiliensis . Two other findings suggest a third hypothesis to explain the absence of antibody detection patients in relapse . The first is the polymorphism of the gene encoding the P . brasiliensis gp43 immunodominant antigen [18] . The second is the finding of two different isolates of P . brasiliensis in different organs of the same armadillo [19] , a finding also observed in a patient with two genetically distinct isolates of P . brasiliensis . Clinical isolates were obtained from injuries on different anatomic sites and characterized by means of the RAPD technique . The genotypic evaluation showed more than 28% variability between these fungal isolates , therefore suggesting that different genotypes of P . brasiliensis may infect the same patient and induce active disease [20] . Thus , it is possible that some patients have been infected by more than one of P . brasiliensis isolate , with a positive serology to one of them; the other isolate , responsible for the relapse , could not be serologically recognized [21] . Other studies confirm these hypotheses . The cryptic species of Paracoccidioides brasiliensis , S1 , PS2 , PS3 and Paracoccidioides lutzii , recently identified [22]–[24] , have implications on PCM immunodiagnosis [25]–[26] , and presented a regional distribution . Such results suggest that there are differences in the fungus antigenic composition of P . brasiliensis . Arantes et al . [27] collected aerosol samples for the environmental detection of Paracoccidioides ssp . , by placing a cyclonic air sampler at the entrance of the armadillo burrow in Botucatu ( SP ) . Most ITS sequences showed a high similarity with homologous sequences of P . lutzii in the GenBank database , suggesting that this species – Paracoccidioides lutzii – may not be exclusive in Midwest Brazil . Studies involving other microorganisms also confirm such hypotheses . Ghannoum et al . [28] analyzed clinical isolates of Cryptococcus neoformans obtained from five patients with recurrent cryptococcal meningitis and demonstrated the different composition of sterols between relapse and pre-treatment isolates , which indicates that the sterols had been modified by therapy or that patients were infected with new isolates with different sterol compositions . Soll et al . [29] monitored Candida albicans strains isolated from different body sites of a single patient in three events of recurrent vulvovaginal candidiasis . The strains were evaluated by using Southern blot hybridization . They observed three different strains of C . albicans colonizing five sites , at the time of the first infection . Although the same strain of C . albicans had been responsible for the three vaginal infections , different colony phenotypes manifested with every new infection . Such findings suggest that the action of antifungal and/or immune response of patients could lead to changes in the antigenic composition of fungi that are responsible for relapse , which would induce the production of antibodies that do not recognize the antigens used in routine serological tests . Hence , diagnostic methods based on proteomics could contribute to the solution of this problem . Evaluations in that direction are already in the course of pilot study in our Service . Another possible method of diagnosis of PCM relapse would be the analysis of circulating [30] and/or urinary [31] antigens , the methods of which have been evaluated successfully in the initial diagnosis of this mycosis and in follow-up with these patients . Nevertheless , such methods are not yet routinely available in clinical laboratories , especially those that are public . Polymerase Chain Reaction ( PCR ) with primers specific for P . brasiliensis detection in clinical materials and other molecular method could also be excellent options for diagnosing PCM relapse , however , these techniques have not been incorporated yet into the routine of clinical laboratories [32] . In addition , PCR sensitivity is low in serum samples [33] , the aim of our study , although higher in other clinical materials [32]–[35] . As a final point , while more sensitive and specific serological methods are assayed , we should continue to give special attention to the mycological diagnosis of PCM relapse . Such methods include the gold standard of diagnosing PCM . Hence , direct mycological [36]–[37] , cytopathological [36]–[38] , histopathological [36] and cultivation tests for P . brasiliensis [36]–[39] must be appropriately and routinely conducted when following up with such patients . | Paracoccidioidomycosis ( PCM ) is a systemic mycosis caused by fungi of the Paracoccidioides brasiliensis and Paracoccidioides lutzii complexes , which live in the soil and affect mainly rural workers in the most productive period of their lives . PCM can relapse after effective treatment because quiescent fungi can reactivate . However , the physician has to differentiate a relapse from another disease . In the diagnosis of relapse , the classical double agar gel immunodiffusion test ( DID ) and the modern enzyme-linked immunosorbent assay ( ELISA ) and immunoblotting ( IB ) were evaluated . The frequency of PCM relapse and its diagnosis were the objective of this study . The prevalence of relapse was low and the new serological tests presented a little higher sensitivity than DID . Since the serological tests presented only a moderate sensitivity , direct mycological , cytopathological , and histopathological examinations and isolation in culture for the fungi must be appropriately and routinely performed when the hypothesis of relapse is considered . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"medicine",
"clinical",
"immunology",
"biology",
"and",
"life",
"sciences",
"immunology"
] | 2014 | Prevalence and Serological Diagnosis of Relapse in Paracoccidioidomycosis Patients |
Candida is the most common human fungal pathogen and causes systemic infections that require neutrophils for effective host defense . Humans deficient in the C-type lectin pathway adaptor protein CARD9 develop spontaneous fungal disease that targets the central nervous system ( CNS ) . However , how CARD9 promotes protective antifungal immunity in the CNS remains unclear . Here , we show that a patient with CARD9 deficiency had impaired neutrophil accumulation and induction of neutrophil-recruiting CXC chemokines in the cerebrospinal fluid despite uncontrolled CNS Candida infection . We phenocopied the human susceptibility in Card9-/- mice , which develop uncontrolled brain candidiasis with diminished neutrophil accumulation . The induction of neutrophil-recruiting CXC chemokines is significantly impaired in infected Card9-/- brains , from both myeloid and resident glial cellular sources , whereas cell-intrinsic neutrophil chemotaxis is Card9-independent . Taken together , our data highlight the critical role of CARD9-dependent neutrophil trafficking into the CNS and provide novel insight into the CNS fungal susceptibility of CARD9-deficient humans .
Human systemic fungal infections are typically opportunistic and primarily affect patients with acquired immunodeficiency ( i . e . commonly due to HIV infection or modern medical interventions such as cancer chemotherapy and transplantation ) or inborn errors of immunity [1 , 2] . Systemic candidiasis , in particular , is the most common deep-seated fungal infection in the developed world and is now the most common cause of nosocomial bloodstream infection in the US [3] . Despite the availability of potent antifungal drugs , clinical outcomes of infected patients are often poor leading to significant morbidity and unacceptably high mortality rates [1] . Adjunctive immune-based therapies are therefore becoming increasingly desirable , yet their development and success depends on improving our understanding of the cellular and molecular basis of human antifungal immunity . Fungi are recognized by the host immune system via innate pattern recognition receptors ( PRRs ) that are predominantly expressed by myeloid cells , including neutrophils , monocytes/macrophages and dendritic cells . C-type lectin receptors ( CLRs ) are a large class of PRRs of which several members including Dectin-1 , Dectin-2 and Mincle have been shown to play crucial roles in antifungal immunity [4] . These receptors bind components of the fungal cell wall and this innate recognition event leads to the initiation of intracellular signaling cascades that stimulate cellular responses such as phagocytosis , induction of the respiratory burst , and production of pro-inflammatory chemokines/cytokines [4] . Although different CLRs exhibit independent functions , they utilize a common signaling pathway involving the kinase Syk and the signaling adaptor , CARD9 [5] . CARD9 is one of the most crucial mammalian antifungal immune molecules identified to date . Genetic deletion of Card9 in mice results in defective pro-inflammatory cytokine production by myeloid cells and leads to a substantial increase in mortality following C . albicans challenge compared to WT mice [6] . CARD9 is also indispensable for antifungal defense in humans . Autosomal recessive CARD9 deficiency due to biallelic missense or nonsense CARD9 mutations leads to profound defects in production of pro-inflammatory cytokines in response to fungal-specific stimuli , including GM-CSF , IL-1β , IL-6 and TNFα [7–14] . Other known CARD9-dependent functions in humans include CR3-dependent killing of unopsonized yeast by neutrophils [7 , 14 , 15] , generation of neutrophilic myeloid-derived suppressor cells [16] , and , in some patients , generation of Th17 cells [8 , 11] . Along with loss of these CARD9-dependent functions , patients deficient in CARD9 are highly susceptible to spontaneous development of systemic fungal infections , predominantly caused by Candida species [7 , 8] , as well as severe cutaneous and subcutaneous infections caused by dermatophytes collectively known as deep dermatophytosis [13 , 17] . Interestingly , a large proportion of CARD9-deficient patients develop systemic candidiasis that specifically targets the central nervous system ( CNS ) without persistent fungemia [7–9 , 12 , 18] . In patients with intact CARD9 signaling , systemic candidiasis is most often associated with the kidney , liver and spleen [19] . Therefore , this striking CNS-targeted manifestation of systemic candidiasis in CARD9-deficient patients suggests that CARD9 is uniquely required for protecting the CNS from fungal invasion . However , the CARD9-dependent immune effector functions operating in the CNS remain undefined . Here , we describe a novel CARD9 missense mutation ( c . 170G>A; p . R57H ) identified in an 11 year-old girl with Candida meningoencephalitis . We extensively characterized the impact of this mutation on the antifungal host response in human myeloid cells , focusing on neutrophils , which are the primary immune effector cells against systemic C . albicans infections . Furthermore , we phenocopied the human CNS susceptibility to C . albicans infection in Card9 knockout ( KO ) mice , and characterized the Card9-dependent immune protective functions in this tissue in vivo . We have found that CARD9 is critical for control of fungal invasion in the CNS , acting to promote fungal-specific and tissue-specific neutrophil trafficking from the blood to the target organ via neutrophil-targeted chemokine production in both mice and humans .
An 11-year-old girl living in North Carolina in the US and born to consanguineous El Salvadorian parents presented at 8 years of age with fever , back pain , headache and vomiting and was diagnosed with T12-L1 spine osteomyelitis with associated diskitis and paraspinal abscess . She had recurrent oral thrush since birth but no other infections or medical conditions . Vertebral body bone biopsy showed granulomatous inflammation and necrosis with abundant Candida pseudohyphae ( Fig 1A ) and culture of the biopsy sample grew pan-sensitive Candida albicans . The patient received a 6-month course of fluconazole with clinical improvement . Approximately 15 months after discontinuation of fluconazole treatment , at the age of 9 , the patient’s infection recurred and she presented with fever , neck and back pain , headache and vomiting . Cerebrospinal fluid ( CSF ) analysis , which was performed this time , revealed pleocytosis ( 213 nucleated cells/μl; 51% lymphocytes , 27% monocytes , 22% eosinophils ) , elevated protein ( 168 mg/dL ) and decreased glucose ( <20 mg/dL ) ; CSF culture grew pan-sensitive C . albicans and beta-D-glucan ( BDG ) in the serum and CSF was markedly elevated at >500 . Magnetic resonance imaging revealed cervical spine osteomyelitis , leptomeningeal enhancement ( Fig 1B , black arrows ) complicated by a large syrinx ( Fig 1B , white arrow ) and obstructive hydrocephalus that necessitated placement of a ventriculoperitoneal shunt , and brain abscesses ( Fig 1C , arrow ) . No liver , splenic , intestinal or renal lesions were noted on abdominal imaging and no heart valve vegetation was seen by echocardiogram . Immunologic analyses of peripheral blood revealed normal monocyte , neutrophil , eosinophil , CD4+ T , CD8+ T , CD19+ B , and natural killer lymphocyte numbers . Phagocyte oxidative burst assessed by the dihydrorhodamine ( DHR ) assay was normal . T-lymphocyte proliferations were normal in response to PHA and PWM but decreased or absent in response to Candida and tetanus antigens . The levels of the patient’s IgA , IgE , IgG and IgM were normal . Neither her parents nor her siblings had a history of infections . Since autosomal recessive CARD9 deficiency has been shown to predispose to Candida meningoencephalitis [7–9 , 12] , we sequenced CARD9 exons ( S1 Table ) and found a homozygous CARD9 missense mutation , c . 170G>A in exon 3 causing a substitution of histidine for arginine at position 57 ( R57H ) , a highly conserved residue within the CARD domain of the CARD9 protein ( Fig 1D and S1 Fig ) . The patient’s parents and brother are healthy and heterozygous for the mutation ( WT/c . 170G>A ) whereas the patient’s healthy sister has WT CARD9 alleles ( Fig 1D and 1E ) , consistent with autosomal recessive CARD9 deficiency with complete clinical penetrance . The c . 170G>A mutation was not found in the 1000 Genomes or ExAC databases covering 117 , 000 chromosomes in the specific CARD9 exon , thus this mutation is unlikely to be an irrelevant polymorphism . In addition , it was predicted to be deleterious by SIFT ( deleterious ) , CADD [PHRED score 17 . 7] [20] and PolyPhen 2 [highest possible score of 1] [21] . Collectively , these data strongly suggest that the patient is homozygous for a rare and deleterious mutant CARD9 allele . Various CARD9 mutations that lead to production of full-length , truncated or absent CARD9 proteins have been reported to date [7 , 8 , 12 , 13] . Hence , we next investigated the impact of the c . 170G>A mutation on CARD9 protein levels by FACS and Western blot analyses on the patient neutrophils and CD14+ monocytes and compared them to that of healthy donor myeloid cells . FACS revealed similar expression of CARD9 in WT and patient neutrophils and CD14+ monocytes ( Fig 1F ) , whereas , as expected , no significant CARD9 expression was observed on T lymphocytes ( S2 Fig ) . Similarly , immunoblot analyses for CARD9 on whole-cell extracts of the patient neutrophils and CD14+ monocytes demonstrated that CARD9 protein expression levels and molecular weight were similar with that of WT cells ( Fig 1G ) . These data are in agreement with previous reports in which nonsense mutations prevented the generation of full-length CARD9 protein , whereas missense mutations did not predictably do so [7 , 8 , 12 , 13] . We next sought to examine the functional consequences of the c . 170G>A mutation by measuring the production of pro-inflammatory cytokines and chemokines from peripheral blood mononuclear cells ( PBMCs ) stimulated with either heat-killed C . albicans or LPS . Consistent with previously reported CARD9 mutations [8 , 12 , 13] , the c . 170G>A mutation resulted in dramatically decreased IL-1β , IL-6 and TNF-α production after 48 hours of stimulation with C . albicans compared with 10 healthy donors , while cytokine production following LPS stimulation was comparable in the patient and healthy donor PBMCs ( Fig 1H–1J ) . Furthermore , the production of GM-CSF , IFN-γ , IL-1α , IL-2 , IL-10 , CCL3 , CCL4 , CCL5 , CCL7 and CXCL8 was also found to be CARD9-dependent in human PBMCs following fungal stimulation , whereas production of IL-4 , IL-7 , IL-15 , IL-16 , CCL2 , CXCL1 , CXCL2 and CXCL5 was CARD9-independent ( S3 Fig ) . Since decreased proportions of IL-17+ CD4+ T cells have been reported in some , but not all , CARD9-deficient patients [7 , 8 , 10 , 12 , 13] , we evaluated the production of IL-17A by the patient’s CD4+ T cells following PMA/ionomycin stimulation ex vivo using FACS , and found no impairment relative to 9 healthy donors ( Fig 1K ) . Taken together , these data show that the c . 170G>A mutation results in the generation of full-length CARD9 protein which exhibits impaired capacity for production of pro-inflammatory mediators upon fungal stimulation . Neutrophils and monocytes/macrophages are the critical cellular mediators of host defense against systemic C . albicans infection [19 , 22 , 23] . Therefore , we next investigated the effects of the c . 170G>A mutation on the ability of neutrophils and monocytes to kill the two major morphological forms of C . albicans , yeasts and hyphae . Hyphae are the invasive filamentous fungal forms present in human infected CNS tissue ( Fig 1A ) . It has been previously shown that CARD9 deficiency impairs the killing capacity of neutrophils against unopsonized , but not opsonized , C . albicans yeasts [7 , 15] . Indeed , we found decreased killing by the patient’s neutrophils of unopsonized C . albicans yeasts whereas neutrophil killing was intact against opsonized yeasts ( Fig 2A ) . However , CARD9-deficient neutrophils did not exhibit impaired killing capacity against opsonized or unopsonized C . albicans hyphae ( Fig 2B ) , which are the predominant form found in infected CNS tissue of mice and humans ( Fig 1A ) [24 , 25] . We then examined the impact of CARD9 deficiency on the ability of CD14+ monocytes to kill C . albicans yeasts and hyphae . CD14+ monocytes demonstrated inferior killing capacity of C . albicans relative to neutrophils ( Fig 2A–2D ) . As shown in Fig 2C and 2D , CD14+ monocytes from the patient exhibited a modest decrease in killing of C . albicans yeasts , but hyphal killing was intact . The decreases observed in killing of yeasts by the patient’s neutrophils and CD14+ monocytes were not due to impaired fungal internalization ( Fig 2E ) . Collectively , these data show that although CARD9-deficient neutrophils have a decreased ability to kill unopsonized C . albicans yeasts , CARD9 is dispensable for killing of the invasive hyphal forms found in infected CNS tissue . Therefore , other factors in addition to CARD9-mediated C . albicans yeast killing by phagocytes are likely to also contribute to the susceptibility of CARD9-deficient patients to Candida infection of the CNS . Notably , we observed that the patient had a striking absence of neutrophils in the infected CSF , despite uncontrolled fungal disease . Fig 3A shows a representative cytopathology image of the infected CSF , in which lymphocytes , eosinophils and mononuclear phagocytes could be found , but neutrophils were absent . Indeed , FACS analysis of the infected CSF revealed that the predominant accumulating leukocytes were T-lymphocytes and eosinophils ( Fig 3B ) , which are not critical mediators of host defense against systemic C . albicans infection [19] . This finding is in agreement with other reports of high numbers and predominance of lymphocytes and eosinophils in the CSF of Candida-infected CARD9-deficient patients [7 , 9] . Monocytes , dendritic cells and B-cells were also present ( 5–10% each ) , whereas neutrophils comprised <1% of total leukocytes ( Fig 3B ) . Over time , the lack of neutrophils was consistent with these cells never exceeding 7% of total leukocytes in the infected CSF ( range: 0–7% ) despite continued inflammation ( range of nucleated cells in CSF: 108-422/μl; range of % of eosinophils within leukocytes in CSF: 3–40%; range of lymphocytes within leukocytes in CSF: 25–72%; range of protein levels in CSF: 161–845 mg/dL ) and extremely high fungal load indicated by persistently elevated CSF BDG ( Fig 3C ) . Persistent Candida tissue hyphal invasion and absence of neutrophil infiltration were confirmed in the patient’s CNS tissue by brain biopsy ( S4 Fig ) . The decreased neutrophil accumulation noted in our patient is grossly suboptimal compared to the robust neutrophil accumulation seen in patients without CARD9 deficiency when they develop C . albicans meningitis , typically post-neurosurgical procedures ( Fig 3C; red bar ) [26–29] . Consistent with that , a patient at NIH who developed C . albicans meningitis post-Ommaya reservoir placement exhibited pleiocytosis ( 133–1091 nucleated cells/μl ) and significant neutrophil mobilization in the infected CSF ( 56–73% of total leukocytes ) ( Fig 3C; grey bar ) . We next sought to investigate the etiology of suboptimal neutrophil accumulation in the infected CSF in the setting of CARD9 deficiency . The patient did not manifest peripheral neutropenia ( range of absolute neutrophil count in the blood: 1 , 960–5 , 340/mm3 ) , ruling out this possibility ( Fig 3D ) . We next examined whether CARD9 is a survival factor for human neutrophils . For that , we harvested neutrophils from the peripheral blood of the patient and healthy donors and determined differential cell apoptosis and death at 3 and 6 hours after serum starvation , TNF-α stimulation or live C . albicans challenge ex vivo . Neutrophils were also directly exposed to the patient’s infected CSF to evaluate for the presence of potential soluble pro-apoptotic factors within the infected tissue . We found that CARD9-deficient neutrophils did not exhibit enhanced apoptosis or death under any of the tested conditions relative to WT neutrophils ( Fig 3E ) . Therefore , these data demonstrate a striking absence of neutrophil accumulation in the Candida-infected CSF in human CARD9 deficiency , which is not caused by peripheral neutropenia or defective neutrophil survival . We reasoned that the decreased neutrophil recruitment in the CNS of the patient could be explained by the net effect of ( 1 ) impaired intrinsic chemotactic ability of CARD9-deficient human neutrophils and/or ( 2 ) defective production of neutrophil-targeted chemoattractants in the infected tissue . Thus , we tested whether CARD9-deficient neutrophils have cell-intrinsic chemotactic defects and/or whether the neutrophil-targeted chemoattractants are not produced in order to generate the chemotactic gradient that is necessary for cell trafficking into the infected CNS . We first assessed neutrophil chemotaxis using both the EZ-TAXIScan system and a modified Boyden chamber , the 96-well Neuroprobe chamber . The EZ-TAXIScan instrument allows for simultaneous measurement of cell velocity and directionality by observing the spatial and temporal movement of individual cells toward a chemoattractant . Using this method , we found no defect in chemotaxis of CARD9-deficient neutrophils toward IL-8 or formyl-methionyl-leucyl-phenylalanine ( fMLP ) ( Fig 4A and 4B ) , two prototypic chemotactic factors for human neutrophils . We then used the 96-well Neuroprobe chamber and examined neutrophil chemotaxis toward IL-8 , fMLP , the leukotriene B4 ( LTB4 ) , and the complement factor C5a . Similar to our EZ-TAXIScan data , no impairment was found in the chemotaxis of CARD9-deficient neutrophils relative to WT neutrophils using the 96-well Neuroprobe chamber when correcting for multiple comparisons ( Fig 4C ) . Since the patient’s neutrophils did not exhibit a primary chemotaxis defect , we next measured levels of the neutrophil-targeted chemokines CXCL1 , CXCL2 , CXCL5 and IL-8 in the infected CSF by Luminex array at different prospective time-points during uncontrolled fungal infection . Strikingly , we found absent to very low levels of these chemokines ( Fig 4D ) . IL-8 was detected at higher levels ( mean: 267 . 6 pg/mL ) than all other chemokines ( mean: 13 . 3 pg/mL for CXCL1 , 10 . 6 pg/mL for CXCL2 , 16 . 3 pg/mL for CXCL5 ) , but these IL-8 levels were comparable to previously reported IL-8 levels in uninfected CSF from healthy donors [30] . Interestingly , the measured levels of IL-8 are significantly lower than those detected in patients with neutrophilic meningitis in the absence of CARD9 deficiency ( Fig 4D; red bar ) [30] . Consistent with the absence of induction of neutrophil-targeted chemokines , CSF from the infected patient did not exhibit any greater chemotactic activity toward WT and patient neutrophils than that observed in buffer , CSF from a healthy subject or CSF from two patients with lymphocytic ( non-neutrophilic ) meningitis ( Fig 4E ) . Instead , our CARD9+/+ patient who developed C . albicans meningitis post-Ommaya reservoir placement associated with neutrophil mobilization in the CSF exhibited significant induction of CXC neutrophil-recruiting chemokines in the infected CSF ( Fig 4D; grey bar ) . Specifically , the CSF levels of IL-8 , CXCL1 , CXCL2 and CXCL5 were 17 . 1–fold ( 4576 . 8 pg/mL ) , 294 . 4–fold ( 3120 . 6 pg/mL ) , 16–fold ( 261 . 2 pg/mL ) and 93 . 4–fold ( 1523 . 1 pg/mL ) greater , respectively , relative to those detected in the CSF of our CARD9-deficient patient . In line with the induction of neutrophil-targeted CXC chemokines , Candida-infected CSF from the CARD9+/+ patient exhibited chemotactic activity toward WT neutrophils ex vivo ( Fig 4F ) . Altogether , these data suggest that the absence of neutrophil accumulation in the infected CNS despite uncontrolled fungal disease correlates with a defect at the level of chemoattractant production in the infected tissue , whereas CARD9 is dispensable for cell-intrinsic neutrophil chemotaxis . To further examine the mechanistic role of CARD9 in antifungal defense of the CNS , we utilized a well-established mouse model of systemic C . albicans infection . This model results in fungal dissemination to the brain associated with an early influx of neutrophils , whose accumulation decreases later on as the infection is controlled [24] . To directly examine whether the recruitment of neutrophils to the CNS is critical for control of fungal proliferation in this tissue , we depleted neutrophils in WT mice using the 1A8 clone of the anti-Ly6G antibody [31] . We found that administration of 1A8 successfully depleted >90% of neutrophils in the brain of infected mice compared to infected mice treated with an isotype control antibody , and had no effect on the recruitment of other myeloid cells such as Ly6Chi monocytes ( Fig 5A ) . Analysis of brain fungal burdens in these animals revealed that neutrophil depletion by 1A8 led to a significant increase in brain fungal load compared to control animals ( Fig 5B ) . Therefore , these data demonstrate that neutrophil recruitment to the CNS post-infection is critical for control of fungal brain infection . We further analyzed the fungal-load dependence of the neutrophil response in the infected WT mouse CNS by infecting WT mice with either a low ( 7x104 ) or a high ( 1x106 ) inoculum of C . albicans . WT mice infected with the high inoculum developed significantly greater brain fungal burden compared to those infected with the low inoculum ( Fig 5C ) . Using FACS , we assessed the number of recruited neutrophils in the brains of animals with high or low tissue burden . We found that animals infected with the high inoculum , which developed high brain fungal proliferation , had nearly 10-fold greater neutrophil numbers in the brain than observed in the mice which received the low inoculum and had lower tissue fungal burden ( Fig 5D ) . Accordingly , we found that the levels of the major mouse neutrophil-targeted chemokines Cxcl1 ( KC ) and Cxcl2 ( MIP-2 ) were significantly increased in the brains of mice with the higher tissue fungal burden ( Fig 5E ) . In contrast , LPS-induced chemokine ( LIX ) , the mouse homologue of CXCL5 which can also recruit neutrophils , was similar in all brains analyzed and had no correlation to the extent of brain fungal burden ( Fig 5E ) , suggesting that Cxcl5 induction is not fungal-load dependent in the infected mouse brain . Taken together , these data show that neutrophil-targeted chemokine production and neutrophil mobilization to the infected CNS directly correlates with the extent of tissue fungal burden . We next sought to understand the functional role of Card9 in the mouse brain during C . albicans infection . For this , we used Card9-/- mice that are significantly more susceptible to systemic candidiasis ( Fig 6A ) , as has been previously shown [6] . Importantly , we phenocopied for the first time the human CNS susceptibility to Candida infection of CARD9-deficient patients in Card9-/- mice which developed uncontrolled brain infection , evidenced by significantly higher brain fungal burdens as early as 24 hours post-infection ( Fig 6B ) . At 72 hours post-infection , Card9-/- mice exhibited dramatically greater brain fungal burdens ( over 1000-fold ) compared to WT animals . In agreement , histopathological analysis of infected brains revealed extensive tissue invasion in the Card9-/- brain by hyphae , which were the predominant fungal morphology observed in the infected brain ( Fig 6C ) . Since our CARD9-deficient patient had limited numbers of neutrophils in the infected CSF , we next investigated whether the same defect was present in the infected Card9-/- mouse brain . Indeed , we found a profound decrease in neutrophil accumulation in the Card9-/- brain at 24 hours post-infection compared with WT animals ( Fig 6D and S5A Fig ) . Notably , while WT mice recruited high numbers of neutrophils to the brain early after infection , there was no significant increase in neutrophil numbers or frequency between uninfected and infected Card9-/- brains at any time point after infection , despite dramatic tissue fungal proliferation ( Fig 6D and S5A Fig ) . Indeed , histopathology revealed that neutrophils could be seen to crowd around hyphae in WT brains at 24 hours post-infection , whereas very few of these cells were observed in Card9-/- brains ( Fig 6E ) . Card9-/- brains had similar neutrophil numbers with WT brains at 72 hours post-infection; strikingly , these numbers were similar to those observed in the uninfected brain ( Fig 6D ) . This reflects a markedly defective response in Card9-/- mice since these animals have over 1000-fold increase in brain fungal burden compared to WT ( Fig 6B ) ; hence , based on the inoculum-dependence of neutrophil accumulation in the mouse brain ( Fig 5 ) , these animals should have mobilized significantly more neutrophils into this tissue . In our patient and in other reported patients’ CSF [7 , 9] , in addition to a lack of neutrophils , we and others also observed high numbers of lymphocytes and eosinophils . To investigate if this was also true in Card9-/- mice , we calculated the relative frequency of multiple lymphoid and myeloid cell populations in the brain following infection by FACS . Besides a significant reduction in neutrophils and a corresponding increase in microglia percentages in Card9-/- brains , we did not find any other significant differences in the frequency of other leukocyte populations between WT and Card9-/- brains at the tested time points ( S5B Fig ) . Collectively , these data show that , consonant to human findings , Card9 is required for the control of C . albicans infection of the CNS and for promoting neutrophil accumulation in the infected brain . We next investigated whether neutrophil accumulation into the C . albicans-infected kidney is Card9-dependent . We found that , in line with previously reported findings [6] , Card9-/- mice had significantly increased kidney fungal burdens at both 24 and 72 hours post-infection ( S6A Fig ) . However , unlike the brain , Card9-/- mice were able to mobilize significant numbers of neutrophils into the kidney at 72 hours post-infection , approximately 3-fold greater than in WT infected kidneys at the same time-point post-infection , and nearly 100-fold greater than in uninfected control kidneys ( S6B Fig ) . Although the extent of neutrophil accumulation may still be to a degree suboptimal relative to the higher kidney fungal load seen in Card9-/- kidneys [22 , 24 , 32] , the data collectively indicate that Card9 plays organ-specific roles in neutrophil accumulation after systemic candidiasis , in agreement with a recent report showing neutrophil mobilization into the C . tropicalis infected Card9-/- kidney [33] . We then assessed whether the defect in neutrophil accumulation into the infected brain was specific to fungi or could be recapitulated with non-fungal pathogens . For this , we infected WT and Card9-/- animals intravenously with Staphylococcus aureus , an important human bacterial pathogen . We found that Card9-/- mice had similar bacterial burdens and accumulation of neutrophils in the brain compared to WT mice at 48 hours post-infection ( S7 Fig ) . These data are in agreement with the lack of development of bacterial CNS infections in patients with CARD9 deficiency and show that the neutrophil accumulation defect we observed during C . albicans infection is specific for fungal pathogens . We next began to investigate possible mechanisms of the Card9-dependent accumulation of neutrophils in the brains of C . albicans infected mice . We first assessed whether Card9 is critical for the production of neutrophil progenitors in the bone marrow and egress of these cells into the peripheral blood after infection . We found that C . albicans infection led to a significant increase in neutrophil progenitors in the bone marrow ( S8 Fig ) and a significant induction of neutrophilia in the blood of WT mice ( Fig 6F ) . No differences were observed in neutrophil numbers in the bone marrow ( S8 Fig ) and peripheral blood ( Fig 6F ) between WT and Card9-/- animals , either at steady state or after infection . Since neutrophil production and mobilization of large numbers of neutrophils from the bone marrow into the blood of Card9-/- mice following C . albicans infection did not translate into accumulation of these cells into the infected CNS ( Fig 6D–6F ) , these data collectively suggest that Card9 is critical for the trafficking of neutrophils from the blood into the infected CNS . We showed earlier that CARD9-deficient human neutrophils had normal chemotaxis toward four major neutrophil-targeted chemoattractants ex vivo ( Fig 4A–4C ) . To analyze the recruitment of WT and Card9-/- neutrophils into the fungal-infected CNS in vivo , we generated mixed bone marrow chimeras in order to directly assess whether Card9-/- neutrophils have a cell-intrinsic chemotactic defect . WT animals were lethally irradiated and subsequently injected with a 1:1 ratio of CD45 . 1+ WT and CD45 . 2+ Card9-/- bone marrow cells . Following the adoptive transfer , animals were left to reconstitute for 8 weeks prior to C . albicans infection ( S9 Fig ) . The relative recruitment of WT and Card9-/- neutrophils to the brain of infected chimeras was then assessed by FACS , with congenic CD45 isoform expression used to distinguish between WT and Card9-/- neutrophils . Using this approach , we found no defect in the ability of Card9-/- neutrophils to traffic into the infected brain , since the relative ratio of WT:Card9-/- neutrophils remained unchanged before and after infection ( Fig 7A ) . These data indicate that neutrophil-intrinsic Card9 is dispensable for neutrophil recruitment during fungal infection of the CNS . The mixed chimera experiment results ( Fig 7A ) implied that Card9-/- neutrophils are able to upregulate key adhesion molecules that allow their transmigration from the blood into the infected brain . To further test this , we next analyzed whether expression of various adhesion molecules that are required for neutrophil adhesion and rolling on the endothelial surface for transmigration into inflamed tissues was disrupted in the absence of Card9 . Using qRT-PCR on whole brain homogenates , we found that all adhesion molecules analyzed were induced in the brain of WT mice following infection ( Fig 7B ) . While we observed a delayed induction in the expression of α-integrin X and α-integrin L in infected Card9-/- brains , these differences were not evident at 72 hours post-infection . Since β-integrin 2 induction was greater in Card9-/- brains at 72 hours post-infection , we assessed whether its induction in the Candida-infected brain is fungal load-dependent by infecting WT mice with high and low brain fungal burdens . Indeed , induction of β-integrin 2 , ICAM-1 and L-selectin was significantly greater in animals with higher brain fungal burden , whereas induction of α-integrin X , α-integrin L and P-selectin was not fungal load-dependent ( S10 Fig ) . These data suggest that the higher levels of β-integrin 2 and the trend toward higher induction of ICAM-1 observed in Card9-/- brains ( Fig 7B ) may reflect greater tissue fungal proliferation in these mice . Collectively , we did not observe consistent or profound defects in the expression of neutrophil-targeted adhesion molecules in the infected Card9-/- brain ( Fig 7B ) . The mixed chimera experiment results ( Fig 7A ) also suggested that Card9-/- neutrophils do not have a survival defect in vivo . To directly rule out the possibility that reduced neutrophil numbers in the Card9-/- brain is caused by enhanced neutrophil apoptosis and/or death following their recruitment into the CNS , we analyzed the extent of apoptosis and death in neutrophils recruited to the brain at 24 hours post-infection using Annexin-V and 7-AAD staining . We found that the frequency of apoptotic and dead neutrophils was similar in WT and Card9-/- brains ( Fig 7C ) , in line with our patient data that showed that CARD9 is not a survival factor in human neutrophils ( Fig 3E ) . Taken together , these data show that Card9 deficiency does not adversely affect cell-intrinsic neutrophil trafficking to the brain , expression of neutrophil-targeted adhesion molecules , or neutrophil survival during C . albicans infection of the CNS in mice . Since Card9-/- neutrophils did not exhibit a cell-intrinsic chemotactic defect , and as we found our patient’s infected CSF to contain low concentrations of neutrophil-targeted CXC chemokines , we next investigated if this defect could also be modeled in the brains of Card9-/- mice . Thus , we analyzed whole brain homogenates for expression of Cxcl1 , Cxcl2 and Cxcl5 by Luminex array in uninfected and C . albicans-infected mice . We found that both Cxcl1 and Cxcl2 were significantly induced following infection in both WT and Card9-/- mice whereas Cxcl5 was poorly induced ( S11 Fig ) . At 24 hours post-infection , WT and Card9-/- brains had similar levels of these chemokines despite a 10-fold higher fungal load in Card9-/- brain tissue ( S11 Fig ) . At 72 hours post-infection , chemokine levels decreased in WT mice in parallel with falling tissue fungal burden ( Fig 6B ) , whereas chemokine levels in Card9-/- brains remained stable relative to 24 hours post-infection ( S11 Fig ) despite a further dramatic increase in brain fungal burden ( Fig 6B ) . When chemokine production was expressed as a concentration relative to fungal load , Cxcl1 and Cxcl2 levels were dramatically decreased in Card9-/- brains ( Fig 8A ) . We then examined CXC chemokine induction in the kidney and found that Cxcl1 and Cxcl2 were significantly increased in the infected Card9-/- kidneys compared to WT ( S6C Fig ) . Although induction of Cxcl1 and Cxcl2 was to a degree impaired relative to the kidney fungal load in Card9-/- mice ( S6A and S6C Fig ) , renal CXC chemokine induction was sufficient for mediating neutrophil recruitment into the Candida-infected Card9-/- kidney ( S6B Fig ) . Since the IL-17/IL-22 pathway has also been shown to promote neutrophil recruitment into infected tissues [34] , and some CARD9-deficient patients have reduced Th17 cells in the peripheral blood [7 , 9 , 13] , we determined the levels of IL-17 and IL-22 in the brains of WT and Card9-/- mice following infection . However , we found no consistent induction of these cytokines compared to uninfected control brain , at either the mRNA or protein levels . Taken together , Card9-/- mice display a marked insufficiency in the induction of Cxcl1 and Cxcl2 in the brain during C . albicans infection relative to the extent of tissue fungal burden . Since we identified that CXC chemokines were reduced in the CNS of both mice and humans with CARD9 deficiency , we next sought to understand the cellular source of these chemokines in a bid to further define the antifungal immune response within the brain . We used FACS-sorting ( S12 Fig ) to purify the major recruited myeloid cell populations from infected brains ( neutrophils and Ly6Chi monocytes ) , and then used qRT-PCR to quantify relative expression of Cxcl1 , Cxcl2 and Cxcl5 . We also FACS-sorted and analyzed resident cell populations from infected brains , as microglia and astrocytes have been shown to be major producers of cytokines and chemokines during CNS inflammation [35–37] . Firstly , we found that all cell populations expressed Card9 , with the highest levels detected in resident microglia ( Fig 8B ) . Interestingly we also found low , albeit detectable , expression of Card9 in the CD45- population ( Fig 8B ) , which is predominantly made up of resident glial cells as determined by the presence of transcripts for glial acidic fibrillary protein ( Gfap; astrocyte marker ) and myelin oligodendrocyte glycoprotein ( Mog; oligodendrocyte marker ) , while endothelial cells were absent as determined by CD31/CD102 staining ( S12C and S12D Fig ) . We found that neutrophils were the major cellular source of Cxcl2 in the infected brain in WT animals , while CD45- cells predominantly produced Cxcl1 and Cxcl5 ( Fig 8C ) . In infected Card9-/- brains , we found a significant decrease in levels of Cxcl1 and Cxcl2 from purified Card9-/- neutrophils , and further profound decreases in Cxcl1 and Cxcl5 in the Card9-/- CD45- population ( Fig 8C ) . To visualize the cellular sources of Cxcl2 in the Candida-infected brain in vivo , we made use of a recently described transgenic Cxcl2-GFP reporter mouse [38] . In line with our qRT-PCR data , we found that neutrophils in the infected WT brain were the predominant producers of Cxcl2 following infection , whereas Cxcl2 was undetectable in other tested cell populations ( Fig 8D and 8E ) . When we infected Card9-/- mice on this genetic background ( Card9-/- Cxcl2-GFP ) , we found decreased proportion of Cxcl2-expressing Card9-/- neutrophils in the infected brain ( Fig 8F ) . Thus , our data indicate that resident CD45- cells and recruited neutrophils are the major producers of neutrophil-targeted CXC chemokines in the C . albicans-infected mouse brain . Taken together , because CNS tissue does not contain large numbers of resident neutrophils at steady state , our data are consistent with a model by which an early wave of Card9-dependent chemotactic signals by resident cells is followed by a subsequent positive feedback loop of neutrophil-driven cell chemoattraction that is defective in the Card9-/- brain .
In the present study , we show that the antifungal adaptor molecule CARD9 is critical for neutrophil recruitment from the blood into the CNS during fungal infection in mice and humans . CARD9 appears to act by promoting production of neutrophil-targeted chemoattractants in the infected CNS tissue , while cell-intrinsic neutrophil chemotaxis is CARD9-independent . Our conclusions are based on detailed analysis of differences in clinical , microbiological , pathological , immunological and molecular parameters between Card9+/+ and Card9-/- mice , and are corroborated by analogous functional defects identified in a CARD9-deficient patient with Candida meningoencephalitis . Our study is the first to uncover the crucial role of CARD9 on neutrophil accumulation in the CNS during systemic fungal infection in mice and humans , and provides novel insight into the mechanisms of spontaneous susceptibility to CNS fungal infection seen in autosomal recessive CARD9 deficiency . Here , we describe a novel CARD9 missense mutation , c . 170G>A ( p . R57H ) , which further adds to the list of CARD9 mutations associated with the development of systemic fungal infections [7–14 , 18] . Multiple missense and nonsense CARD9 mutations have now been described that reside in both the coiled-coil and CARD domains [7–14] . There has been thus far no correlation between specific CARD9 mutations and development of systemic fungal infections , and therefore more research will be required to define the potential differential functional roles of the coiled-coil and CARD domains of CARD9 on promoting mucosal versus systemic immunity against C . albicans and other fungi . Despite several reports of CARD9-deficient patients with CNS infections , the mechanisms of susceptibility in these patients are not well understood , which prompted us to investigate phagocyte function , since these cells are required components of systemic antifungal immune defense against Candida species [24] . We found that monocytes from the CARD9-deficient patient had a profound defect in pro-inflammatory cytokine production specifically in response to fungal components , similar to other reported findings [7–14] . In addition , we show for the first time a modest , yet significant , defect in killing of C . albicans yeasts by monocytes , the mechanisms of which merit future investigation . Similarly , in agreement with a previous report [7] , we found a defect in killing of unopsonized yeasts by CARD9-deficient human neutrophils , although opsonization could restore this defect . We found no impact of CARD9 deficiency on the killing of hyphal C . albicans forms by either neutrophils or monocytes . These results confirm recently reported findings using CARD9-deficient neutrophils from a patient infected with Phialophora , in which only killing of unopsonized conidia was defective , while hyphal killing was intact [14] . Collectively , the impaired killing of CARD9-deficient neutrophils against unopsonized Candida yeasts has led to the hypothesis that CARD9 deficiency may predispose humans to CNS infection since opsonization is suboptimal in this tissue . However , because the predominant Candida morphology in the infected brain in mice and humans is hyphae ( Fig 1A , Fig 6C and S4 Fig ) [9 , 24] , and because phagocyte hyphal killing is intact in these patients , this current explanation appears insufficient to fully account for why these patients develop spontaneous CNS fungal disease . Strikingly , we found that CARD9 deficiency resulted in a profound absence of neutrophils in the infected CNS ( Fig 3 , Fig 6 and S4 Fig ) ; instead , eosinophils and lymphocytes were the predominant leukocytes in the infected CSF , in agreement with recently published findings in other CARD9-deficient patients with Candida CNS infection [7 , 9] . This profound lack of neutrophil accumulation in the infected CNS is in stark contrast to the robust accumulation of neutrophils in the Candida-infected CNS of mice and humans without CARD9 deficiency [24 , 29 , 39] . In our CARD9-deficient patient , this CNS-specific neutropenia could not be explained by a decrease in circulating neutrophils or neutrophil survival , nor a cell-intrinsic chemotaxis defect . Instead , we found that neutrophil-targeted CXC chemokines were not induced in the infected CSF . In patients with neutrophilic meningitis , levels of IL-8 in the CSF have been reported to reach up to 45 , 000 pg/mL ( median , 13 , 900 pg/mL ) [30] , and the infected CSF has been shown to be chemotactic toward neutrophils ex vivo [40] . Instead , our patient’s CSF lacked chemotactic activity toward neutrophils ex vivo and had approximately 300 pg/mL of IL-8 , which is within the range of concentrations seen in the CSF of uninfected control individuals [30] , and less than 20 pg/mL of CXCL1 , CXCL2 and CXCL5 . This is in contrast to the induction of CXC chemokines in the Candida-infected CSF of another patient without CARD9 deficiency who developed meningitis post-Ommaya reservoir placement at NIH , whose CSF was chemotactic toward neutrophils ex vivo . Therefore , our data indicate that CARD9 is required for proper production of these chemoattractants in response to CNS fungal infection . However , more work assessing CSF leukocyte immunophenotype and measuring these chemoattractants in the CSF of additional patients with CNS fungal infections , with and without CARD9-deficiency , is needed . To directly examine the role of CARD9 in CNS antifungal defense , we moved to a mouse model of systemic candidiasis using Card9-/- mice [41] . We show for the first time that these animals develop significant brain fungal infection with a profound absence of neutrophil accumulation in the infected CNS . Furthermore , we also found that Card9-/- neutrophils could mobilize from the bone marrow into the blood , thus demonstrating a Card9-dependent defect of cell trafficking from the blood into the CNS , which is tissue- and fungus-specific . In line with our human data , we showed that Card9-/- neutrophils did not have a survival defect nor were they intrinsically defective in trafficking from the blood into the Candida-infected brain , since they were able to mobilize into WT brains in mixed bone marrow chimera experiments . Since our human data suggested that neutrophil-targeted chemokine production depended on CARD9 , we investigated the production and source of these molecules in the infected CNS using our mouse model . We found that production of neutrophil-targeted chemokines was grossly suboptimal in the Card9-/- brain relative to the tissue fungal burden . These data suggest that the lack of neutrophils in the Card9-deficient CNS is caused , at least in part , by impaired chemoattractant production . Recent work has demonstrated dependency on both Card9 and IL-1R/Myd88 signaling for neutrophil-targeted chemokine production and neutrophil accumulation in the Aspergillus-infected mouse lung [38 , 42] . Specifically , Card9-/- mice infected with Aspergillus recruited normal numbers of neutrophils to the lung in the early phase of the infection driven via IL-1R/Myd88–dependent CXC chemokine production , whereas at the later stages of the infection a 40% decrease in neutrophil accumulation was observed , which was due to decreased CXC chemokine production by Card9-/- hematopoietic cells [38 , 42] . In stark contrast to this partial dependency on Card9 for neutrophil recruitment during pulmonary aspergillosis , we show here that Card9 deficiency almost abolishes neutrophil recruitment to the CNS throughout the entire course of C . albicans infection . Taken together , these data further underscore the tissue-specific nature of antifungal innate immune responses [24] and provide a potential explanation why CARD9-deficient patients , including our patient reported here , have not been reported to have increased susceptibility to pulmonary infections with ubiquitous inhaled molds . We next investigated the cellular source of neutrophil-targeted chemokines in the CNS , in a bid to further understand how neutrophils are recruited to this site during fungal infection . We found that neutrophils were the predominant source of Cxcl2 and to a lesser extent of Cxcl1 , in agreement with previous reports that neutrophils can create positive-feedback loops for their continued recruitment into fungal-infected tissue [32] . We next investigated if Card9 was required for this phenomenon , and found that Card9-/- neutrophils produced less Cxcl1 and Cxcl2 . Interestingly , we also found significant decreases in Cxcl1 and Cxcl5 transcript levels in the CD45- resident cell population purified from Card9-/- brains , which we found were positive for transcripts encoding glial markers Gfap and Mog . Therefore , our data point to a two-step chemokine-driven neutrophil recruitment process in the WT CNS , initially promoted by resident stromal cells and followed by cell-intrinsic neutrophil chemokine production . In Card9-/- brains , an initial ‘hit’ of defective chemokine production by the resident CD45- compartment may lead to poor neutrophil recruitment , which appears to be further exacerbated by the absence of neutrophil-mediated Cxcl1 and Cxcl2 production in the Card9-/- brain . Future studies should also aim to examine other neutrophil-targeted chemoattractants that may also contribute to Card9-dependent neutrophil recruitment into the infected CNS , as well as their myeloid and non-myeloid cellular sources and their Card9-dependent organ-specific localization in tissue . In particular , our work has indicated that the resident cells in the brain are integral for neutrophil-targeted chemokine production , and their identification and Card9-dependent functions within antifungal immunity is an area of ongoing investigation in our laboratory . Additionally , future work should also focus on whether CARD9 is equally critical for neutrophil recruitment to the CNS in CARD9-deficient patients and mice infected with other fungal pathogens , such as the recently reported Exophiala [10] . In summary , we have identified a novel critical role for CARD9 in tissue- and fungus-specific neutrophil recruitment to the CNS during C . albicans infection in mice and humans . CARD9 is important for the production of chemoattractants in the CNS by neutrophils and resident cells , whereas other neutrophil-intrinsic functions such as phagocytosis , oxidative burst , killing , chemotaxis and survival , appear largely CARD9-independent . Our work highlights the importance of the CLR/CARD9 immune pathway in the neutrophil-dependent control of C . albicans infection of the CNS , and forms the foundation for devising immune-based therapies for bypassing CARD9 in the production of neutrophil-targeted chemokines such as via the potential direct intrathecal delivery of these molecules in infected CARD9-deficient patients .
The patient , her relatives and the healthy donors were enrolled in protocols approved by the National Institute of Allergy and Infectious Diseases and National Cancer Institute Institutional Board Review ( IRB ) committees , and provided written informed consent for participation in the study . This study was conducted in accordance with the Helsinki Declaration . Animal studies were performed in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health , under the auspices of protocol LCID14E approved by the Animal Care and Use Committee of the National Institute of Allergy and Infectious Diseases . Animals were euthanized by cervical dislocation following administration of ketamine/xylazine cocktail . DNA was harvested from Ammonium-Chloride-Potassium ( ACK ) -lysed whole blood from the CARD9-deficient patient , her parents and two siblings using the Gentra Puregene Blood DNA isolation Kit ( Qiagen ) per the manufacturer’s instructions . Genomic amplification was performed in 15 μL reactions using Platinum PCR SuperMix High Fidelity ( Life Technologies ) , 0 . 625 mM of each primer and 10–200 ng of DNA . Cycling conditions were 95°C for 3 minutes followed by 35 cycles of 95°C for 20 seconds , and 68°C for 2:45 . PCR products were purified using ExoSAP-IT ( USB products by Affymetrix ) , sequenced using Big Dye Terminators v3 . 1 ( Life Technologies ) , cleaned up using Performa DTR Ultra spin plates and run on an ABI 3730XL . Amplification and sequencing primers are shown in S1 Table . PBMC were harvested from whole blood by gradient centrifugation using Lymphocyte Separation Media ( Lonza ) , according to the manufacturer’s instructions . PBMC were washed in PBS and quantified prior to use in stimulation assays and flow cytometry ( see below ) or downstream sorting of CD14+ monocytes . CD14+ monocytes were magnetically sorted using negative selection via the Monocyte Isolation Kit II ( Miltenyi ) according to the manufacturer’s instructions . Purity and viability of sorted monocytes was >84% and >92% , respectively . Neutrophils were isolated using 3% Dextran in 0 . 85% sodium chloride and red blood cells lysed using sequential exposure to 0 . 2% and 1 . 6% NaCl solutions . Purity and viability of isolated neutrophils exceeded 95% . CD14+ monocytes and neutrophils were quantified prior to use in FACS or Western blot analyses or downstream functional assays , including internalization assays , killing assays , survival studies and chemotaxis studies ( see below ) . CSF was centrifuged and cells were washed , then stained with a Live/Dead fluorescent dye ( Invitrogen ) for 10 minutes in PBS at 4°C , followed by Fc receptor blockade with Fc Blocking Reagent ( Miltenyi ) in FACS buffer ( PBS supplemented with 0 . 5% BSA and 0 . 01% sodium azide ) at 4°C . Surface antigen staining was then performed by incubating cells with fluorochrome-conjugated ( eFluor 605 NC , PE-Cy7 , APC , APC-eFluor 780 , Alexa Fluor 700 , eFluor 450 , FITC , PE , PerCP-Cy5 . 5 , Biotin/Streptavidin Brilliant Violet 570 ) antibodies against human CD45 ( HI30 ) , CD56 ( CMSSB ) , CD3 ( SK7 ) , CD19 ( HIB19 ) , CD11b ( ICRF44 ) , CD11c ( 3 . 9 ) , CD123 ( 6H6 ) ( eBioscience ) ; CD14 ( M5E2 ) , CD16 ( 3G8 ) , HLA-DR ( G46-6 ) ( BD Biosciences ) ; CD141 ( M80 ) ( BioLegend ) for 30 minutes on ice . After incubation , the cells were washed 3 times with FACS buffer and sample acquired using a 5-laser BD LSRFortessa , equipped with BD FACS Diva software ( BD Biosciences ) . FlowJo ( TreeStar ) was used for the final analysis . Cell numbers were quantified using PE-conjugated fluorescent counting beads ( Sperotech ) . To determine whether CARD9 deficiency affects the ability of PBMCs to produce pro-inflammatory cytokines and chemokines , we used a Luminex-based assay . In brief , 5 x 105 PBMCs from healthy donors or the CARD9-deficient patient were incubated in duplicate in a round-bottom 96-well plate ( Corning Inc . ) at 37°C in a 5% CO2 incubator in RPMI 1640 containing 10% fetal bovine serum ( FBS , Gibco ) , 100 U/mL of penicillin , and 100 μg/mL of streptomycin ( unstimulated ) , or RPMI 1640 with 10% FBS/antibiotics containing LPS ( 100 ng/mL ) , or heat-killed C . albicans SC5314 yeasts ( 1 x 106 /mL ) . After 48 hours of stimulation , PBMCs were pelleted and the supernatant was collected and stored at -80°C until analysis . Luminex analysis was done via a multiplex bead array assay with antibodies and cytokine standards to generate known concentration curves ( R&D Systems , Peprotech ) . Individual Luminex bead sets ( Luminex ) were coupled to cytokine-specific capture antibodies according to manufacturer’s protocols and biotinylated polyclonal antibodies were used at twice the recommended concentrations for a classical ELISA according to the manufacturer’s instructions . The assay was run with 1200 beads per set of cytokines in a volume of 50 μL . The plates were read on a Luminex MAGPIX platform where more than 50 beads were collected per bead set . The median fluorescence intensity of the beads was then measured for each individual bead , which was analyzed with the Millipex software using a 5P regression algorithm . To determine whether CARD9 deficiency affects the capacity of peripheral blood neutrophils and/or CD14+ monocytes to internalize C . albicans , 5 x 104 cells from healthy donors or the CARD9-deficient patient were incubated in a polystyrene round-bottom FACS tube ( BD Falcon ) in a 37°C shaking water bath for 60 minutes in RPMI 1640 containing 100 U/mL of penicillin and 100 μg/mL of streptomycin with dTomato-expressing Candida that had been suspended in RPMI 1640 with antibiotics without ( unopsonized ) or with 5% human serum ( opsonized ) at a cell:yeast ratio of 1:25 or 1:1 , respectively . Samples were then stained with a FITC-conjugated anti-Candida antibody ( cat #21164; Abcam ) and APC-conjugated anti-CD14 ( for monocytes ) or anti-CD15 ( for neutrophils ) antibodies and subsequently washed . FACS was performed on a 5-laser LSRFortessa . The percentage of Candida internalization by myeloid cells was then determined by initially gating on APC+ monocytes or neutrophils and then by defining the percentage of APC+ cells that had internalized Candida ( APC+PE+FITC- ) versus those that had Candida bound on the cell surface ( APC+PE+FITC+ ) . To determine whether CARD9 deficiency affects the capacity of peripheral blood neutrophils and/or CD14+ monocytes to kill Candida , we used the alamarBlue-based fluorescence assay as previously described [22] . Briefly , neutrophils and monocytes were harvested from peripheral blood as described above , and 5 x 104 cells from healthy donors or the CARD9-deficient patient were incubated in flat-bottom 96-well plates ( Genesee Scientific ) for 2 . 5 hours with opsonized or unopsonized Candida yeasts or 3-hour pre-grown pseudohyphae at a cell:Candida ratio ranging from 1:4 to 4:1 ( see figure legends ) . The wells were then treated with 0 . 02% Triton X-100 in water to lyse the neutrophils/monocytes , washed twice with PBS and incubated with 1x alamarBlue ( Invitrogen ) in PBS for 18 hours at 37°C . Fluorescence was measured using a POLARstar OPTIMA plate reader ( BMG Labtech ) . Killing was calculated by comparing the fluorescence of Candida incubated with neutrophils or monocytes with that of Candida incubated without myeloid cells . Neutrophils from healthy donors and the CARD9-deficient patient were harvested as described above . To determine whether CARD9 deficiency impairs survival of neutrophils , cells were cultured for 3 or 6 hours at 37°C in a 5% CO2 incubator in RPMI 1640 without FBS ( serum starvation ) or in RPMI 1640 with 10% FBS with tumor necrosis factor-α ( 50 ng/mL ) or live Candida albicans SC5314 ( 1 x 105 cells/mL ) or in infected CSF from the CARD9-deficient patient to assess the extent of cell apoptosis and death . Cells were stained with Annexin V and propidium iodide ( BD Biosciences ) per the manufacturer’s instructions , and flow cytometry was performed on a 5-laser LSRFortessa . Chemotaxis assays were performed on freshly isolated neutrophils using the EZ-TAXIScan ( ECI , Inc . , Japan ) or a modified Boyden chamber ( NeuroProbe ) . In brief , 5 x 103 neutrophils were placed in each well of the EZ-TAXIScan and 1 . 0 μL of either buffer or fMLF ( 5x10-8 M ) or IL-8 ( 5x10-8 M ) was added to the opposing well . Images of cellular migration were captured every 30 sec for 60 min at 37°C and individual cells ( 10 picked at random ) were tracked digitally using ImageJ software . The paths of the migrating cells were plotted with the position of each cell at t = 0 anchored at the origin . Using the coordinates obtained at each time point , the migratory path of each cell was resolved into a random migrational vector ( orthogonal to the direction of the chemoattractant , along the x-axis ) and a directed migrational vector ( parallel to the direction of the chemoattractant , along the y-axis ) . In the absence of chemoattractant , the directed migrational vector should be equivalent to the random migrational vector . For studies analyzing chemotaxis to several chemoattractants simultaneously , chemoattractants ( fMLF , LTB4 , C5a , and IL-8 ) were prepared at the indicated concentrations in PBS with 0 . 1% human serum albumin and loaded in triplicate to the bottom well of a 96-well ChemoTx disposable chemotaxis plate ( Neuro Probe , Inc . ) . Neutrophils were labeled with calcein-AM ( 10 μg/mL ) and 7 . 5 x104 added on top of the wells , separated from the chemoattractants by a 5 μm polycarbonate filter . Plates were incubated for 45 min at 37°C , and the fluid in upper chamber removed . Cells attached to the underside of the filter were detached by washing in 2 mM EDTA . After removing the filter , fluorescence of each well was determined in a multi-well fluorescent plate reader ( SpectraMAX Gemini EM ) . The number of migrating cells was determined using a standard curve made with known quantities of labeled cells . To test whether CSF from our CARD9-deficient Candida-infected patient or CSF from a healthy uninfected donor or CSF from our CARD9+/+ Candida-infected patient post-Ommaya reservoir placement or CSF from two patients with non-neutrophilic ( lymphocytic ) meningitis was chemotactic itself toward WT and/or CARD9-deficient neutrophils , the NeuroProbe protocol was used , using a 50% dilution of CSF as the chemoattractant . Neutrophils and PBMCs were harvested from healthy donors and the CARD9-deficient patient as described above . Cells were first stained with Live/Dead fluorescent dye ( Invitrogen ) for 10 minutes on ice , followed by anti-CD16/32 for 10 min to block Fc receptors , and then by APC-conjugated anti-human CD14 ( clone M5E2; BD Biosciences ) . Cells were then fixed with 2% paraformaldehyde ( Affymetrix ) at room temperature for 10 minutes and subsequently permeabilized with 1X FACS permeabilizing solution 2 ( BD Biosciences ) according to the manufacturer’s instructions . The cells were then incubated for 30 minutes with rabbit monoclonal anti-CARD9 ( clone EPR6489; Origene ) or monoclonal rabbit IgG isotype control ( clone EPR25A; Abcam ) , washed and then stained with PE-conjugated secondary anti-rabbit antibody ( eBioscience ) for 30 minutes . Samples were acquired using a BD LSRFortessa . Neutrophils and CD14+ monocytes were harvested from healthy donors and the CARD9-deficient patient as described above . Cells were washed once with ice-cold PBS and lysed in buffer containing 50 mM HEPES , 50 mM NaCl , 10% glycerol , 0 . 5% Nonidet P-40 , 2 mM EDTA , and a Protease and Phosphatase Inhibitor Cocktail ( cat# 78440; Thermo Scientific ) . Cell lysates were centrifuged at 13 , 200 g for 10 minutes at 4°C , and equal protein amounts from the supernatant were resuspended in SDS loading buffer with 5% β-mercaptoethanol for Western blot analysis . Proteins were analyzed by SDS-PAGE and transferred to Immobilon P , polyvinylidene difluoride membrane ( Millipore , Billerica ) . Membranes were blocked in TBS-T containing 5% nonfat dry milk , and then incubated with polyclonal rabbit anti-human CARD9 ( cat# 12892-1-AP; ProteinTech Group , Chicago , IL ) and rabbit monoclonal antibody against GAPDH ( cat# 2118; Cell Signaling Technology ) in TBS-T containing 5% BSA for 2 hours at room temperature under continuous agitation . Membranes were then washed 3 times with TBS-T , and incubated with horseradish peroxidase–conjugated anti-rabbit secondary antibody ( Southern Biotech ) in TBS-T containing 5% nonfat dry milk for 1 hour at room temperature . Membranes were subsequently washed , and detection of immunoreactive bands was performed using the SuperSignal West Fempto Maximum Sensitivity Substrate chemiluminescence ( ECL ) detection kit according to the manufacturer’s instructions ( Pierce Biotechnology ) . PBMCs from healthy donors and the CARD9-deficient patient were isolated as described above and cultured in RPMI 1640 with 10% FBS , 2mM L-glutamine , 100 U/mL of penicillin , and 100 μg/mL of streptomycin ( Gibco ) at 37°C in a humidified 5% CO2 incubator . Intracellular staining for IL-17A was performed on PBMCs stimulated for 6 hours with PMA ( 20 ng/ml; Sigma ) and ionomycin ( 1 μM; Life Technologies ) in the presence of Brefeldin A ( 10 μg/ml; Sigma ) . FITC-labeled anti-CD4 ( clone RPA-T4; eBioscience ) was added for the final 30 minutes of culture before the fix and permeabilization step . Cells were then washed with PBS and then fixed and made permeable with Foxp3 staining set ( eBioscience ) according to the manufacturer's instructions . Cells were then incubated with PE-labeled anti-IL-17A ( clone eBio64CAP17; eBioscience ) for 1 hour at 4°C , and acquired on a FACSCanto ( BD Biosciences ) . 8–12 week old female C57BL/6 ( Taconic ) and Card9-/- [41] mice were maintained in individually ventilated cages under specific pathogen-free conditions at the National Institutes of Health ( Bethesda , MD , USA ) . Cxcl2-GFP reporter animals [38] were bred and housed in the Memorial Sloan Kettering Cancer Center Comparative Medicine Shared Resources under specific pathogen-free conditions . Candida albicans strain SC5314 was used for all infections . Yeast was serially passaged 3 times in YPD ( yeast extract , bacto-peptone and dextrose ) broth , grown at 30°C with shaking for 18–24 hours at each passage . Yeast cells were washed in PBS , counted , and injected intravenously via the lateral tail vein . Each animal received 0 . 7–1 . 3 x 105 yeast cells ( details for doses used in individual experiments are presented in the figure legends ) . At 24 or 72 hours post-infection , animals were euthanized and following analyses performed: brain fungal burdens; FACS analysis on blood , brain and bone marrow cells; quantification of chemokines by Luminex array; histopathology; cell sorting; RNA isolation from brain tissue followed by qRT-PCR analysis . WT mice were given 100 μg of 1A8 or 2A3 ( isotype ) antibody ( Bio X Cell ) intravenously at 24 hours prior to infection , and again at the time of infection . Antibody-treated mice were infected with 1 x 105 CFU of C . albicans SC5314 , and analyzed at 24 hours post-infection for brain fungal burden and FACS analysis ( see below ) . 6–8 week old recipient C57BL/6 . SJL ( CD45 . 1+ ) mice were irradiated with 900 rad and left to rest for 4 hours . Bone marrow from gender-matched C57BL/6 . SJL and Card9-/- ( CD45 . 2+ ) donor mice was isolated from the femurs , washed and counted using Trypan blue exclusion , and resuspended as a 1:1 ratio for transfer . Flow cytometry was used to confirm the ratio of CD45 . 1+ to CD45 . 2+ cells prior to use . 5 x 106 total bone marrow cells were transferred to each irradiated recipient mouse intravenously via the lateral tail vein approximately 4 hours post-irradiation . Recipient animals were given trimethoprim/sulfamethoxazole in the drinking water for the first 4 weeks of reconstitution , and then switched to antibiotic-free water . Mice were left to reconstitute for a total of 8 weeks post-transfer , and chimera status was assessed by flow cytometry on a sample of peripheral blood . Chimeras were then infected with 1 . 3 x 105 CFU of SC5314 , and euthanized for analysis at 24 or 72 hours post-infection . Brains and 300 μL peripheral blood were stained for flow cytometry analysis as above . WT and Card9-/- neutrophils ( CD45+ CD11b+ Ly6Ghi ) were separated based on expression of congenic CD45 . 1 and CD45 . 2 markers and the ratio between them calculated . At 24 or 72 hours post-infection , animals were euthanized and brains and kidneys were weighed , homogenized in PBS , and serially diluted before plating onto YPD agar supplemented with Penicillin/Streptomycin ( Invitrogen ) . Colonies were counted after incubation at 37°C for 24–48 hours . Leukocytes were isolated from brain , kidney , blood and bone marrow using previously described methods [24 , 43] , resuspended in PBS and stained with fluorochrome-conjugated antibodies as described for the human samples . Anti-mouse antibodies used in this study were: CD45 ( 30-F11 ) , CD11b ( M1/70 ) , CD11c ( N418 ) , MHC Class II ( M5/114 . 15 . 2 ) , CD45 . 1 ( A20 ) , CD3 ( 145-2C11 ) , CD19 ( eBio1D3 ) , F480 ( BM8 ) , all from eBiosciences , and CD45 . 2 ( 104 ) , Ly6G ( 1A8 ) , Ly6C ( AL-21 ) , all from BD Biosciences . Brain-isolated leukocytes were Fc-blocked and stained with fluorochrome-conjugated antibodies as described above , washed twice with PBS , and resuspended in Annexin V binding buffer ( BD Biosciences ) . Samples were incubated at room temperature for 15 minutes with 5 μl each of FITC-conjugated Annexin V and 7-AAD per the manufacturer’s instructions ( BD Biosciences ) . FACS was performed on a 5-laser LSRFortessa . Brains and kidneys were weighed and then homogenized in PBS/0 . 05% Tween20 and a protease inhibitor cocktail ( Roche ) . Cell debris was removed by two sequential centrifugation steps ( 3600 rpm for 5 minutes , 13200 rpm for 10 minutes ) , and the resulting supernatant filtered through 0 . 2 μm filters , snap-frozen on dry ice and stored at -80°C until analysis . Samples were analyzed by Luminex assay , as described above , and cytokine/chemokine concentrations were determined per gram of tissue . Brains were removed from infected mice at the indicated time points and fixed in 10% formalin for 24 hours before embedding in paraffin wax . Tissue sections were stained with periodic acid-Schiff ( PAS ) and hematoxylin and eosin ( H&E ) . WT animals were infected with 1 . 3 x 105 CFU SC5314 and euthanized at 24 hours post-infection . Brains were isolated and leukocytes stained as above with sterile antibodies . Neutrophils ( CD45hi CD11b+ Ly6Ghi Ly6Cint ) , Ly6Chi monocytes ( CD45hi CD11b+ Ly6Chi Ly6G- ) , microglia ( CD45lo CD11b+ Ly6G- Ly6C- ) and stromal cells ( CD45- ) were FACS-sorted into sterile sorting buffer ( HBSS supplemented with 2 mM EDTA , 10% FCS , 100 U/mL penicillin , 100 μg/mL streptomycin ) using a FACS Aria instrument . Purity of cells , on average , were as follows: neutrophils , 90%; monocytes , 88%; microglia , 95%; stromal cells , >99% ( S12 Fig ) . Cells were then centrifuged ( 1500 rpm 5 minutes , 4°C ) and resuspended in Trizol for RNA purification . 3–5 animals were pooled for individual sorts , and 3–6 independent sorts were performed in total . Cxcl2-GFP reporter and non-transgenic WT littermates were infected as described above with 1 . 3 x 105 CFU SC5314 intravenously and analyzed at 24 hours post-infection . Brains from infected mice were processed as previously described [24] , and stained for FACS analysis as described above . Cxcl2-GFP expression was determined in the major myeloid populations ( neutrophils , Ly6Chi monocytes and microglia ) by comparison with WT ( GFP- ) cells gated in the same way . RNA was extracted from either sorted brain myeloid cells or homogenized brain tissue using Trizol ( Invitrogen ) and the RNeasy kit ( Qiagen ) per the manufacturer’s protocol . Purified RNA was used as a template for cDNA generation using the qScript cDNA SuperMix kit ( Quanta Biosciences ) with oligodT and random primers . Quantitative PCR was performed by SYBR Green ( PerfeCTa SYBR Green FastMix ROX; Quanta BioSciences ) or TaqMan detection ( PerfeCTa qPCR FastMix ROX; Quanta BioSciences ) with the 7900HT Fast Real-Time PCR System ( Applied Biosystems ) . All qPCR assays were performed in duplicate and the relative gene expression of each gene was determined after normalization with GAPDH transcript levels using the ΔΔCT method . TaqMan primers/probes ( Sell , Selp , Icam1 , Itgax , Itgb2 , Itgal , Gfap , Mog , Card9 , Il17a , Il22 , Gapdh ) were predesigned by Applied Biosystems . Primers for SYBR Green detection were as follows: Gapdh: FOR 5’-aactttggcattgtggaagg , REV 5’-acacattgggggtaggaaca; Cxcl1: FOR 5’-actgggattcacctcaagaa , REV 5’-tctccgttacttggggacac; Cxcl2: FOR 5’-aagtttgccttgaccctgaa , REV 5’-aggcacatcaggtacgatcc; Cxcl5: FOR 5’-gaaagctaagcggaatgcac , REV 5’-gggacaatggtttccctttt . Community-acquired methicillin resistant Staphylococcus aureus USA300 ( CA-MRSA , LAC strain from F . DeLeo , Rocky Mountain Laboratories , NIAID ) was grown in brain heart infusion broth ( BHI , Difco Laboratories , Detroit , Mich . ) at 37°C with shaking at 230 rpm for 18 hours . The culture was then centrifuged at 3000 rpm for 10 minutes at room temperature and the pellet washed twice with PBS . WT and Card9-/- mice were infected intravenously with 2 x 107 CFU via the lateral tail vein , and animals euthanized at 48 hours post-infection and analyzed as described above . Statistical analyses were performed using GraphPad Prism 6 . 0 software . Details of individual tests are included in the figure legends . In general , data was tested for normal distribution by Kolmogorov-Smirnov normality test and analyzed accordingly by unpaired two-tailed t-test or Mann Whitney U-test . In cases where multiple data sets were analyzed , two-way ANOVA was used with Bonferroni correction . In all cases , P values <0 . 05 were considered significant . | CARD9 is a molecule expressed by mammalian immune cells and is centrally positioned downstream of several C-type lectin receptors , which sense fungi . The critical role of CARD9 in the activation of antifungal host defense has been highlighted by the demonstration that human mutations that disrupt the function of CARD9 are associated with the development of spontaneous life-threatening fungal infections , many of which target the central nervous system ( CNS ) . However , why CARD9-deficient patients develop fungal disease in the CNS remained unclear . Here , we show that CARD9 is required for the recruitment of neutrophils , an immune cell critical for antifungal host defense , to the fungal-infected CNS in both mice and humans . Furthermore , we show that CARD9-deficient mice and humans are unable to produce the molecules ( chemoattractants ) needed to recruit neutrophils to the site of infection , thus identifying the mechanism for the striking absence of these cells in the infected CNS . Our work helps explain why CARD9 deficiency associates with CNS fungal disease , and furthers our understanding of how antifungal immunity operates in this complex tissue . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | CARD9-Dependent Neutrophil Recruitment Protects against Fungal Invasion of the Central Nervous System |
Recent studies have suggested that the thermodynamic stability of mRNA secondary structure near the start codon can regulate translation efficiency in Escherichia coli , and that translation is more efficient the less stable the secondary structure . We survey the complete genomes of 340 species for signals of reduced mRNA secondary structure near the start codon . Our analysis includes bacteria , archaea , fungi , plants , insects , fishes , birds , and mammals . We find that nearly all species show evidence for reduced mRNA stability near the start codon . The reduction in stability generally increases with increasing genomic GC content . In prokaryotes , the reduction also increases with decreasing optimal growth temperature . Within genomes , there is variation in the stability among genes , and this variation correlates with gene GC content , codon bias , and gene expression level . For birds and mammals , however , we do not find a genome-wide trend of reduced mRNA stability near the start codon . Yet the most GC rich genes in these organisms do show such a signal . We conclude that reduced stability of the mRNA secondary structure near the start codon is a universal feature of all cellular life . We suggest that the origin of this reduction is selection for efficient recognition of the start codon by initiator-tRNA .
Synonymous mutations are frequently used as a neutral baseline to detect selection pressures at the amino-acid level [1] . Yet many mechanisms are now known that cause selection pressure on synonymous sites . Translationally preferred codons are selected for accurate and efficient translation in bacteria , yeast , worm , fly , and even in mammals [2]–[13] . Selection on synonymous sites acts to increase the thermodynamic stability of DNA and RNA secondary structure [14]–[18] , to improve splicing efficiency [19]–[21] , and to assist protein co-translational folding [22]–[27] . Synonymous codon choice can also affect translation initiation . Most of the sequence elements that control translation initiation ( e . g . the Shine-Dalgarno sequence in prokaryotes and the 5′ cap and Kozak consensus sequence in eukaryotes ) are located in 5′ untranslated regions ( UTRs ) [28]–[30] , where high conservation and AU-richness have been observed [31]–[34] . Yet Zalucki et al . found a significant bias towards usage of the AAA codon at the second amino acid position in Escherichia coli secretory proteins [35] . They proposed that selective pressure for high translation-initiation efficiency accounts for this codon usage bias . Other studies have demonstrated altered expression levels in E . coli after changing synonymous codons in the region downstream from the start codon [36]–[39] . Kudla et al . synthesized a library of 154 genes of green fluorescent protein ( GFP ) that had random changes at synonymous sites without any change in the amino-acid sequence [40] . They found that the GFP expression level varied 250-fold across the library . In this library , the stability of mRNA secondary structure near the start codon explained more than half of the variation in expression level: mRNAs with more stable local structure in this region had reduced protein expression [40] . These observations suggest that translation initiation is facilitated by a choice of synonymous codons that destabilize local mRNA secondary structure . Here , we analyzed the local mRNA secondary structure at the 5′ end of the coding region in 340 species , including bacteria , archaea , fungi , plants , fishes , birds , and mammals . We used computational methods to predict the thermodynamic stability of local mRNA secondary structure in sliding windows downstream from the start codon , and used permutation tests to assess deviation from random expectation . We addressed the following questions: ( i ) Is there a selection pressure on synonymous sites to reduce the stability of local mRNA secondary structure at the translation-initiation region ? ( ii ) Is such a selection pressure a general characteristic for all organisms ? ( iii ) Does 5′ mRNA stability correlate with GC composition , codon usage bias , or gene expression level ? ( iv ) In prokaryotes , does 5′ mRNA stability vary with the optimal growth temperature of the organism ?
We calculated the local folding energy ( ) along the mRNA sequence using a sliding window of 30 nucleotides ( nt ) in length , moving from the start codon to the downstream nucleotide in steps of 10 nt ( for a total of 13 windows ) . To quantify the deviation from expectation given a gene's amino-acid sequence and codon usage bias , we also calculated for 1000 permuted mRNA sequences . We obtained permuted sequences by randomly reshuffling synonymous codons within each gene . We then calculated a -score , , by comparing the of the real mRNA segment to the distribution of values of the permuted sequences ( see Materials and Methods ) . measures the extent to which local mRNA stability deviates from expectation . A positive means that local mRNA stability is reduced , and a negative means that it is increased . For each window , we calculated a genome-wide mean by averaging the corresponding values over all genes in a genome . We performed the sliding window analysis in 340 species , which included 276 bacteria , 35 archaea , 11 fungi , 2 plants , 2 insects , 4 fishes , 2 birds , and 8 mammals . Figure 1 shows an example of the mean for 13 windows in E . coli . We observed a significant positive deviation of from zero in the first two windows ( t-test: in both cases ) . The positive values of suggest selection for reduced mRNA stability at the 5′ end of the coding region . The values further downstream decrease quickly and we observe negative values in most downstream windows . Most species we studied showed a similar pattern to the one we observed in E . coli ( Figure S1 and Table S1 ) , except for plants and warm-blooded animals ( birds and mammals ) . There was a clear increase in mean for windows close to the start codon . Because the at the very start of the coding sequence generally showed the strongest signal of reduced mRNA stability , we will focus on this value for the remainder of this study . In the following , we refer to the at the very start of the coding sequence also as the 5′ . In prokaryotes , 262 out of 276 bacteria and 28 out of 35 archaea showed a positive 5′ ( Figure 2 ) . In eukaryotes , 10 out of 11 fungi , 1 out of 2 plant , both insect species , and all four fish species we analyzed showed this pattern as well . All warm-blooded animals showed a negative 5′ throughout the coding sequence . We list the mean and standard error of for all species and all windows in Table S1 . To investigate whether window size affected our results , we redid our analysis for four species ( two bacteria , one archaeon , one fungus ) using sliding windows of 20 nt and 40 nt , respectively . Results for these two window sizes were comparable to those obtained with a window size of 30 nt ( Figures S2 and S3 ) . For the same four species , we also recalculated controlling for dinucleotide content , by using the DicodonShuffle algorithm [41] . The results were virtually unchanged compared to our standard shuffling method . ( Figure S4 ) . We found substantial variation in the mean 5′ among different species ( Figure 2 ) . Therefore , we next aimed to identify the determinants of 5′ in different genomes . We first considered genomic nucleotide content . We compared the mean 5′ in each genome to the genome's GC content in coding sequences . We observed a strong positive correlation between the mean 5′ and the genomic GC content ( Spearman's , ) when plants and warm-blooded animals were excluded ( Figure 3 ) . Genomes with higher GC content had comparatively less stable mRNA secondary structure at the translation-initiation region . Since the thermodynamic stability of RNA secondary structure tends to be correlated to the RNA's GC content , we also looked into local deviations in a gene's GC content . We calculated , which measures the deviation in GC content in a 30 nt window relative to the average in the gene ( see Materials and Methods ) . We found a negative correlation between genomic GC content and the mean of the first window ( Spearman's , ) . Thus , in GC-rich genomes , the sequence regions immediately downstream of the start codon were particularly GC poor ( Figure S5 ) . Because mRNA stability was reduced only near the translation-initiation region , we expected that similarly GC content was reduced only near the start codon . Therefore , the correlation between genomic GC content and mean should decrease for windows further downstream . We found that indeed the correlation declined continuously and reached approximately zero at the window ( Figure S6 ) . Besides the statistical measure , we also considered directly ( Table S2 ) . We found that the mean value of the first window ( 5′ ) varied greatly among different species and was largely determined by genomic GC content ( Spearman's , , Figure S7 ) . As expected , mRNA stability increased with increasing genomic GC content . In fact , we found similar relationships for windows further downstream , but the stability at the 5′ end of the mRNA was generally lower than the stability further downstream ( Figure S8 ) . Moreover , the difference between the mean 5′ and the mean of the downstream windows increased with increasing genomic GC content . As an example , Figure S9 shows the relationship between the mean GC content and the difference in mean between the first and tenth window ( Spearman correlation , , excluding birds and mammals ) . In summary , the results for generally mirrored the ones for . For prokaryotes , we analyzed whether the 5′ correlated with the optimal growth temperature ( Figure 4 ) . We found a significant negative correlation ( Spearman's , ) . Similarly , the difference in between the first and the tenth window declined significantly with temperature ( Spearman's , ) . Thus , prokaryotes living in colder environments tended to have comparatively less stable mRNA secondary structure at the translation-initiation region . We found no correlations between temperature and either genomic GC content ( Spearman's , ) or the in the first window ( Spearman's , ) . The lack of a correlation between temperature and genomic GC content agrees with the results of Ref . [42] . In the previous subsections , we considered the mean 5′ over all genes in a genome . But we expected that there should also be variation in mRNA stability among genes within one genome . Therefore , we next investigated the potential within-genome factors that may affect mRNA stability near the start codon . We first considered gene GC content . We compared the mean 5′ between genes with the highest 5% and the lowest 5% GC content in each species . In almost all genomes , including birds and mammals , the mean 5′ in GC-rich genes was higher than it was in GC-poor genes ( Figure 5 and Table S3 ) . The differences became weaker as we considered windows further downstream ( Figure S10 ) . Interestingly , even though the whole-genome mean 5′ was negative in birds and mammals , GC-rich genes in these animals showed a positive 5′ ( Figure 5 ) . Next we considered codon usage bias . We used the effective number of codons ( ENC ) to measure the codon usage bias of each gene [43] . Lower ENC values indicate stronger bias . By comparing the bottom 5% of genes with the lowest ENC to the top 5% of genes with the highest ENC , we found that , in most species , genes with stronger codon bias had higher 5′ ( Figure S11 and Table S3 ) . Finally , we tested whether the reduction in 5′ mRNA stability increased with gene expression level . We compared the mean between genes with the highest 5% and the lowest 5% expression level in E . coli , Drosophila melanogaster , Saccharomyces cerevisiae , and Homo sapiens . In all species except H . sapiens , the mean for the highest-expressed genes tended to be higher than that for the genes with the lowest expression level ( Figure 6 ) . Since GC content , codon bias , and expression level all correlated with , we tried to determine whether these quantities are independent sources of variation . We carried out a principle component regression [44] and found that GC content , ENC , and gene expression level contributed nearly equal to variation in in E . coli , S . cerevisiae , and D . melanogaster ( Figure S12 ) . In human , GC content and ENC , but not gene expression level , contributed to the variation .
We have completed a broad survey of mRNA stability near the translation-initiation region of protein-coding genes . We have considered the complete genomes of 340 species , including bacteria , archaea , fungi , plants , insects , and vertebrates . We have found a general tendency for reduced mRNA stability in the first 30–40 nt of the coding sequence . In this region , mRNA stability tends to be less than expected given a gene's amino-acid sequence and codon-usage bias . Experimental work had previously suggested that increased local mRNA stability at the translation-initiation region could prevent efficient translation initiation and hence decrease gene expression level [38] , [40] . We have found that there is variation in the extent to which mRNA stability is reduced both among and within genomes . Among genomes , GC content of coding sequences is a major predictor of the reduction in mRNA stability . The higher the GC content , the larger the reduction in mRNA stability at the 5′ end of the coding sequence ( i . e . , the larger 5′ ) . For prokaryotes , the optimal growth temperature also predicts 5′ . The lower the optimal growth temperature , the larger the reduction in mRNA stability . Within genomes , 5′ also increases with increasing GC content . In addition , it increases with increasing codon usage bias and gene expression level . The region with reduced mRNA stability is located right downstream from the start codon and has a length of 30 to 40 nt ( the first two windows in our analysis ) . This region is similar to the one identified by Kudla et al . [40] . Kudla et al . studied primarily a library of sequences encoding green fluorescent protein , but they also carried out a computational analysis of mRNA stability across the E . coli genome . They found that across the genome , was significantly more positive ( indicating reduced stability ) in the region from nt to than immediately downstream [40] . Our work shows that Kudla et al . 's observation applies to most organisms with known genomes , including bacteria , archaea , and both single- and multi-celled eukaryotes . Further , by focusing on scores relative to the expectation in permuted sequences , our analysis excludes biases such as amino-acid content or preferred-codon usage as the cause of this signal . Past the first two windows , decayed quickly towards a negative asymptotic value . Thus , mRNA stability near the start codon is less than expected , but elsewhere in the gene it is generally higher than expected . The latter result is comparable to observations made by Chamary and Hurst [16] in the mouse genome and by Seffens and Digby [15] in individual genes from several species . Interestingly , for many organisms , in windows 3 to 5 dips below the negative asymptotic value further downstream ( Figure S1 and Table S1 ) . This behavior seems to reflect a selection pressure for particularly stable local mRNA structure right after the translation-initiation region . This increased stability may compensate for the reduced mRNA stability in the translation-initiation region . Previous works have identified AT-biased translation enhancers in prokaryotes [37] , [45]–[47] and preferred nucleotide sequences regulating translation in eukaryotes [48] within the first 30 to 40 nt of the coding sequence . The mechanism by which these sequence motifs work is not currently known . We suggest that the primary mechanism may be destabilization of the mRNA structure near the start codon . By contrast , some motifs work by known mechanisms unrelated to RNA secondary structure . For example , alanine is preferred at the second amino-acid position in highly expressed proteins in several organisms [49] and its codon might bind to a complementary sequence in the 18S ribosomal RNA [50] . We found that the higher the GC content of a genome , the more was mRNA stability reduced at the translation-initiation region . This finding makes thermodynamic sense . GC-rich RNAs tend to fold into more stable structures than AU-rich RNAs , simply because a GC pair has three hydrogen bonds whereas an AU pair has only two . Thus , assuming that selection targets the same low 5′ mRNA stability in all organisms , we would expect that the decrease in stability is larger in GC-rich RNAs , simply because they start from a more-stable baseline . Whether selection actually targets the same low 5′ mRNA stability cannot be determined by our analysis . We found that the mean 5′ increased with increasing GC content . This increase could imply either that organisms with higher GC content can tolerate a higher 5′ mRNA stability or that the selection pressure to reduce 5′ mRNA stability in those organisms is counterbalanced by other selective forces or mutation pressures that increase GC content . For prokaryotes , we addressed the question whether the optimal growth temperature affects 5′ . Thermodynamics predict that the lower the temperature at which an organism grows , the stronger should mRNA stability interfere with translation initiation . In agreement with this prediction , we found that the optimal growth temperature correlated negatively with 5′ . The organisms growing at the lowest temperatures showed the biggest reduction in mRNA stability at the beginning of the coding sequence . This result was independent of the relationship between 5′ and genomic GC content . In our data set , the optimal growth temperature was not correlated with GC content . Even though some authors have argued that GC content correlates with temperature [51] , [52] , more recent studies have disputed this finding [42] , [53] . Our results agree with these more recent studies . Within individual genomes , the reduction of mRNA stability at the translation-initiation region was greater in GC-rich genes than in GC-poor ones . Besides GC content , we found that codon usage bias and gene expression level correlated with 5′ . Because codon usage bias is correlated with gene expression level , in particular in fast-growing microbes [2] , [3] , [8] , [11] , these two correlations likely reflect the same underlying effect . The correlation with expression level mirrors the general observation that evolutionary constraints tend to increase with gene expression level [8] , [11] , [13] , [54]–[58] . Whether expression level , codon usage bias , and GC content contribute independently to 5′ is unclear . These three quantities tend to all be correlated with each other , and we cannot easily disentangle which of these quantities is most important for reduced 5′ . For example , in mammals , high GC content in genes can increase mRNA levels through increased efficiency of transcription or mRNA processing [59] . Using principal component regression , we showed that in E . coli , yeast , and fly , the three quantities codon usage bias , GC content , and gene expression level all contribute equally to reduced 5′ , whereas in humans only GC content and codon-usage bias seem to contribute . We found reduced mRNA stability near the start codon in a wide range of organisms , including both prokaryotes and eukaryotes . Yet warm-blooded animals ( birds and mammals ) showed no such trend on the whole-genome level , even though their genomic GC content is well within the range in which we found reduced mRNA stability in bacteria , archaea , fungi , insects , and fishes . We believe that our finding for birds and mammals was caused by the isochore structure of their genomes [60] . Gene GC content in these organisms ranges from 20% to 95% and is much more varied than in organisms without isochores . The whole-genome average of 5′ may not be meaningful in organisms with isochores . When we considered only to top 5% most GC-rich genes , we did find a moderate signal of reduced mRNA stability in these organisms as well . What is the biological mechanism that links mRNA stability near the start codon to efficient protein translation ? There are two possibilities . First , strong local mRNA secondary structure could interfere with ribosome binding . Second , it could interfere with start-codon recognition . We believe that the currently available evidence favors the latter explanation . In prokaryotes , ribosome binding occurs at the Shine-Dalgarno sequence , located a few nucleotides upstream from the start codon [28] . Kudla et al . [40] showed that synonymous mutations near the start codon can regulate protein expression . They concluded from computational modeling that the primary determinant of protein expression was the stability of local mRNA secondary structure near the start codon , not occlusion of the Shine-Dalgarno sequence by RNA secondary structure [40] . In eukaryotes , translation initiation follows a scanning mechanism . The 40S ribosomal subunit enters at the 5′ end of the mRNA and migrates linearly until it encounters the first AUG codon [30] . If synonymous mutations near the start codon could affect ribosome entry at the 5′ cap , there should be a correlation between 5′ UTR length and mRNA stability near the start codon . The further away the start codon is from the 5′ cap , the less should local mRNA stability near the start codon affect ribosome entry . However , we did not find such a relationship , neither within genomes nor among genomes ( data not shown ) . Therefore , we suggest that both in prokaryotes and in eukaryotes , reduced mRNA stability at the translation-initiation region primarily facilitates efficient start-codon recognition .
We collected the genomes for 276 bacteria , 35 archaea , 11 fungi , 2 plants , 2 insects , 4 fishes , 2 birds , and 8 mammals . The genomic sequences of the bacteria , archaea , fungi , plants , and insects were downloaded from the NCBI FTP server ( ftp://ftp . ncbi . nih . gov/ ) , while the sequences of the vertebrates were obtained from Ensembl ( http://www . ensembl . org/ ) . We only considered coding sequences longer than 50 codons . We collected previously published expression data for four species: for E . coli , we obtained gene expression levels measured in mRNAs per cell from Ref . [61]; for S . cerevisiae , we used expression data from Ref . [62]; for D . melanogaster , we used as expression level the geometric mean of expression data from different tissues obtained in Ref . [63]; and for H . sapiens , we also measured expression level as the geometric mean of expression among different tissues [64] . We obtained optimal growth temperature data for 80 bacteria and 14 archaea from Ref . [42] , which is a collection from multiple sources , including original publications , American Type Culture Collection , German Collection of Microorganisms and Cell Cultures , and Prokaryotic Growth Temperature Database . We calculated RNA folding energies using the RNAfold program in the Vienna package [65] , [66] . We used default settings: folding occurred at ; GU pairs were allowed; unpaired bases could participate in at most one dangling end; energy parameters were as reported in Ref . [67] . We evaluated only the minimum-free-energy structure . is the change in free energy from the unfolded state to this structure . If synonymous selection acts on mRNA folding near the start codon , then on average the secondary structure in this region should be less stable for the naturally occurring sequence than for permuted sequences . For each gene , we randomly reshuffled synonymous codons among sites with identical amino acids , to control for amino-acid sequence , codon usage bias , and GC content . We repreated this process 1000 times to obtain 1000 permuted sequences for each gene . For the wild-type sequence and each permuted sequence , we then calculated local mRNA folding energies in a sliding window of 30 nt ( 20 nt and 40 nt were also used in some species ) . To determine the deviation of the wild-type sequence from the permuted ones , we calculated the Z-score of the local mRNA stability ( ) for each sliding window by: ( 1 ) Here , is the folding free energy for the naturally occurring sequence in the window under consideration , is the folding energy of the corresponding window of the permuted sequence , and is the mean of over all permuted sequences . The variable represents the total number of permuted sequences . Here , . Similarly , we evaluated the difference between the local mRNA GC composition of the wild-type sequence and the permuted sequences . The Z-score of local mRNA GC content ( ) for each window can be expressed as: ( 2 ) The definitions for , , and are analogous to , , and but refer to GC content rather than to free energy of folding . | Synonymous mutations are mutations that change the nucleotide sequence of a gene without changing the amino-acid sequence . Because these mutations don't alter the expressed protein , they are frequently also called silent mutations . Yet increasing evidence demonstrates that synonymous mutations are not that silent . In particular , experimental work in Escherichia coli has shown that the choice of synonymous codons near the start codon can greatly influence protein production . Codons that allow the mRNA to fold into a stable secondary structure seem to inhibit efficient translation initiation . This observation suggests that selection should prefer reduced mRNA stability near the start codon in many organisms . Here , we show that this prediction generally holds true in most organisms , including bacteria , archaea , fungi , plants , insects , and fishes . In birds and mammals it doesn't hold true genome-wide , but it does hold true in the most GC-rich genes . In all organisms , the extent to which mRNA stability is reduced increases with increasing GC content . In prokaryotes , it also increases with decreasing optimal growing temperature . Thus , it seems that all organisms have to optimize their synonymous sites near the start codon to guarantee efficient protein translation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/molecular",
"evolution",
"molecular",
"biology/translational",
"regulation",
"genetics",
"and",
"genomics/bioinformatics",
"evolutionary",
"biology/genomics"
] | 2010 | A Universal Trend of Reduced mRNA Stability near the Translation-Initiation Site in Prokaryotes and Eukaryotes |
The segmentation gene network in Drosophila embryo solves the fundamental problem of embryonic patterning: how to establish a periodic pattern of gene expression , which determines both the positions and the identities of body segments . The gap gene network constitutes the first zygotic regulatory tier in this process . Here we have applied the systems-level approach to investigate the regulatory effect of gap gene Kruppel ( Kr ) on segmentation gene expression . We acquired a large dataset on the expression of gap genes in Kr null mutants and demonstrated that the expression levels of these genes are significantly reduced in the second half of cycle 14A . To explain this novel biological result we applied the gene circuit method which extracts regulatory information from spatial gene expression data . Previous attempts to use this formalism to correctly and quantitatively reproduce gap gene expression in mutants for a trunk gap gene failed , therefore here we constructed a revised model and showed that it correctly reproduces the expression patterns of gap genes in Kr null mutants . We found that the remarkable alteration of gap gene expression patterns in Kr mutants can be explained by the dynamic decrease of activating effect of Cad on a target gene and exclusion of Kr gene from the complex network of gap gene interactions , that makes it possible for other interactions , in particular , between hb and gt , to come into effect . The successful modeling of the quantitative aspects of gap gene expression in mutant for the trunk gap gene Kr is a significant achievement of this work . This result also clearly indicates that the oversimplified representation of transcriptional regulation in the previous models is one of the reasons for unsuccessful attempts of mutant simulations .
The segmentation gene network in early Drosophila embryo provides a powerful model system to study the role of genes in pattern formation . This network solves the fundamental problem of embryonic patterning: how to establish a periodic pattern of gene expression , which determines both the positions and the identities of body segments [1] , [2] . The developmental process which performs this task is called segment determination . The fruit fly segments are arranged sequentially along the anterior-posterior axis of the embryo . All segments are determined simultaneously during the blastoderm stage , just before the onset of gastrulation [3] . The segmentation genes have been subdivided into 4 classes based on their mutant phenotype [1] , [2] . The maternal coordinate genes are expressed from the mother and form broad protein gradients in the anterior , posterior or terminal regions of the embryo [4]–[7] . Other genes , which belong to gap , pair-rule and segment-polarity classes , are zygotic , i . e expressed in the embryo . Most of segmentation genes encode transcription factors , which in turn regulate the expression of many other genes , including segmentation genes themselves . It was demonstrated by genetic analysis that segmentation genes form a hierarchical regulatory cascade , in which genes in higher layers ( e . g . maternal coordinate genes ) regulate genes in lower layers ( e . g . gap genes ) , but not vice versa . In addition genes in the same hierarchical level interact with each other . The gap gene system establishes discrete territories of gene expression based on regulatory input from a long-range protein maternal gradients , Bicoid ( Bcd ) and Hunchback ( Hb ) in the anterior and Caudal ( Cad ) in the posterior of the embryo [8] , [9] . Gap genes Kr , kni , hb , gt and tll are expressed in from one to three domains , each about 10–20 nuclei wide [10] . Early gap gene expression of the trunk gap genes Kr , hb , gt and kni is established through feed-forward regulation by maternal gradients , after initial establishment gap domain borders sharpen , moreover both sharpening and maintenance of gap domain boundaries requires gap-gap cross-regulatory interactions [11] . This process is accompanied by the anterior shift of Kr , kni and gt expression domains in the posterior region of the embryo [10] , [12] , [13] . Kr plays a central role in segmental pattern formation as indicated by strong alteration of expression patterns of almost all zygotic segmentation genes in mutants [14]–[16] . Kr null mutants show deletion of thoracic and anterior abdominal segments as well as frequent mirror duplications in the abdomen [15] , [17] . At the level of gene expression this mutation manifests in the large shift of posterior Gt domain , resulting in overlap of positions of posterior Gt and Kni domains [15] , [18] . During sharpening and maintenance stage of gap gene expression Kr acts a repressor of gt and hb [15] , [19] , [20] . The repression of gt , which expression domains are strictly complementary to those of Kr , is strong , while the effect of Kr on hb is more subtle [21]–[24] . It was observed in assays with cell lines carrying reporter constructs that the regulatory effect of Kr is concentration-dependent: Kr monomer is transcriptional activator , while at high concentrations Kr forms a homodimer and becomes a repressor that function through the same target sequence as the activator . However it is difficult to establish whether such an effect occurs at physiologically relevant regulator's concentrations [25] . The segmentation gene network is one of the few examples of developmental networks studied using data-driven mathematical modeling [13] , [26]–[28] . These models fall into two categories . The phenomenological models do not require any a priory information about regulatory mechanism [29] , [30] and try to reconstruct it by solving the inverse problem of mathematical modelling . A major shortcoming of these models is that their parameters have no explicit connection to the genomic DNA sequence . The second modelling approach seeks to extract information about gene regulation from the sequences of cis-regulatory regions and the measured or inferred binding of sequence-specific transcription factors to these elements [26]–[28] , however it still neglects major features of the transcription process , such as chromatin structure and modifications , binding site orientation and proximity to transcription start site , etc . Current simplifications and unknown features limit the predictive power of these models , but more powerful and complex models may be generated in future using better datasets such as in vivo transcription factors occupancy , relative accessibility of different DNA regions , in vivo data on interplay between different transcription factors , nucleosome and chromatin remodelling enzymes . In this paper we apply a phenomenological model known as gene circuits to reconstruct the gap gene network in Kr null mutants . This model considers a row of nuclei along the A-P axis of the embryo . Between nuclear divisions the model describes three basic processes , namely protein synthesis , protein decay and diffusion of proteins between neighboring nuclei of syncitial blastoderm . A few basic assumptions about eukaryotic transcriptional regulation were incorporated into the model . First a sigmoid regulation-expression function was used to introduce regulatory inputs into the model . Secondly , each regulatory interaction can be represented by a single parameter which sign indicates the type of regulatory interaction: activation ( if it is positive ) , repression ( if negative ) , no interaction ( if it is close to zero ) . Third it was assumed that regulatory inputs are additive and independent of each other . The gene circuit models were successfully applied to correctly reproduce the quantitative features of gap gene expression in wild type [12] , [13] . This study revealed five regulatory mechanisms responsible for sharpening and maintenance of gap gene expression domains: broad activation by maternal gradients of Bcd and Cad; gap gene auto-activation; strong mutual repression between gap genes which show complementary expression patterns ( hb and kni; Kr and gt ) ; weaker asymmetric repression between overlapping gap genes ( Hb on gt , Gt on kni , Kni on Kr , Kr on hb and Hb on Kr ) and repression by terminal gene tll at the embryo termini . The asymmetric repression between overlapping gap genes is responsible for shifts of gap gene domains in the posterior region of the embryo . It is important to note that the wild type gap gene circuit model has the predictive power when molecular fluctuations of the input factors are taken into account [31] , [32] . It is evident that to understand the gap gene network we need not only to describe the mechanism underlying its functioning in intact state , but also to comprehend what happens when certain stimuli or disruptions occur . Recently Papatsenko and Levine ( 2011 ) constructed a dynamic model based on a modular design for the gap gene network , which involves two relatively independent network domains with elements of fractional site occupancy . This model requires only 5–7 parameters to fit quantitative spatial expression data for gap gradients in wild type and explained many expression patterns in segmentation gene mutants obtained in studies published mainly in the late 1980s and early 1990s . However these patterns were characterized qualitatively by visual inspection , that may not capture the fine details of gene expression . For example , previous studies based on qualitative visual analysis of gene expression patterns showed that a Kr null mutation results in large shift of posterior Gt domain , overlap of positions of posterior Gt and Kni domains and decrease in the level of gt expression in the second half of cycle 14A [15] . Here we obtained a large dataset on gap gene expression in Kr null mutants and extracted quantitative gene expression data using a data pipeline established previously [33] . The analysis of this data allowed us to characterize the expression of other gap genes at unprecedented level of detail . In particular we showed that the significant decrease in the level of gene expression in the second half of cycle 14A is common to all gap gene expression domains . This novel biological result seems counterintuitive , because genetics studies show that Kr acts as a repressor , and therefore should come under close scrutiny . The most serious limitation of the gap gene circuit models is their inability to correctly reproduce the expression patterns in trunk gap gene null mutants at quantitative level , although a theoretical study had shown previously that such prediction is possible if gene circuits models were fit to simulated , noise-free data [29] and simulating null mutants of the terminal gap genes tll and hkb was successful [12] , [13] , [34] . A variety of reasons could be responsible for the failure , of which , from our point of view , the most important is the oversimplified representation of transcriptional regulation in the model . Indeed , as was already mentioned above , the action of regulator on its target gene is represented by a single parameter , whereas it is well known that the cis-regulatory elements ( CRE ) of segmentation genes often reproduce only one of expression domains of an endogenous gene when placed upstream of a reporter gene [35]–[37] . Moreover different CREs of one gene can have different transcription binding site composition , i . e . different regulatory inputs . For example , computational prediction of transcription factor binding sites showed that regulatory sequences which drive expression of gt in the anterior and posterior domains have different transcription binding site composition: the anterior gt domain has regulatory inputs from Bcd and Kni , while the posterior domain contains inputs from Hb and Cad , which are absent in the sequences responsible for anterior expression [37] . Similar to gt , two CREs essential for hb expression in anterior domain and in central stripe and posterior domain differ in transcription binding site composition [38]–[40] . It is evident that current gene circuits models do not consider the mechanism of gene regulation at such a level of detail . This defect does not interfere with the ability of these models to fit gap gene expression patterns in wild type , however in mutant background with deficient set of regulators the failure of the model to take into account such features may suddenly become essential . To avoid such problems we use a revised model which builds on separate treatment of domains with different regulatory inputs . This is possible by narrowing down the spatial domain of the model and considering only the posterior half of the blastoderm ( region from 47 to 92% embryo length ( EL ) ) , in which each of the trunk gap genes is expressed in one domain . As opposed to previous gap gene circuit models , which have a constant Bcd gradient and did not consider Cad data from late time points just before the onset of gastrulation [12] , [32] , and similar to approach used in [30] , we implement Bcd as a time-variable input and use data on late Cad expression to represent the rapidly changing expression dynamics of these two genes . After cleavage cycle 12 Bcd nuclear gradient starts to decay [41] . Analysis of data from fixed embryos showed that Bcd protein reached its maximal level near the beginning of cycle 14A and thereafter starts to decrease slowly that culminates in an almost twofold decline by gastrulation [10] . From the second quarter of cleavage cycle 14A onward the cad expression in abdominal region start to gradually decrease and by gastrulation cad expression in the posterior region sharpens to a stripe which spans from 75 to 90% EL [10] . The gene circuit models do not require any assumption about regulatory interactions within a gene network . Instead the regulatory topology of the network is obtained by solving the inverse problem of mathematical modeling , i . e . by fitting the model to the data [29] . To obtain the estimates for regulatory parameters that predict a specific network topology in mutants we fitted the model to gap gene expression patterns in wild type and in embryos with homozygous null mutation in Kr gene simultaneously . The logical justification of such an approach is to use the parameters of the wild type gap gene network as specific constraints on regulatory weights in mutants in order to obtain the consistent parameter estimates for both genotypes on one hand and on the other hand to preserve the characteristic features of gene regulation in mutant . The parameter estimates obtained in such a way were further studied by applying identifiability analysis , that confirmed that fitting to two genotypes simultaneously substantially increases the statistical significance of parameter values . We use the modeling framework outlined above to explain the characteristic features of gap gene expression in Kr null mutants and in the posterior half of the blastoderm . In what follows we describe the expression patterns of gap genes in Kr null mutants and analyze quantitative gene expression data extracted from these patterns . We then use these data as input to a new gap gene circuit model . We show that in contrast to earlier models , this model correctly reproduces the characteristic features of gap gene expression in Kr mutants . In particular , it reproduces correctly the greater shift of posterior Gt domain than in wild type and significant decrease in the level of gap gene expression in the second half of cycle 14A . We next obtain the parameter estimates for the model ( and hence the predicted gap gene network topology in wild type and mutant ) and perform identifiability analysis to understand how reliable are these estimates . We study the dynamical behavior of our model and analyze the role of individual regulatory loops in gap gene expression in wild type and mutants . We show that a remarkable transformation of gap gene expression patterns in Kr mutants can be explained by dynamic decrease of activating effect of Cad on a target gene and exclusion of Kr gene from the complex network of gap gene interactions , that makes it possible for other interactions , in particular , between hb and gt , to come into effect . Our model also predicts the derepression of the anterior border of Hb posterior domain in Kr;kni double mutants , that is established in the absence of key repressors . We validate this prediction and show the correctness of network topology inferred in this work .
In wild type Drosophila embryos gap genes are expressed as large intersecting domains along the A-P axis ( Figure 1A ) . In general , all these domains exhibit similar temporal dynamics: after formation they start to grow , reach maximum expression levels around mid-cycle 14A and decline by gastrulation . In the course of cycle 14A gap gene domains change their positions and shift to the anterior [10] . We have shown that the asymmetric gap-gap cross-repression with the posterior dominance is responsible for these shifts [12] . In Kr null mutants gap gene expression is significantly altered ( Figure 1B ) . It has been previously reported that in these mutants the posterior domain of gt is expanded towards the center of the embryo [42] , [43] . We detect that in the course of cycle 14A the posterior domain of gt shifts dynamically on 15% embryo length ( EL ) in the anterior direction and overlaps with Kni domain . Thus , by gastrulation , the difference in position of gt domain in mutants and wild type embryos constitutes approximately 10% EL . The anterior shift of the Kni domain maxima in Kr mutants constitutes only 1 . 8% EL . Hb posterior domain in mutants is formed at the beginning of cycle 14A and shifts on about 3% EL in the anterior direction during this cycle . Thus , the positional dynamics of this domain in mutants and wild type is similar . The level of hb posterior expression in mutants is nearly the same as in wild type until time class 3 , but declines afterwards . By gastrulation it constitutes only a half of the wild type expression level ( Figure 1C , F ) . Gt posterior domain is initially lower than in wild type , it grows up to time class 4 and significantly declines thereafter ( Figure 1D , G ) . The level of kni expression remains constantly low throughout cycle 14A ( Figure 1E , H ) with a slight decrease at the very end of this cycle ( not shown ) . The features of the gap gene expression in Kr mutants described above raise many questions . Namely , the regulatory mechanisms underlying a decrease in gene expression levels , as well as a much larger shift of Gt posterior domain should be explained . In the following sections , we describe our modification of the gene circuit model [12] , [13] , [29] , [44] that correctly reproduces the gap gene expression in Kr mutants and hence can serve as a tool to answer these questions . The gene circuit model used in this work differs from previous implementations in several aspects . First we narrowed down the spatial domain of the model by considering only the posterior half of the blastoderm ( region from 47 to 92% embryo length ( EL ) ) , in which each of the trunk gap genes is expressed in one domain . This allows us to avoid the inherent limitation of the model , in which the action of regulator on its target gene is represented by a single parameter . Secondly , as opposed to previous gap gene circuit models which have a constant Bcd gradient and did not consider Cad data from late time points just before the onset of gastrulation [12] , [32] , we implement Bcd as a time-variable input and use data on late Cad expression to represent the rapidly changing expression dynamics of these two genes at that stage . We used bcd and cad profiles from FlyEx database for cycle 13 and eight temporal classes of cycle 14A as external inputs to our model equations . We used the modeling framework outlined above to explain the characteristic features of gap gene expression in Kr null mutants and in the posterior half of the blastoderm . To obtain the estimates for regulatory parameters that predict a specific network topology in mutants the model was fitted to gap gene expression patterns in wild type and in embryos with homozygous null mutation in Kr gene simultaneously . DEEP method was applied to minimize the sum of squared differences between experimental observations and model patterns [45] , [46] and find all parameters of model equations , i . e . regulatory weights , synthesis rates , decay and diffusion constants , that allow to reproduce the characteristic features of gap gene expression in Kr null mutants as closely as possible . We performed over 200 runs with different initial parameter approximations and control variables . The search space was sampled uniformly for each parameter in the interval defined by biologically relevant limits . Two step procedure was applied to construct the ensemble of parameter sets . On the first stage , the residual mean square ( RMS ) was checked and the sets with RMS less than 5% of the maximal gene expression value ( equals 255 in our data ) were accepted for further analysis . Secondly , we inspected the model expression patterns visually . Consequently , the ensemble of 11 parameter sets was obtained that correctly reproduces the dynamics of gene expression in wild type and mutant embryos , in particular the decrease of gap gene expression levels and the anterior shift of gt domain . To infer the topology of regulatory network we classified the estimates of regulatory weights ( the elements of T and E matrices ) into the following three categories: ‘activation’ ( parameter values greater than 0 . 005 ) , ‘repression’ ( parameter values less than −0 . 005 ) and ‘no interaction’ ( between −0 . 005 and 0 . 005 ) . This leads to a predicted regulatory topology of the network based on which category a majority of parameter estimates falls into ( summarized in Figure 2 ) . There are several networks in the ensemble called consensus networks , in which the signs of regulatory parameters coincide with the predicted network topology inferred from the fits . Figure 2A shows simulation results together with experimental data and Table S1 presents parameters for one of such networks . It is evident that in spite of some patterning defects especially at early stages the model correctly reproduces the dynamics of gene expression in wild type and mutant embryos . Some basic features of the gap gene network topology in wild type and mutant become immediately obvious from inspection of Figure 2B and Table S1 . First , Cad activates zygotic gap gene expression . Second , hb , Kr , kni , and gt show autoactivation . Third , Bcd activates Kr , gt , kni in all parameter sets , however in case of hb it shows activation in approximately the same number of circuits as it shows repression . Fourth , all reciprocal interactions among trunk gap genes are either zero or repressive . An important exception is activation of hb by Gt . Finally tll represses kni and Kr and weakly activates gt . The identifiability analysis was conducted with respect to parameters of the model estimated by fitting to experimental data . The model considers the time evolution of protein concentrations of four gap genes hb , Kr , gt , and kni in two genotypes: wild type and in embryos with homozygous null mutation in Kr gene . The total parameter set that minimizes the cost functional 0 consists of 40 parameters and is denoted as . The set includes four subsets , , , and of 10 parameters each , that describe regulatory action on each target gene . In mutants the model is only fitted to quantitative gene expression data for 3 genes , gt , hb and kni , and hence the parameters from are estimated using data points from wild type embryos only ( half of all data points ) . All the other parameter subsets are estimated from the whole dataset . Due to lack of space we denote the elements of inter-connectivity matrices T and E by single-letter notations of genes , namely , H , K , G , N , B , C , and T stand for gt , Kr , gt , kni , bcd , cad , and tll , respectively . For example , characterizes the regulatory action of Hb on kni . The sensitivity of the model solution to parameter changes is characterized by the size of confidence intervals . The confidence intervals ( 2 ) ( see Methods ) are constructed under the assumption of normally distributed error in data , that is not satisfied for gene expression data . The error in data almost linearly increases with the mean concentration that is typical rather for the Poisson than for normal distribution . To make the error independent of the mean we applied the variance-stabilizing transform to both data and model solution . The transformed objective functional was minimized using the parameter estimates obtained for non-transformed functional as initial values for the optimization procedure . The new solutions were found in a very close vicinity of initial parameter sets . The 11 parameter sets , which minimize the transformed model functional , are given in Table S2 and will be referred to as circuit parameter sets . The newly estimated regulatory weights were classified into regulatory categories as described in subsection Gene Network Topology . This classification results in predicted regulatory topology of the network ( Figure 2C ) , which is largely the same as in Figure 2B , however not all the entries in two tables coincide . The estimates of some parameters not-uniquely determine the type of gene regulation in different circuits , i . e . in some circuits the parameter estimates exceed the threshold 0 . 005 in absolute value , while in the others are below the threshold . It is also true for new parameter sets , however , the number of such circuits is different than those given in Figure 2B . The confidence intervals for individual parameters are constructed in the vicinity of the model solution . The results for one representative circuit are presented in Figure 3 . Most of the values of regulatory parameters are very close to zero , and it is important to make sure whether the value ( more precisely , the sign ) of a regulatory parameter is significant . The hypothesis that the parameter estimate is non-zero is tested as follows: if a confidence interval includes both positive and negative values , the hypothesis is rejected , otherwise , accepted . Our classification method to infer the topology of regulatory network used in this work , was based on comparison of the values of regulatory parameters with the threshold . However , as it has already been mentioned , estimates of some parameters take values , which only exceed the threshold in part of circuits . By exploration of confidence intervals for these parameters we came to the conclusion , that almost all the estimates , that are close in absolute value to the threshold , are insignificant . This result explains the discrepancies between the network topologies presented in Figures 2B and 2C: the conclusions about the type of gene interaction that are based on insignificant parameter estimates are unreliable . The analysis of confidence intervals conducted for all the circuits ( Figures S2 and S3 ) allowed us to refine the predicted regulatory network topology ( Figure 2C ) . We classify parameters as insignificant activation/repression if the parameter estimates are positive/negative in almost all the circuits but their confidence intervals contain zero , and hence the parameter sign cannot be identified . As a result the non-identifiable regulatory parameters are , , , , , , and and therefore we cannot draw any conclusion about these interactions . Interestingly most of these interactions involve Kr as a target gene or Bcd as a regulator of gap gene domains located in the posterior of the embryo . Other regulatory parameters are well identifiable and , hence , the identifiability analysis corroborates the gene network topology drawn from classifying parameter values only . It should be stressed that the confidence intervals provide the full information about the parameter estimates only in case of parameter independency , otherwise the intervals are overestimated . Moreover , strong correlation between parameters may lead to their non-identifiability , because a change in one parameter value can be compensated by the appropriate changes of another parameters and , hence , does not significantly influence the solution . In view of this we investigate the dependencies between parameters using the collinearity analysis of the sensitivity matrix . This method allows to reveal correlated and hence non-identifiable subsets of parameters . The sensitivity matrix defined in Methods was analyzed in the vicinity of 11 points in the parameter space that define the optimal model solutions . The collinearity index ( equation ( 3 ) in Methods ) was computed for all the subsets of dimension k of the parameter set . The threshold value for was chosen equal to 7 . This value in case of corresponded to approximately 99% pairwise Pearson correlation between columns of the sensitivity matrix . The method allowed to detect subsets of dimension 2 and 3 with the collinearity index exceeding the threshold value , i . e . subsets of poorly or non-identifiable parameters . Most of parameter combinations in these subsets were the same for all 11 circuits ( see Table 1 ) . Almost all the pairs of parameters in subsets of dimension 2 belonged to , i . e . were related to Kr target gene . To explain this result we additionally compute the collinearity indices between columns of the upper half of the sensitivity matrix , that only include the partial derivatives computed at 1532 wildtype observations . The method detected much more parameter subsets with collinearity indices greater than the threshold , that included the parameters characterizing the input of all four genes . The parameters of the full model fitted to two genotypes thus are better identifiable than those of the model that solely describes the wild type data . However , the parameters from cannot be identified from the full model as are just estimated from the wild type observations . The subsets of dimension 3 with the highest collinearity indices ( ) are also common for the most of the circuits . Most of these combinations are related to gt and Kr , i . e . , include parameters from and . The two approaches applied to characterize parameter identifiability are closely connected and complement each other . By exploration of confidence intervals we can see to what extent the model solution is sensitive to parameter changes and test the significance of parameter sign , but this method does not give any explanations to the sources of non-identifiabilities . One of such explanations can be provided by collinearity analysis . The correlation between parameters revealed by this approach can clarify insignificance or unreliability of parameter estimates with large confidence intervals . For example , we derive non-identifiability of from the large size of its confidence interval and at the same time the analysis of the sensitivity matrix allows us to detect the subset of parameters and with high mean collinearity index equal to 7 . 76 ( see Table 1 ) . Thus , poor identifiability of can be explained by correlation between two regulatory parameters , that is reflected in their high collinearity index . The gene network topology inferred from both classifying the parameter values and parameter identifiability analysis is presented in Figure 2C . As Figure S4 shows it is largely in agreement with topologies predicted by earlier models [13] , [31] , [34] , [47] . Strong constraints for mutual repression are present for kni and hb , which show complementary expression patterns . Besides , strong repressive action exert both Kr on hb and Hb on gt . Some previous models had predicted the repressive action of Kr on hb [31] , while most showed no interaction [13] , [34] , [47] . Many repressive interactions between gap genes show weaker constraints toward repression , and interestingly we have found very weak or no dynamical constraints for repression of Gt on Kr , the interaction with strong constraint for repression in all wild type gene circuit models [13] , [31] , [34] , [47] , [48] . In addition our model predicts weak repressive interactions between Kni and gt and Kr and kni . In earlier gap gene circuit models the first interaction was predicted as no interaction [13] , activation [34] or activation in about half the circuits , and repression in the other half [31] . The repressive action of Kr on kni is only observed in our model , all other models predicted no interaction between the two genes . In addition in current model Bcd shows activation of hb in approximately the same number of circuits as it shows repression , while in all previous models this interaction was predicted as activation . Weak activation of gt by Tll is now present in 10 parameter sets , while previous results predicted this interaction as repression . Finally our model predicts no interaction between Tll and hb . Some previous models had classified this interaction as activation [31] , [34] , while other predicted it as repression or no interaction [13] , [48] , [49] . Null mutation in Kr gene results in strong alteration of expression patterns of almost all zygotic segmentation genes . In gap gene network this mutation manifests in significant reduction of gap gene expression levels in cycle 14A , as well as in large shift of posterior Gt domain and overlap of positions of posterior Gt and Kni domains . Previous gap gene circuit models fail to correctly model the gap gene expression patterns in the embryos homozygous for null mutation in a trunk gap gene . A new model introduced here correctly reproduces the characteristic features of gap gene expression in Kr null mutants and in the posterior half of the blastoderm . To investigate the mechanism responsible for strong alteration of the expression patterns of gap genes in Kr null mutants we have performed the detailed graphical analysis of gap gene regulation in the posterior of the embryo . This analysis revealed the following regulatory principles . In both Kr mutants and wild type the posterior Hb domain is the last gap domain to form; its expression is initiated in cleavage cycle 13 and the domain retracts from the posterior pole at temporal class 2 of cycle 14A . Later the expression level in the posterior Hb domain increases gradually up to temporal class 7 and diminishes at the very end of cycle 14A in wild type embryos , while in mutants the level of expression is nearly the same as in wild type until time class 3 , but declines afterwards . In the model , combined activating inputs by Cad and Gt are responsible for hb expression in early cycle 14A . At later time ( from time class 3 onward ) hb autoactivation starts to play role and it gradually supplements activation by other factors , which strength decreases in both genotypes . In mutants the positional dynamics of Hb domain resembles that in wild type , while the anterior shift of Gt domain is much larger . Larger anterior shift of Gt domain causes stronger decrease in activation contribution by gt to the Hb domain ( Figure S5 ) . This effect together with smaller autoactivation level may cause the fall in accumulation of Hb in Kr mutants . We do not include hkb , the gene responsible for formation of the posterior boundary of the Hb posterior domain , in our model and therefore only the mechanism underlying the formation of the anterior boundary can be analyzed . In wild type this boundary is formed by joint repression by Kr and Kni , while in Kr mutants Kni is the only repressive input , which strength diminishes with time due to decrease in kni expression level . Gt posterior domain forms in cycle 13 . In wild type embryos the expression of gt in this domain reaches maximum at time class 5 and then declines . In Kr null mutants gt expression is lower than in wild type , it grows up to time class 4 and significantly declines thereafter ( Figure 1D , G ) . In mutants the anterior shift of gt posterior domain is much larger than in wild type: by gastrulation , the difference in position of gt domain in mutants and wild type embryos constitutes 10% EL and Gt domain overlaps Kni domain . Cad and Gt autoactivation contributes activating inputs on posterior Gt domain ( Figure 4 ) . Both in Kr mutants and wild type embryos the strength of Cad activating input decreases by gastrulation , however in mutants this reduction is larger , as Gt domain shifts closer to the anterior end of the embryo against gradient of Cad concentration ( Figure 4C ) . It is obvious that weaker activation of gt by Cad will lead to lower level of gt autoactivation within its domain . Indeed , by time class 7 autoactivation contributes strongly to expression of gt only in wild type . This provides a straightforward mechanism for reduction in the level of gt expression in the posterior of the Kr mutant embryos: the decrease of activating contribution by Cad and diminishing gt autoactivation result in downregulation of gt expression . It should be noted that in mutant a small level of Hb repression is evident across the middle region of the model spatial domain at all times ( Figure 4B ) . This repression is caused by the spurious expression of hb in the region of 60–77%EL ( see Figure 1 ) and could be responsible for decrease in the gt expression level . However the exclusion of this elevated expression from the model ( by setting Gt input into hb expression to zero ) does not lead to increase in the gt expression level ( Figure S6 ) in mutants , that makes it unlikely that Hb repression contributes significantly to the low levels of gt . Kr and Kni repression is involved in the positioning of the anterior boundary of the Gt posterior domain in wild type embryos ( Figure 4 ) . In mutants Kni does not significantly contribute to this boundary formation , that can be accounted for its small expression level . Besides by time class 7 Gt shifts to the anterior border of the model spatial domain . All this precludes the conclusions on the possible mechanisms of the anterior boundary formation of Gt domain in Kr mutants . The posterior boundary of the Gt domain depends almost exclusively on very strong repression by Hb both in wild type embryos and Kr mutants ( Figure 4 ) . The accumulation of Hb in the posterior region causes increase in both levels and extent of this repression over time . This in turn leads to an anterior shift of Gt domain . In Kr null mutants the lack of gt repression by Kr and very weak repression of gt by Kni allows Gt posterior domain to move further than in wild type to the territory of kni expression . Thus , the mechanism underlying the shift in posterior Gt domain in Kr mutants is equivalent to those of other gap domains in wild type embryos: shift happens because of the almost absence of repression by the adjacent anterior domain ( Kni ) , while it becomes increasingly repressed posteriorly ( by Hb , in this case ) . In wild type embryos kni expression is first detected in cycle 13; it reaches maximum by temporal class 5 of cycle 14A . In Kr mutants the level of kni expression remains constantly low throughout cycle 14A ( Figure 1E , H ) and the anterior shift of the Kni domain maximum constitutes only 1 . 8% EL . In the model cad and kni autoactivation provides activating inputs on Kni domain . Both in Kr mutants and wild type embryos the strength of Cad input decreases by gastrulation , however in mutants this decrease is stronger , happens faster and is accompanied by diminishing kni autoactivation . By temporal class 7 autoactivation of kni is present at significant level only in wild type . Similar to Gt domain a small level of hb repression evident across the middle region of the model spatial domain ( Figure 5B ) is unlikely to contribute to the low levels of kni as the exclusion of this spurious expression from the model does not lead to increase in the kni expression level ( Figure S6 ) . In wild type embryos Kr repression is responsible for positioning the anterior boundary of Kni domain , while in Kr mutants this boundary forms outside the model spatial domain . The posterior boundary of this domain depends on repression by Gt , Hb and Tll in both genotypes , however Tll repression is only retained in a region posterior of 80% EL . In Kr mutants Gt repression spreads into a territory where kni expression domain forms , preventing the increase of gene expression level in this domain . The analysis performed above points on the central role of hb in gap gene regulation in embryos . To support the validity of this prediction the in silico experiments were done . In these experiments , instead of setting to zero ( as is usually done to model mutant genotype ) , we multiplied it to the scaling coefficient , which gradually decreases from 1 to 0 , and inspected changes in gap gene expression in the posterior of the embryo . It is evident in Figure 6 that decrease in Kr regulatory input is most important for hb dynamics . Besides , this experiment demonstrates the monotonous dependence of change of gap protein concentrations on , that may be an additional argument for validity of numerical results , obtained independently in two genotype modeling . It should be stressed that though such an experiment does not has a special biological sense , it makes it possible to predict , which component of the network is subject to biggest impact by mutation . The last hardly could be revealed in biological experiments .
The failure of gene circuit models to correctly reproduce expression patterns in gap gene null mutants could be due to a variety of reasons: it is possible that our data does not correctly reflect the absolute concentrations of gap gene proteins and some scaling of expression patterns in the data is necessary or production delays may be required in the model to eliminate premature initiation of gap-gap gene interactions . Alternatively , a more detailed consideration of molecular mechanisms responsible for gap gene expression may be required for mutant simulation . At present we do not have a satisfactory understanding of such mechanisms for any of the gap genes , however some important regulatory principles emerged from experiments with reporter constructs , DNAse protection assays , Chip-chip experiments and large-scale computational screens to identify and analyze gap gene enhancers . These experiments demonstrated that cis-regulatory elements ( CRE ) of segmentation genes often reproduce only one element of an endogenous gene expression pattern when placed upstream of a reporter gene [35]–[37] and that different CRE of one gene can have different transcription binding site composition , i . e . different regulatory inputs . For example , three gt CREs drive reporter gene expression in the posterior ( gt_ ( -3 ) ) and distinct anterior domains ( gt_ ( -6 ) , gt_ ( -10 ) ) , respectively , while another element ( gt_ ( -1 ) ) reproduces endogenous gt expression in both anterior and posterior domains [36] , [37] . Moreover computational screens predict that the anterior and posterior gt domains have different regulatory inputs . It is currently unclear how gt CREs interact in regulation of the endogenous gt gene , however it is obvious that the representation of the gap gene regulatory interactions by a single parameter in the gene circuit model is hardly suitable for theoretical description of such a complex mechanism and should be substituted by more realistic representation . As a first step in this direction we introduced a revised model which builds on gene circuit method but treats domains with different regulatory inputs separately . A straightforward way to implement such a modification is to narrow down the spatial domain of the model by considering only the posterior half of the blastoderm , in which each of the trunk gap genes is expressed in one domain . Here we demonstrated that the new model correctly reproduces the characteristic features of gap gene expression in Kr mutants , the greater shift of posterior Gt domain than in wild type and significant decrease in the level of gap gene expression in the second half of cycle 14A in particular . The successful modeling of expression patterns in a mutant for a trunk gap gene is a significant achievement of this work . This result also clearly indicates that the oversimplified representation of transcriptional regulation in the previous models is one of the reasons for unsuccessful attempts of mutant simulations . Most of previous gap gene circuit models represent Bcd as a time-constant gradient and did not consider Cad data from late time points just before the onset of gastrulation [12] , [32] . It was reasonable to implement such an approach when the mechanism for precise positioning of segmentation gene expression domains was investigated , however an intriguing feature of Kr mutants is the reduction in the level of gap gene expression in the second half of cycle 14A , a phenomenon which understanding requires a precise consideration of activators responsible for gap gene expression . As we have shown before [10] , Bcd protein reaches its maximal level near the beginning of cycle 14A and thereafter starts to decline slowly , while Cad expression in abdominal region starts to gradually decrease from time class 3 onward . Accordingly in the model we implement Bcd as a time-variable input and use data on late Cad expression to represent the rapidly changing expression dynamics of these two genes . This allows us to demonstrate that decrease of activating input by Cad and weakening of autoactivation are responsible for reduction in the level of gt and kni expression in the posterior of the Kr mutant embryo . In gene circuit models the regulatory topology of the network is obtained by solving the inverse problem of mathematical modeling , i . e . by fitting the model to the data [29] . To obtain the estimates for regulatory parameters that predict a specific network topology in mutants we fitted the model to gap gene expression patterns in wild type and in embryos with homozygous null mutation in Kr gene simultaneously . The rationale behind such an approach is that , as it was shown in the Parameter identifiability and Correlations section , using the parameters of the wild type gap gene network as specific constraints on regulatory weights in mutants substantially increases the statistical significance of fitted parameter values . Our results demonstrate the existence of parameter sets describing gap gene expression in two genotypes simultaneously and thus the applicability of the gap gene circuit formalism to model genotypes of trunk gap gene mutants . One finds hard to say whether the overfitting is the reason why these parameter sets were not discovered during the fit to the wild type data alone . Overfitting is defined by a fine balance between the number of model parameters and the level of details used to describe the system . The qualitative models usually require small amount of parameters . However , when the data under modelling becomes more quantitative , the number of parameters usually increases , and in the general case there are no methods to find the optimal number of parameters , exceeding which will lead to overfitting . We treat the possible overfitting problem by applying the practical identifiability analysis of the found parameter values . The parameter estimates obtained in such a way were further studied by applying identifiability analysis . Two approaches were used . First the sensitivity of the model to parameter changes and identifiability of parameters in the vicinity of the model solution were analyzed on the basis of confidence intervals of parameter estimates . This analysis showed that the most of regulatory parameters are well identified and their estimates can be used to make conclusions about the type of gene interaction . Secondly , as parameter non-identifiability can be a consequence of their strong correlation we applied the collinearity analysis of the sensitivity matrix to reveal the subsets of correlated parameters . We found that non-identifiability of some parameters detected by the method based on confidence intervals can be explained by the correlations between different parameters . Our analysis also demonstrated that parameters of the model fitted to two genotypes are better identifiable than those of the model fitted to wild type data only . Our quantitative analysis of gap gene expression in mutants confirms and extends results from earlier studies . It was previously reported that in cycle 14A the shift of posterior Gt domain is much larger than in wild type and that this domain overlaps Kni domain [15] , [18] . It was also demonstrated that the level of Kni expression in Kr mutants is reduced . Here we showed that by gastrulation the difference in position of gt domain in mutants and wild type embryos constitutes approximately 10% EL . Contrary to posterior Gt domain , the anterior shift of the Kni domain maxima in Kr mutants constitutes only 1 . 8% EL , as a result these domains overlay each other . The level of kni expression remains constantly low throughout cycle 14A . Decrease in the expression levels of both kni and gt in Kr mutants was shown in earlier qualitative studies [15] , [42] , [50] , [51] . High temporal resolution of our dataset enabled us to find out that the decline in gene expression level in the second half of cycle 14A turned out to be an intrinsic property of all gap domains . The regulatory mechanisms for expression of the trunk gap genes in the posterior of the embryo predicted by our model r summarized in Figure 7A: ( 1 ) Cad activates zygotic gap gene expression . ( 2 ) Autoactivation is involved in maintenance and sharpening of hb , Kr , kni , and gt domains . ( 3 ) Trunk gap genes either repress each other or do not interact . An important exception is activation of hb by Gt . ( 4 ) Bcd activates Kr , gt , kni in all parameter sets , however identifiability analysis showed that this activation is insignificant . In case of hb Bcd shows activation in approximately the same number of circuits as it shows repression . ( 5 ) In the posterior terminal region of the embryo Tll represses Kr and does not interact with hb . In general this regulatory principles are in agreement with the results from previous studies ( see Figure S4 ) , however some differences exist . This is not surprising , if to consider that contrary to previous studies the model was fitted to two genotypes simultaneously . The rationale behind such an approach is to use the parameters of the wild type gap gene network as specific constraints on regulatory weights in mutants in order to obtain the consistent parameter estimates for both genotypes on one hand and on the other hand to preserve the characteristic features of gene regulation in mutant . Previous models predict the mutual repression between non-overlapping gap genes hb and kni , as well as gt and Kr . Indeed , our model showed strong constraints for mutual repression between kni and hb , however identifiability analysis classified the action of Gt on Kr as insignificant repression . This discrepancy can be explained by the fact that in our model the parameters corresponding to Kr gene are estimated using data points from wild type embryos only and therefore their identifiability is inferior to that of parameters estimated from the whole dataset . The non-identifiability of parameters describing the action of Bcd on gap genes in Kr-deficient gap network may be explained by the exclusion from the model of the anterior half of the blastoderm , where Bcd contributes stong activating inputs on the anterior and central gap gene domains [12] , [52] . Bcd activating inputs on posterior domains are smaller and the interactions of gap genes come into play to dynamically position the posterior gap domains . In the gap gene circuit models the motif which includes Tll is the most variable component of the gap network ( see Figure S4 ) . The network reconstructed in this work constitutes no exception to this pattern: our model predicts no interaction between Tll and hb , while some previous models had classified this interaction as activation [31] , [34] or predicted it as repression or no interaction [13] , [48] , [49] . In previous models Tll exerts repressive action on gt and kni , however in our model these parameters are non-identifiable . In addition our model predicts repression of Kr by Tll , however in the other model [34] this parameter was classified as non-identifiable . Interestingly that in spite of the differences in gene regulation between the models discussed above the asymmetric cascade of cross-repressive interactions between gap genes with overlapping expression domains is preserved in the current two genotype model . As is evident from inspection of Table S1 the regulatory weights and are larger than the reciprocal weights and in all circuits , while the regulatory weight is larger than the reciprocal weight in 9 out of 11 circuits . These asymmetrical interactions lead to anterior shifts in domain positions both in wild type and mutant , as will be discussed below . The regulatory mechanisms predicted by the model are mainly in agreement with experimental evidences . During cleavage cycle 14A both Cad and Bcd continue to activate gap genes , however the Bcd gradient starts to rapidly decay about 10–15 min before gastrulation [41] , [53] . The evidence for autoactivation of gap genes is not so clear . In Kr and gt mutants expressing non-functional proteins Kr domain is narrowed and weakened [16] , and the intensification of Gt domains during cycle 13 is delayed [42] . Besides computational studies predict that both Kr and Gt bind to some of their own regulatory elements [37] . The autoactivation is not required for expression of the posterior Hb domain [21] . Repressive feedback between hb and kni and gt and Kr is suggested by many experimental results [15] , [42] , [54] . There is experimental evidence for additional repressive interactions between gap genes with overlapping expression domains . Repression of gt by Hb is supported by the fact that the posterior Gt domain fails to retract from the posterior pole of the embryo around mid-cycle 14A [15] , [42] , while no gt expression can be detected in embryos over-expressing hb [42] . The central Kr domain expands posteriorly into regions with reduced or lacking kni activity in mutants [22] , [55] , [56] . There is a Kni binding site in the Kr regulatory region , which overlaps with a Bcd activator site [50] . It has been proposed that Kr is required for kni activation , however , this effect turned out to be indirect [51] , [57] . Kr and hb are the only pair of overlapping gap genes that show mutual repression , however there is some ambiguity in the genetic evidence . Some authors have reported a posterior expansion of the anterior Hb domain in Kr mutants [54] , [58] . Repression of Kr by Hb is suggested by an anterior expansion of the central Kr domain in hb mutants [22] , [55] , [56] , [59]–[61] and multiple Hb binding sites have been identified in the Kr regulatory region [60] . Evidences on repression of kni by Gt and repression of gt by Kni are ambiguous [15] , [42] , [51] , [62] , e . g . a posterior expansion of the abdominal Kni domain was reported in one study [42] , this effect was not seen in another [62] . Terminal gap genes have strong repressive effects on trunk gap gene gt , kni and Kr [38] , [57] , [63] . In contrast the posterior domain of Hb is present and expanded to the anterior in embryos over-expressing tll [64] . This suggests that Tll activates hb expression in its posterior domain , however , this interaction is probably indirect , since posterior Hb is present in tll;kni double mutants . The important corollary that follows from the inferred topology ( Figure 7A ) is the prediction that in Kr , kni double mutants the anterior border of Hb posterior domain will not be properly set as both regulators responsible for formation of this border are deficient . We confirmed this prediction in experiment ( Figure 7B ) . This experimental result strongly supports the model . To explain mechanisms responsible for alteration of the gap gene expression pattern in Kr mutants we implement the analysis of regulatory loops and study the dynamical change of Cad contribution to a target gene regulation in wild type and mutants . Cad is the main activator of gap genes in the posterior of the embryo . This analysis revealed two mechanisms responsible for alteration of the posterior gt expression pattern in Kr mutants . First , gt expression level decreases because the activating effect of Cad on gt diminishes with time ( Figure 4 ) . Secondly , two interactions from Kr to gt and hb responsible for formation of the anterior borders of Gt and Hb posterior domains correspondingly are impaired in embryos ( Figure 7A ) , that makes it possible for Gt posterior domain to move forward to the anterior due to repression by Hb . The mechanism responsible for reduction of kni expression level in Kr mutants is essentially the same as described for gt posterior domain: Kni level reduces because the activating effect of Cad on kni decreases with time ( Figure 5 ) and because the interaction between Kr and gt is deficient that makes it possible form Gt repression to spread into a territory where kni expression domain forms ( Figure 7A ) . It should be noted that the difference in Cad regulatory effect on the posterior gt and kni expression between mutant and wild type is not very large ( see Figures 4C and 5C ) . This fact could raise doubt on the role of Cad in the reduction of gap gene expression levels in mutants . However recently it was demonstrated that only a 3% overlap exists between transcription factor occupancy and gene response to TF knockout [65] , [66] . This and other results ( discussed in detail in [67] ) point that the relations between TF concentration and function are non-linear and that weak regulatory events and small differences in regulation may play a biologically significant role in the quantitative control of complex biological processes . Our model can also explain the mechanism , which provides for decrease of the hb posterior domain expression level: such a reduction happens because Gt stops to activate hb due to its shift ( Figure S5 ) . However it should be noted that the activation of hb by Gt predicted by all gene circuit models [13] , [31] , [34] , [48] is currently not supported by the literature . In our model hb activation by Gt leads to the elevated and spurious expression of hb in the region of 60–77% EL ( see Figure 1 ) . The exclusion of this elevated expression from the model does not cause an increase in both gt and kni expression levels ( Figure S6 ) , that makes it unlikely that Hb repression contributes significantly to the low levels of these domains in mutant . We note finally that the reduction of gap gene expression levels is peculiar not only for Kr mutants . For example hb expression ( see Figure 7B ) and gt posterior domain levels [26] are reduced in kni mutants . In wild type embryos the gap gene expression levels stop to grow in the second half of cycle 14A and slightly decrease by gastrulation [10] . Mutations in gap genes aggravate this effect , that underlines the importance of intact network for maintenance of normal gap gene expression levels .
We obtained embryos from loss-of-function allele ( FlyBase ID FBal0005790 ) [17] . embryos were collected either from Df ( 3L ) ri-79c or Df ( 3L ) ri-XT1 , ru[1] st[1] e[1] ca[1] stocks . Kr;kni double mutant embryos were made by crossing and Df ( 3L ) ri-79c flies . 3–4 hr old embryos from flies carrying mutation [17] were collected , fixed and stained as described elsewhere [68] , [69] . We used primary antibodies against Kr , Knirps ( Kni ) , Giant ( Gt ) , Hunchback ( Hb ) and Even-skipped ( Eve ) [68] , [70] and secondary antibodies conjugated to Alexa Fluor 488 , 555 , 647 , and 700 ( Invitrogen ) . Each embryo was additionally stained with either anti-histone H1-4 antibody ( Chemicon ) or Hoechst 34580 ( Invitrogen ) to mark the nuclei . Laterally oriented Kr null embryos , showing zero level of Kr expression and severely transformed Eve pattern [15] , were scanned using Leica TCS SP2 and Leica TCS SP5 confocal microscopes as described [69] . For each experiment , the microscope gain and offset were set on maximum expression level of a given gene in wild type patterns and then these settings were applied for mutants . The 8-bit digital images of gene expression in Kr mutants were acquired for cleavage cycle 14A . For spatial registration and data integration , embryos from cleavage cycle 14A were distributed into 8 time classes about 6 . 5 min each on the basis of measurement of degree of membrane invagination , as well as characteristic features of the even-skipped gene expression pattern [53] . The quantitative gene expression data and integrated patterns for each temporal class of cycle 14A were obtained as previously described [33] , [69] , [71] , [72] using recently developed packages ProStack and BREReA [73] , [74] . The one-dimensional integrated patterns of gene expression in wild type were taken from FlyEx database ( http://urchin . spbcas . ru/flyex/ , [75] ) . Gene circuit models [12] , [13] , [31] , [76] , [77] describe the dynamics of segmentation gene expression in the syncytial blastoderm of Drosophila melanogaster . The circuits used in this paper consider the time evolution of protein concentrations of gap genes hb , Kr , gt , and kni in two genotypes: wild type and in embryos with homozygous null mutation in Kr gene . To make separate treatment of domains with different regulatory inputs possible we narrowed down the spatial domain of the model by considering only the posterior half of the blastoderm ( region from 47 to 92% embryo length ( EL ) ) , in which each of the these genes is expressed in one domain . We consider a one-dimensional row of nuclei along the anteroposterior axis of the embryo , as anteroposterior ( A-P ) and dorsoventral ( D-V ) patterning systems are largely independent of each other in the presumptive germ band of the blastoderm embryo . The modeled region extends over 45% of the A-P axis , from the minimum of gt expression inbetween third and fourth gt stripes to the posterior border of the posterior hb domain ( Figure 1 ) . Gene circuits function according to three rules: interphase , mitosis and division [29] . During mitosis , only protein transport and protein decay govern the dynamics as transcription shuts down and nascent transcripts are destroyed [78] . Mitotic division is modeled as a discrete change in the state of the system . At the end of a mitosis , each nucleus is replaced with its daughter nuclei , the inter-nuclear distance is halved and the daughter nuclei inherit the state of the mother nucleus . During interphase the change in concentration for each gap gene product a in each nucleus i over time t is described by the following system of ordinary differential equations ( ODEs ) ( 1 ) The three terms on the right-hand side of the equation represent protein synthesis , protein diffusion and protein decay . is the total regulatory input to gene a . is the number of gap genes in the model ( hb , Kr , kni and gt ) , is the number of external regulatory inputs ( bcd , cad and tll genes , which are not regulated by gap genes , but regulate these genes ) . and are genetic inter-connectivity matrices that characterize the action of regulator b or external input e on gene a . The sizes of these matrices are and correspondingly . is a threshold parameter of the sigmoid regulation-expression function . is the maximum synthesis rate , the diffusion coefficient , and the decay rate of the product of gene a . The gap gene circuits used in this study consider events occurring during cleavage cycles 13 and 14A and ending at the onset of gastrulation [3] . The divisions are carried out according to a division schedule based on experimental data . Time t is measured in minutes from the start of cleavage cycle 13 . The interphase of cycle 13 lasts for 16 . 0 min , and its mitosis from 16 . 0 to 21 . 1 min . At min , the thirteenth division is carried out by applying the division rule . The interphase of cycle 14A starts immediately after division , and lasts until gastrulation at min . The cleavage cycle 14A is subdivided into 8 temporary equivalent classes , as a result the model is compared to data at 9 time points , one time point for cycle 13 ( C13 ) , and eight points for cycle 14A ( time classes T1–T8 ) . Kr , gt , and kni are exclusively zygotic , and are not present at significant levels before cycle 13 [10] , thus they have initial conditions of zero . For hb , the expression data from cycle 12 is used as the initial condition . hb , which is expressed both maternally and zygotically , shows a large increase in expression in cycle 13 [38] , [39] , indicating commencement of its zygotic expression . Non-zero initial conditions for external inputs Bcd , Cad and Tll are obtained by piecewise linear interpolation of integrated expression data at midpoint of C12 ( t = −6 . 2 min ) and midpoint of C13 ( t = 10 . 55 min ) . Moreover , in order to solve the right hand side of equation ( 1 ) , the concentrations of external inputs must be supplied for any time in the duration of the model . This is implemented by linear interpolation between data points at C13 and eight time classes of cycle 14A ( T1–T8 ) , with data points corresponding to midpoints of C13 and each time class . As was suggested in previous studies [79] the zero flux boundary conditions were chosen at both ends of modeling interval because other numerically feasible alternatives , such as periodic boundary conditions , are not biological . The actual flux through the boundaries is nonzero but depends on gene expression in a complicated manner that may add the unneeded overhead for numerical simulations . The model parameters are estimated by fitting the model output to experimental data . This is performed by minimization of cost function based on the sum of squared differences between gap protein levels in the model and data . DEEP - Differential Evolution Entirely Parallel method is applied to biological data fitting problem . We introduce a new migration scheme , in which the best member of a branch substitutes the oldest member of the next branch , that provides a high speed of the algorithm convergence [46] . For the comprehensive analysis of modeling results it is necessary to know how reliable the parameter estimates are . In practice insufficient or noisy data , as well as the strong parameter correlation or even their functional relation may prevent the unambiguous determination of parameter values . Such parameters are related to as non-identifiable . To reveal non-identifiable parameters the method based on confidence intervals [49] , [82] is applied . The confidence intervals are constructed for the parameter estimates in the vicinity of model solution and are given by ( 2 ) where is the sensitivity matrix , the matrix of partial derivatives of the model solution with respect to the parameter vector; is the objective functional; is an -quantile of -distribution with m and N-m degrees of freedom . The size of confidence intervals characterize the sensitivity of the solution to parameter changes: the shorter is the confidence interval the more reliable is the parameter estimate . If the most important feature of the parameter estimate is its sign , the identifiable estimate must have the confidence interval bounded away from zero . It is important to mention that the confidence intervals are only estimated precisely in case of independent parameters , if some parameters are strongly correlated the confidence intervals are overestimated . In other words the confidence interval ( 2 ) is the whole area of the parameter variation as the other parameters take any possible values from the m-dimensional confidence area ( see equation ( 1 ) and Figure 1 in Text S2 ) . Besides , correlation of parameters causes calculational errors due to ill-conditionality of the sensitivity matrix . This issue is discussed in more detail in Text S2 . The other method to detect interrelations between parameters is the collinearity analysis presented in [83] . The method is suitable for models with large number of parameters . The aim of the method is to reveal the so-called near collinear columns of the sensitivity matrix , the matrix of partial derivatives of the model solution with respect to the parameter vector , and thus detect subsets of non-identifiable parameters . Identifiability of a parameter subset is characterized by collinearity index defined as ( 3 ) whereis the minimal eigenvalue of the submatrix of the Fisher information matrix . High values ofindicate that the subset of parameters is poorly identifiable due to relations between at least two parameters . The aim of the analysis is to detect all the parameter subsets of any dimension with high collinearity index such that they do not contain subsets of lower dimension for which the collinearity index is also high . Thus we reveal all the non-identifiable parameters . For more detailed description of the methods see Text S2 . | Systems biology is aimed to develop an understanding of biological function or process as a system of interacting components . Here we apply the systems-level approach to understand how the blueprints for segments in the fruit fly Drosophila embryo arise . We obtain gene expression data and use the gene circuits method which allow us to reconstruct the segment determination process in the computer . To understand the system we need not only to describe it in detail , but also to comprehend what happens when certain stimuli or disruptions occur . Previous attempts to model segmentation gene expression patterns in a mutant for a trunk gap gene were unsuccessful . Here we describe the extension of the model that allows us to solve this problem in the context of Kruppel ( Kr ) gene . We show that remarkable alteration of gap gene expression patterns in Kr mutants can be explained by dynamic decrease of the activating effect of Cad on a target gene and exclusion of Kr from the complex network of gap gene interactions , that makes it possible for other interactions , in particular between hb and gt , to come into effect . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"systems",
"biology",
"developmental",
"biology",
"gene",
"regulation",
"gene",
"expression",
"regulatory",
"networks",
"molecular",
"genetics",
"biology",
"computational",
"biology",
"genetics",
"gene",
"networks",
"pattern",
"formation",
"genetics",
"and",
"genomics"
] | 2012 | Modeling of Gap Gene Expression in Drosophila Kruppel Mutants |
Every year , influenza epidemics affect millions of people and place a strong burden on health care services . A timely knowledge of the onset of the epidemic could allow these services to prepare for the peak . We present a method that can reliably identify and signal the influenza outbreak . By combining official Influenza-Like Illness ( ILI ) incidence rates , searches for ILI-related terms on Google , and an on-call triage phone service , Saúde 24 , we were able to identify the beginning of the flu season in 8 European countries , anticipating current official alerts by several weeks . This work shows that it is possible to detect and consistently anticipate the onset of the flu season , in real-time , regardless of the amplitude of the epidemic , with obvious advantages for health care authorities . We also show that the method is not limited to one country , specific region or language , and that it provides a simple and reliable signal that can be used in early detection of other seasonal diseases .
Seasonal influenza is a worldwide infectious disease estimated to be the cause of 3 to 5 million cases of severe illness and up to half a million deaths every year [1] , also placing a strong economic burden on health services [2] and [3] . To deal with these epidemics , the beginning of the flu season has to be declared . Following the official alerts , hospital emergency rooms and health care centres activate appropriate flu response protocols and prepare for possible overcrowding . However , and despite occurring yearly , the onset of the influenza outbreaks is unpredictable and this uncertainty poses logistic problems to most public health services , often already under high demand due to excess winter mortality . Therefore , reliable and timely information tools on current influenza activity are of the utmost interests to health services and to health-related decision makers . In Europe , the European Influenza Surveillance Network ( EISN ) , implemented and coordinated by the European Centre for Disease Control ( ECDC ) , is the leading responsible entity for gathering and reporting data on influenza activity , during each season . This surveillance mechanism relies on a network of sentinel medical doctors , spread throughout all European Union ( EU ) and European Economic Area ( EEA ) Member States . These sentinel doctors report on the number of patients with influenza-like illness ( ILI ) who self-referred to primary health care services , from October to May of each year , and also send samples for laboratory testing . With this information , the ECDC generates a weekly report , referring to the previous week , which includes the estimated number of ILI cases per 100 , 000 inhabitants , and other indicators such as trend , types and subtypes of circulating influenza viruses , or geographical spread [4] , [5] . The EISN-ILI method is arguably one of the best surveillance systems in the world and a systematic source of reliable data . However , it faces several challenges . First , only an unknown sized sample of those with ILI seek medical care , and this sample can change depending on the circulating virus subtype , from season to season and from country to country; second , the number of medical professionals participating in the sentinel network is small ( 1–5% of physicians working in the country or region [5] ) , which can result in low statistical significance and unpredictability; third , even if consultations happened with no delay , the data would be available at best with one week lag . Thus , this system can lead to under-reporting , especially early in the season , when both medical doctors and the general population haven’t been alerted to the presence of a circulating Influenza-Like Virus . This means that between the actual onset of the seasonal epidemics and the official alert , several weeks can elapse . These limitations have been recognized by others and past studies have focused on forecasting the ILI incidence rate independently of clinical consultations , by using data from on-line volunteer participants [6] , ILI-related queries on Google [7] , Wikipedia logs [8] or a combination of several data sources [9] . All these systems were designed to give the best , real-time , ILI rate estimates . However , and irrespective of the data source used , these studies often focus on the USA or in one single country , and might be difficult to generalize to different regions . Moreover , influenza dynamics research [10] is often mainly concerned with simulating the flu season’s number of cases , as well as the peak’s timing , neglecting timely onset identification . We argue that , from the health policy stand point , it is fundamental to be able not only to track changes in incidence rate , but also to accurately know when the flu season has started , as a major concern with the outbreak of influenza is the immediate over-burden of health resources . An early detection of the flu onset could a ) anticipate the provision or reinforcement of health professionals and facilities; b ) confidently advise the generally healthy population to stay home , redirecting them from the likely to-be crowded emergency services [11]; and c ) signal the entire EINS network , possibly even improving the surveillance system . In fact , the cited 2014 review [10] listed 24 papers that focus of seasonal influenza forecasting or that could be applied to seasonal data . Several identify onset prediction as an important goal [12][13] but only one [14] tries to do on onset prediction , although not in real time . This has been the case of few other studies , that develop methods to identify the onset , or show that different systems and data sources could be used to it , either with real data or just by testing different models , such as [15][16][17][18][19] , and particularly [20] , which focuses on the potential of GFT . But , to our knowledge , no other work has focused on developing and testing a real-time onset detection system . Thus , in this study we present a different approach and describe a method to identify the onset of the seasonal influenza epidemics , using alternative and real-time data-sources . In this context , we also present a new source of data , highly correlated with the ILI rate . Saúde 24 is a Portuguese national triage call centre service , established to give free and real-time telephone health advice [21] . From the symptoms collected in each phone-call , we can not only have an estimate of the Portuguese ILI rate , but also use our method to signal the flu season outbreak . First we identify the onset of the flu season by fitting the EISN-ILI data using a modified version of the classic SIR model ( MSIR ) , with a dynamic transmission rate ( A and B in Fig 1 ) . This marks the beginning of the flu in past seasons and in several countries . Second , we use alternative data sources , that do not have the described limitations of the EISN-ILI data , to identify the onset , using the MSRI fitted EISN-ILI as our ground truth ( C in Fig 1 ) . Our model selects the combination of features that minimizes the difference between the onset identified using the EISN-ILI fit ( orange lines ) and the one obtained with the different data sources ( blue line ) . Third , we test this model in real-time and compare our predictions to the target . Finally , we compare our real-time identified onsets to the official flu season alerts , as published by the different countries analysed . By using these different data sources , and optimizing each source’s strengths , we can produce an accurate signal that identifies , in real-time , the onset of the flu season and that anticipates the official alerts by several weeks . We show that the model performance depends not only on the quality of the input data , but also on its diversity . We also show that the model is not region-specific and that , depending on the quality of the data , can be applied to different countries . With such a reliable method , complementary to the current system , public health authorities could significantly anticipate their respective protocols and timely respond to the upcoming flu peak .
We collected influenza related data from three different and independent sources . These are the 1 ) EISN-ILI incidence rate per 100 , 000 inhabitants , considered the ground truth in this study , 2 ) Google Trends for four influenza related search-terms and 3 ) Saúde 24 phone calls logs , only available in Portugal . When possible , we collected this data in all countries under consideration and for five consecutive influenza seasons: 2010/2011 , 2011/2012 , 2012/2013 , 2013/2014 and 2014/2015 . The main goal of our approach is to build a mechanism able to timely identify the flu onset . Our method relies on a function that signals , in real-time , the likelihood that the season has started . This signal function receives as input influenza related data , and outputs a normalized sigmoid-like activation function that informs about the likelihood of the onset . To build such a signal we devised a 3-step method ( Fig 1 ) . First , and in order to construct the training data sets , we identified the week that marks the beginning of each season , or onset . This is done by fitting a Susceptible-Infected-Recovered-like compartmental model to the EISN-ILI data of all influenza seasons under consideration . Second , we introduce a signal function , the Identified Onset ( IO ) , centred at the previously found onset week . This is used as the ground truth or target function , to which all other simulations will be compared . Third , by using alternative data sources , a training process is explored to fit a Predicted Onset ( PO ) signal function by repeating and testing over all seasons . Finally , we compare both the Identified and the Predicted Onsets to the official alert periods , described in the Data section . The following sections describe each step in detail . We have used Mathematica 10 . 1 [29] to perform all calculations
The influenza epidemic varies in timing and amplitude from season to season and from country to country , making it difficult to predict and identify its onset . Fig 4 reveals this difficulty by showing , for two geographically distinct countries , the unscaled ILI incidence , as obtained from the EISN , and comparing five seasons with the official alert dates . S11 Fig shows the same data for the other countries analysed . To help identify the onset , we applied the modified SIR model ( MSIR ) to all 23 countries available from EISN . From these , 11 countries resulted in a good fit ( all 5 seasons fits must result in a SIR shape distribution and AR2 > 0 . 9 ) , and for 8 countries we had at least two independent data sources . Fig 2 shows the countries analysed and the time series used as inputs ( columns 3 to 8 ) . These countries have different climates , cover a large geographical area and have significant cultural and social differences , from language to school year . Fig 5 panels A , C , E , G , I , K , M , and O show the best Modified-SIR model ( MSIR ) fitting results for seasons 2010–2015 , for each of the analysed 8 countries ( we show in S12 Fig the equivalent results for Iceland , Luxembourg and Latvia ) . The corresponding best fit parameters are summarized in S4 Table . Each fit is accompanied by the respective transmission rate , β , obtained from Eq 1 ( orange line in Fig 5 panels B , D , F , H , J , L , N , and P ) . The dashed vertical lines , connecting both panels , show the transmission rate sigmoidal inflection point , which marks the epidemic onset . We show also in S13 Fig the fitting result for all initial analysed 23 countries . As can be seen , there are large differences in amplitude and smoothness between ( and within ) seasons , but as long as our MSIR fits the EISN-ILI time series , we can identify the beginning of the season as the inflection point of the sigmoid ( dashed line ) in all 8 countries . Given it’s sigmoidal shape , the transmission rate is a strong candidate indicator of the epidemic onset: it results in an approximated step function that triggers at the outbreak . Thus , we can define “signal functions” , with a similar shape , and collect alternative sources of data , that do not require post-processing , such as the one provided by Google Trends ( GT ) . By using a combination of time-series and by training our functions , we can identify the onset in real-time . The figure also shows the week difference between our identified onset and the MSIR fit maximum value , which can be interpreted as the time difference between the beginning of the season and its peak . This week difference is consistent in several countries ( Hungary and Spain being the best examples ) , with an overall average ( and median ) distance of 8 weeks . S1 to S8 Figs show each GT query time series , for the respective country , and the corresponding EISN-ILI rate , between June/2010 and June/2015 . As mentioned before , GT does not release weekly data for every search term , in every country . Fig 2 shows the countries for which only monthly data was available and notes whether they were removed from the study . Similar to previous work in other countries [7] , we find a good peak correspondence between the EISN-ILI incidence and searches for influenza related terms , particularly for the words “flu” and “cough” , in the different countries and languages analysed . However , there have been some concerns regarding the use of GT to predict and track the flu season [32][33] . These are mainly focused on its sensitivity to media reporting , which can lead to an artificial increase in searches , and the covered demographics , which in some countries is heavily biased towards a young and educated population . To overcome these issues , we also took advantage of the Saúde24 ( S24 ) phone service , which covers a broader demographic , including elders , also offering detailed information about the callers . These calls are in real-time and the call logs provide both structured and unstructured information about the callers’ symptoms as gathered by highly trained nurse practitioners . Moreover , S24 became well known in Portugal during the 2009 H1N1 pandemic and it is still broadly used by people with ILI symptoms . S9 Fig shows the total number of phone calls received by S24 services between June/2010 and February/2015 , and the reported EISN-ILI incidence rate for Portugal in the same time period . Similarly to GT , we find a good correspondence between the number of calls that S24 received and the EISN-ILI incidence . We then combined these different input time series , or features , to determine which combination offered the best onset identification ( see Methods for more details ) . These are the input combinations that minimize the difference between our Identified Onset ( IO ) and our Predicted Onset ( PO ) , and are shown in Fig 3 . S5 Table shows , for these best selected features , the resulted weights set {a , bk} . In half of the countries , using ILI provided best fits than using GT ( CZ , ES , IE and NO ) . The reverse was true in Belgium and Portugal , with no noticeable differences in the cases of Hungary and Italy . Using a combination of features , either from ILI , from different GT term-searches , or S24 , proved to be the best option for Italy and Portugal . This consistency between the IO and the PO was particularly good in the cases of Italy ( RSM = 0 . 06 ) and Spain ( RMS = 0 . 11 ) , which were also the countries that had presented the best AR2 in the MSIR fit ( Fig 5 ) . This is not a coincidence , as well-shaped data clearly offers the best prediction results . In the case of Portugal , that did not have a particularly good AR2 , adding the features from S24 alone , improved the signal by 2 to 3-fold when compared with the other features’ results and by 2-fold when compared with the average best RSM of all the other countries that , to our knowledge , do not have an S24 equivalent . The previous results show that it is possible to select different features and to use them to identify the onset of the flu season , in different countries . We then trained our model to simulate a real-time signal prediction , using the previously selected features . Fig 6 shows the signal prediction simulation for 2010–2015 seasons , and for the different countries . It is possible to see the predicted onset ( PO ) function , in blue , and the identified onset ( IO ) function , in orange , for all five seasons and for all 8 countries . The onsets are chosen as being the curve’s inflection points , at 0 . 5 . The prediction simulations were very good for Spain , Italy and Portugal , followed by Hungary , which presented a sigmoid shape and a stable signal . Again , the best simulation cases agreed with the best MSIR fitting scenarios ( Spain and Italy resulted in a Adjusted R2 of 0 . 99 and 0 . 97 respectively ) or , in the case of Portugal , when the S24 data was included . It should also be noted that , except in a few cases discussed below , all calculated feature weights are within the range of the standard error , for the respective season and country ( S5 Table ) . This is a good indication that our results are not over-fitted . The same figure also shows the week difference between the orange IO and the blue PO trigger week position . A plus signal means that the prediction failed by lateness , and the minus sign that it failed by anticipation . We found a close to perfect match ( 0 and plus or minus one week ) in approximately half of the seasons analysed . The model misses the onset by 4 or more weeks in only 8 out of the 40 season/country combinations , most notably in the cases of Czech Republic , Ireland and Norway , with 4 of these instances being a signal anticipation , or false positive . In fact , one of the few exceptions to the quality of the feature weighs is found in season 10/11 , in Hungary . In this case , the resulted fitted a constant ( S5 Table ) is very far from the other comparable four results , most likely a consequence of the fitting optimization process . Thus , the result is a poor generalization for the season prediction simulation and , in this particular case , it resulted in a 6 week false positive . However , it is worthwhile noticing that this might not be a real failure of the model . S4A and S11D Figs , show that there is a systematic ( and most likely artificial ) decrease in the registered number of cases every season on , and around , week 50 , in Hungary . Our method identified week 48 as the beginning of the season 10/11 and it is possible that this was actually the case . S11 Fig also shows that often the official alerts happen at or even past the peak . To compare our method to the timings of the current official alert system , we plotted the same IO and PO curves and added the official season alerts , as shown in Fig 7 and separately in S14 and S15 Figs , respectively . The IO matches or anticipates the official alert in all of the cases studied and anticipates the alert signal by at least 2 weeks in 90% of the cases ( S14 Fig ) . In the case of Spain , these were very consistent , and the IO anticipated the alert by exactly three weeks in all seasons analysed . Similarly , the PO calculated by our real-time prediction , anticipates or closely matches ( at most one week difference ) the official alert in all but three of the seasons ( BE 10/11 , BE 12/13 and NO 12/13 ) , and anticipates the alert by at least 2 weeks in 70% of the cases . These differences are particularly large in Czech Republic or Ireland , but consistent in countries such as Spain , Italy and Portugal , where we observe systematic predictions . This result would be even more striking if we removed Belgium from the analysis . In the case of Belgium , the correspondence between the PO and the Alert is very good , suggesting that the official alerts in Belgium happen with a very short delay . This clearly indicates that the official alerts are systematically delayed by at least a few weeks when compared to the actual beginning of the epidemic and that our method can improve the current system by more than 3 weeks , in most cases .
The proposed model is not designed to detect the peak: there is no focus on the amplitude of the curve as it depends on external factors that were not considered . This can also explain why the search engine Google has shown to be an accurate tool . The model is also not designed to detect off-season events , such as the 2009 pandemic , as it requires defining a baseline from which the onset deviates . Finally , the fact that the method works differently in different countries can be described as both an asset and a limitation . Our model is flexible enough to be generalized to many countries and realities , contrary to the majority of the previous work that has been limited to the USA and other specific countries . However , it requires fine-tuning to each countries’ data sources and requires that the seasonal epidemics follow a well-behaved SIR-like curve . Overall , the system that we have developed , due to its accuracy and simplicity , by providing one single , easy to interpret , output , can be very useful for public health authorities , in tracking and identifying the beginning of the flu season . In fact , our system is currently being tested in real-time , together with the relevant Portuguese health organizations , and should be easy to implement in other countries . Moreover , and also due to both its simplicity and to the fact that it can be used with different input data , this method should be easy to apply to other SIR-like contagious or seasonal diseases . | Influenza , generally referred to as the flu , is a common infectious disease that affects millions of people . Every year , we expect this seasonal disease to occur during the Winter , but exactly when it will start and how severe it will be is not known . This places a strong burden on health services , as often the spread can be felt as very fast and emergency rooms become flooded with patients . With this work , we propose a new method that identifies the beginning of the yearly flu season . This is done by using several different data sources , including searches for flu-related symptoms on Google and phone call logs to a specialized medical phone service . These data sources , together with our method , can provide a daily or weekly report , making it much faster than current methods , which require lab testing or centralized medical reports . Our method was applied to different European countries and can anticipate current official alerts by several weeks . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"influenza",
"applied",
"mathematics",
"geographical",
"locations",
"epidemiological",
"methods",
"and",
"statistics",
"simulation",
"and",
"modeling",
"algorithms",
"seasons",
"mathematics",
"infectious",
"disease",
"control",
"research",
"and",
"analysis",
"methods",
"curve",
"fitting",
"infectious",
"diseases",
"mathematical",
"functions",
"epidemiology",
"mathematical",
"and",
"statistical",
"techniques",
"critical",
"care",
"and",
"emergency",
"medicine",
"portugal",
"people",
"and",
"places",
"infectious",
"disease",
"surveillance",
"earth",
"sciences",
"disease",
"surveillance",
"viral",
"diseases",
"physical",
"sciences",
"europe"
] | 2017 | Early and Real-Time Detection of Seasonal Influenza Onset |
Although hypoxia is a major stress on physiological processes , several human populations have survived for millennia at high altitudes , suggesting that they have adapted to hypoxic conditions . This hypothesis was recently corroborated by studies of Tibetan highlanders , which showed that polymorphisms in candidate genes show signatures of natural selection as well as well-replicated association signals for variation in hemoglobin levels . We extended genomic analysis to two Ethiopian ethnic groups: Amhara and Oromo . For each ethnic group , we sampled low and high altitude residents , thus allowing genetic and phenotypic comparisons across altitudes and across ethnic groups . Genome-wide SNP genotype data were collected in these samples by using Illumina arrays . We find that variants associated with hemoglobin variation among Tibetans or other variants at the same loci do not influence the trait in Ethiopians . However , in the Amhara , SNP rs10803083 is associated with hemoglobin levels at genome-wide levels of significance . No significant genotype association was observed for oxygen saturation levels in either ethnic group . Approaches based on allele frequency divergence did not detect outliers in candidate hypoxia genes , but the most differentiated variants between high- and lowlanders have a clear role in pathogen defense . Interestingly , a significant excess of allele frequency divergence was consistently detected for genes involved in cell cycle control and DNA damage and repair , thus pointing to new pathways for high altitude adaptations . Finally , a comparison of CpG methylation levels between high- and lowlanders found several significant signals at individual genes in the Oromo .
Hypoxia is a major stress on human physiological processes and a powerful homeostasis system has evolved in animals to cope with fluctuations in oxygen concentration [1] . High altitude ( HA ) hypoxia , such as that experienced at 2500 m of altitude or greater , engages this system and elicits physiological acclimatization when lowlanders become exposed to hypoxia . In addition to lower oxygen levels , lower biodiversity and extreme day-to-night temperature oscillations challenge HA living . The classic response is an increase in hemoglobin ( Hb ) concentration that during acclimatization compensates for the unavoidable lowered percent of oxygen saturation ( O2 sat ) of Hb due to ambient hypoxia . Acclimatization is not completely effective , however . For example , birth weights are lower than at low altitude ( LA ) as is physical exercise performance [2] , [3] . In addition , lowlanders residing at altitudes higher than 2500 meters ( m ) are at risk for chronic health problems arising in part from acclimatization processes . For example , long-term high Hb levels increase blood viscosity as well as the risk of thrombosis and stroke [4] , [5] , [6] and poorer pregnancy outcomes [7] . These results taken together suggest that the acclimatization response does not assure that fitness is unaltered at HA . Several distantly related human populations have survived for 5–50 thousand years ( ky ) [8] , [9] , [10] at altitudes above 2500 m . Indeed , in most cases , sufficient time has elapsed since HA settlement for natural selection to have changed the frequency of adaptive alleles . Interestingly , even though all HA residents are exposed to the same , constant , ambient hypoxia , indigenous highlander populations show distinctive physiological characteristics thought to offset HA stress: Andeans show some reduction in O2 sat , but a marked increase in Hb levels [11] , Tibetans present markedly low O2 sat , but relatively little increase in Hb levels [12] , and Amhara in Ethiopia present little reduction in O2 sat or increase in Hb levels [13] , [14] . Whether these phenotypic contrasts reflect different genetic adaptations across populations remains an open question . Substantial evidence in Tibetan highlanders suggests that variation in Hb levels and O2 sat is adaptive . In the Tibetan population , a major gene effect on O2 sat was detected , with the inferred genotypes associated with higher O2 sat also associated with higher reproductive success [15]; though the locus underlying this effect has not yet been identified , its effect on phenotypes directly related to fitness points to the presence of adaptive variation . With regard to Hb levels , which are surprisingly similar between Tibetan highlanders and lowlanders at sea level [11] , population genetics and genotype-phenotype association analyses have identified alleles at two loci ( endothelial PAS domain protein 1 ( EPAS1 ) and egl nine homolog 1 ( EGLN1 ) ) that are consistently associated with signatures of positive natural selection and with lower Hb levels , suggesting that natural selection in Tibet favored variants that counteract the deleterious effects of long-term acclimatization [16] , [17] , [18] , [19] . An analysis of genome-wide genotype data in Tibetan and Andean highlanders suggested that natural selection acted on largely distinct loci in the two populations [20] . In addition , a recent study comparing Ethiopian Amhara highlanders with other ethnic groups at LA identified yet another set of candidate targets of selection [21] . However , the Tibetans remain unique with regard to the strength of the evidence for natural selection and the marked genetic effects on the Hb level phenotype . HA populations offer a rare opportunity to investigate the impact of natural selection on the genetic architecture of adaptation because independent realizations of the adaptive process can be examined in different parts of the world . Specifically , the Ethiopian highlands offer a unique opportunity to study HA adaptation because individuals from distinct , but closely related ethnic groups have communities at HA and LA , thus allowing more informative genetic and phenotypic comparisons . In this study , we extended genomic analysis to two Ethiopian ethnic groups , Amhara and Oromo , with the goal of determining whether Tibetans and Ethiopian highlanders share the same adaptations and of elucidating the genetic bases of adaptive HA phenotypes in Ethiopia . We also measured genome-wide methylation levels to explore the contribution of epigenetic modifications to HA adaptations .
We obtained phenotype data in two distinct , but closely related ethnic groups , the Amhara and the Oromo ( Texts S1 , S2; Figures S1 , S2 , S3 , S4 , S5 ) , that include communities of HA and LA residents . All individuals were born and raised at the same altitude where they were sampled . These samples allow comparing phenotypes across altitudes within ethnic groups as well as across ethnic groups . While historical records indicate that the Oromo have moved to HA only in the early 1500 s [22] , [23] , the Amhara have inhabited altitudes above 2500 m for at least 5 ky and altitudes around 2300–2400 m for more than 70 ky [24] , [25] . Therefore , sufficient time has elapsed for the Amhara to have evolved genetic adaptations to hypoxia . As shown in Figure 1 , the HA samples of both ethnic groups had higher Hb than the LA samples , however the Oromo had twice as much elevation in Hb as the Amhara . The elevation in Hb levels is particularly evident for the measurements in males , raising the possibility that other factors ( e . g . menstrual cycle ) in females affect the power to detect significant phenotypic differences between groups . With regard to O2 sat , HA Amhara had a 5 . 6% lower O2 sat compared to LA Amhara while HA Oromo had 10 . 5% lower O2 sat than their LA counterparts . Therefore , we detected significant phenotypic differences not only between populations from the same ethnic group that live at different altitudes , but also across populations from closely related ethnic groups ( Oromo and Amhara ) that live at the same altitude . Given the low genetic divergence between these two ethnic groups at the genome-wide level ( mean FST = 0 . 0098 ) , the phenotypic differences between Amhara and Oromo highlanders are unlikely to be due to independent genetic adaptations in these ethnic groups; rather they are likely to reflect genetic adaptations that evolved in the Amhara , due to their longer residence at HA . We also measured pulse and calculated arterial oxygen content , but these phenotypes did not show significant differences across ethnic groups or altitudes and were omitted from further analyses . For details on the phenotypic variation in Amhara and Oromo , see Text S3 , Tables S1 and S2 and Figure S6 . To learn about the genetic bases of variation in Hb level and O2 sat in Ethiopia , we tested for an association between SNP genotype in the 260 unrelated Ethiopian samples and Hb levels or O2 sat . We considered the total Ethiopian sample ( i . e . HA and LA Amhara and Oromo ) as well as each ethnic group and each altitude separately ( Figures S7 , S8 , S9 , S10 , S11 , S12 , S13 , S14 , S15 ) . No genome-wide significant signal was observed for either Hb levels or O2 sat in the Oromo and in the total Ethiopian sample and for O2 sat in the Amhara ( Figures S7 , S8 , S9 , S10 , S11 , S12 , S13 , S14 , S15 and Tables S3 , S4 , S5 , S6 , S7 , S8 , S9 , S10 , S11 , S12 , S13 , S14 , S15 , S16 , S17 , S18 , S19 , S20 ) . Likewise , no excess of low p-values was observed in these association analyses relative to null expectations obtained by permutations ( Figures S7 , S8 , S9 , S10 , S11 , S12 , S13 , S14 , S15 ) . In contrast , the Amhara showed an excess of low association p-values in the analysis of Hb levels compared to expectations obtained by permutations ( Figure 2A ) , indicating a genetic contribution to variation in Hb levels . In addition , one SNP ( rs10803083 ) on chromosome 1 was associated with variation in Hb levels ( p = 4 . 96×10−8 ) in Amhara; this association is genome-wide significant after correction for the 985 , 385 tests performed ( Figure 2B and Table S3 ) . In addition , the next six most strongly associated SNPs are in strong LD with rs10803083 ( r2≥0 . 69 ) . There are no known genes within 600 kb of these SNPs , and the closest genes , i . e . phospholipase D family member 5 ( PLD5 ) and centrosomal protein 170 kDa ( CEP170 ) , are not obvious candidate genes for variation in Hb levels . These SNPs do not reside within an ultra conserved sequence element . Notably , the effect size of SNP rs10803083 ( ∼0 . 83 g/dL , Figure 2C ) is half that of the EGLN1 SNPs [16] but comparable to that of Hb associated EPAS1 SNPs in Tibetans [17] . Although SNP rs10803083 reaches genome-wide significance levels , replication studies will be needed to further assess the evidence for an association with Hb levels . Our sample size is small relative to traditional GWAS , thus liming the power of our analysis . In addition , due to the correlation between linked SNPs , the Bonferroni correction we applied is overly conservative . Therefore , many true Hb concentration variants may not reach genome-wide levels of significance . In this regard , it is interesting to note that the second strongest association signal ( rs2899662 ) is located within the established hypoxia-candidate gene retinoid-related orphan receptor alpha ( RORA ) . RORA encodes a protein that induces the transcriptional activation of hypoxia-inducible-factor-1alpha ( HIF-1α ) [26] , thus it plays a significant role in the same pathway where adaptations were detected in Tibetans . This SNP is , therefore , an excellent candidate for variation in Hb levels . Additional SNPs of interest for follow up analyses ( Table S3 ) include: the solute carrier family 30 member 9 ( SLC30A9 ) , which is regulated by hypoxia [27] , the collagen type VI alpha 1 ( COL6A1 ) , which is associated with performance during endurance cycling [28] and is a HIF response gene [29] , and the hepatocyte growth factor ( HGF ) , which is induced by hypoxia [30] , activates HIF1 DNA binding [31] , plays a role in angiogenesis and protects against hypoxia induced cell injury [32] , [33] . The above association analyses allow comparing the genetic architecture of Hb levels between Ethiopians and Tibetans [16] , [17] , [18] . First , we focused on the EPAS1 and EGLN1 SNPs that were previously associated with variation in Hb levels in Tibetans , with effect sizes of 0 . 8 g/dL and 1 . 7 g/dL , respectively [16] , [17] . None of these SNPs were significantly associated with Hb levels in Ethiopians ( Table 1 ) . Because we have complete or nearly complete power to detect a genotype-phenotype association in our Ethiopian samples ( see Text S4 and Table S21 ) , we infer that the SNPs associated with variation in Hb levels in the Tibetans do not make a contribution in Ethiopians . Second , because the association signal in the Tibetans may be due to an untyped variant that is tagged by different SNPs in Tibetans and Ethiopians , we also considered all SNPs within 10 kb of the EPAS1 and the EGLN1 genes and repeated this analysis applying a Bonferroni correction for the number of tests performed . None of the EPAS1 or EGLN1 SNPs was significantly associated with Hb levels in Ethiopians . Based on power analyses ( Figure 3 ) , we can exclude associated variants with the same effect size as in Tibetans if the MAF in Ethiopia is greater than 10% and 5% , respectively , for the EPAS1 and EGLN1 genes ( see Figure S16 for Amhara and Oromo ) . Therefore , this more comprehensive analysis suggests that genes shown to contribute to variation in Hb levels in Tibetans either do not influence variation in the Ethiopian populations or if they do , their effect sizes are lower than those reported for the Tibetans . Third , we considered all SNPs within 10 kb of the candidate genes in the “Response to Hypoxia” Gene Ontology ( GO ) category ( 26 genes ) . None of these SNPs is significantly associated with Hb levels after multiple test correction ( p<0 . 05/1309 = 3 . 81×10−5 ) . Because of the larger number of SNPs tested , this analysis has a relatively high multiple testing burden . Nonetheless , we find that we have greater than 80% power to detect a SNP significantly associated with Hb levels and effect size 0 . 8 g/dL if its MAF is at least 20% and 100% power if its effect size is 1 . 7 g/dL Hb ( Figure 3C and Figure S16 for Oromo and Amhara ) . Therefore , we conclude that variation within the “Response to Hypoxia” GO category genes is unlikely to have the same effect on Hb levels in Ethiopians as that observed in Tibetans . A widely used family of approaches for the detection of beneficial alleles uses information about the haplotype structure around the selected site [34] , [35] . However , these approaches have adequate power only in the case of new advantageous alleles that were driven to high frequency by natural selection , i . e . ≥70% [34] , [36] . Because the largest allele frequency differences observed between HA and LA among Amhara or Oromo is less than 40% , these approaches are unlikely to be powerful in this setting . Therefore , to identify alleles that contribute to genetic adaptations to HA in Ethiopians , we used two complementary approaches that focus on the divergence of allele frequency between HA and LA populations . One of these approaches was previously used to successfully identify adaptive alleles in Tibetan highlanders [18] . The first approach is based on the population branch statistic ( PBS ) [18] , which summarizes information about the allele frequency change ( PBSA_BC ) at a given locus in the history of a population ( population A ) since its divergence from two populations ( population B and C ) so that a high PBSA_BC value represents a marked change in allele frequency on the branch leading to population A . This approach was previously used to detect advantageous alleles in Tibetans relative to Han Chinese and Europeans [18] . We tested for an excess of high allele frequency differentiation ( i . e . large PBS values ) on the branch leading to the Ethiopian populations in SNPs within candidate genes for response to hypoxia ( i . e . , genes within “Response to Hypoxia” GO category ) relative to SNPs in all other genes ( Table S22 lists all the population trios tested ) . Specifically , we calculated the ratio of the proportion of SNPs in hypoxia genes versus the proportion of SNPs in all other genes in the top 0 . 5% , 1% and 5% of the distribution of PBS values and used bootstrap resampling to assess the significance of the excess of large PBS values . Although an excess was observed in most population trios ( Table S22 ) , this excess was rarely statistically significant; this finding suggests that levels of linkage disequilibrium in hypoxia genes tend to be higher than in other genes and that this feature may be a confounder in tests for selection [18] . A significant excess of large PBS values in hypoxia genes was observed only in the HA Amhara and the entire Amhara sample ( Table S22 and Figure S17A and S17B ) , thus suggesting that HA Amhara indeed evolved genetic adaptations to hypoxic environments . When we extended this analysis of these same population trios to additional gene classifications ( i . e . BioCarta , KEGG , Gene Ontology ) , we found significant enrichments for SNPs in gene sets related to cell cycle control , response to DNA damage and DNA repair ( Table 2 ) . Interestingly , however , the SNPs with the highest PBS values are found in genes with a well-established role in pathogen response ( Tables S23 and S24 ) . More specifically , the SNPs with the highest PBS values are located within the major histocompatibility complex class II DR alpha ( HLA-DRA ) . Moreover , the null allele ( FY*0 ) at the Duffy blood group locus , which protects against Plasmodium vivax malaria [37] and predicts white blood cell and neutrophil counts [38] , has the second highest value . Consistent with expectations based on the protective effects of the FY*0 allele against malaria , its frequency is lower at HA compared to LA , where malaria is endemic ( 51 . 5% versus 74 . 1% ) . Therefore , these results suggest that , in Ethiopian populations , differences in pathogen loads between LA and HA environments result in stronger selective pressures compared to differences in oxygen levels . SNPs with large , even though not extreme PBS scores and lying within genes known to play an important role in hypoxia are of potential interest for follow up studies . These genes include: Cullin3 ( CUL3 ) , which potentiates HIF-1 signaling [39] , as well as adrenergic beta receptor kinase 1 ( ADRBK1 ) [40] , coronin actin binding protein 1B ( CORO1B ) [41] , anti-silencing function 1 homolog B ( ASF1B ) [42] and MAPK-activated protein kinase MK2 ( MAPKAPK2 ) [43] , which are all down-regulated under hypoxia ( Tables S23 and S24 ) . None of those large PBS SNPs were significantly associated with Hb or O2 sat ( Tables S25 and S26 ) , but a SNP within utrophin A ( UTRN ) - rs7753021 - reached nominal levels of significance with O2 sat ( p = 0 . 005; Table S26 ) . UTRN expression correlates with oxidative capacity [44] and increases with chronic physical training [45] . Slow-twitch muscles , which are associated with endurance performance , have high levels of UTRN [46] . In a complementary analysis , we developed a multiple regression ( MR ) approach to identify SNPs that show high allele frequency differentiation in HA populations relative to predictions based on a large set of worldwide population samples . This method should also be able to predict allele frequencies appropriately in a situation where the target population is admixed . In this approach , we used allele frequency data from 61 LA populations ( including the HGDP and several other populations ) to predict the expected allele frequencies in the HA Amhara . We focused on the HA Amhara because they have lived at HA for a longer period of time and exhibit distinct patterns of Hb and O2 sat levels compared to the Oromo ( Figure 1 ) . In addition , we omitted the LA Amhara in an attempt to reduce the effect of gene flow between altitudes , which could potentially reduce our power to detect adaptive divergence . We used all SNPs to estimate the best-fitting regression coefficients for each population: that is , these are the coefficients that generate the lowest mean square error in predicting the HA Amhara allele frequencies . The populations with the largest regression coefficients in the Amhara regression model are from geographically proximate populations in East Africa ( Maasai , Luhya and LA Oromo ) and from the Middle East and Southern Europe ( see Figure S18 ) . We reasoned that changes in allele frequencies due to high altitude adaptation would be detectable as departures ( i . e . large residuals ) from the predicted allele frequencies based on all other populations As for the PBS analysis , we tested for an excess of SNPs with high allele frequency differentiation using the MR statistic for genes within “Response to Hypoxia” GO category relative to SNPs in all other genes and we used a bootstrap procedure to assess the significance of the observed excess . An excess was observed for all tail cut-offs , but only one reached statistical significance ( Table 2 ) . Other gene sets that showed a significant enrichment of SNPs with strong MR signals include chromosome organization and biogenesis , DNA repair , histone modification . These findings are consistent with the pattern observed in the PBS analysis , indicating that they are robust to the choice of populations used in the test ( Table 2 ) . Among the SNPs with the largest MR scores , there are several SNPs in hypoxia genes , which may be of potential interest for follow up studies ( Table S27 ) . The SNP ( rs12510722 ) , which shows the 4th highest MR scores , lies within the alcohol dehydrogenase 6 ( ADH6 ) gene , whose expression is affected by pseudohypoxia [47] . The SNPs with the 8th and the 15th highest MR score ( rs2660342 and rs2660343 , respectively ) are within 100 kb from solute carrier family 30 member 9 ( SLC30A9 ) and transmembrane protein 33 ( TMEM33 ) . SLC30A9 is up-regulated by hypoxia [27] while TMEM33 is down-regulated under hypoxia and up-regulated after knockdown of HIF1A [41] . Methylation is an epigenetic modification that is known to play a crucial role in the cellular response to hypoxia [48] . Since HA adaptation could be in part maintained by methylation , we measured methylation levels at 27 , 578 CpG sites in 17 HA and 17 LA Amhara and 17 HA and 17 LA Oromo . CpG methylation levels were tested in DNA extracted from blood in the Amhara and from saliva in the Oromo . To avoid confounding due to differences in methylation across tissues , we performed the comparison across altitudes within each ethnic group . In Oromo , four CpG sites reached significance after multiple test correction ( p<1 . 85×10−6 ) , but the closest genes are not known hypoxia candidate genes: apolipoprotein B mRNA editing enzyme catalytic polypeptide-like 3G ( APOBEC3G ) , metallothionein 1G ( MT1G ) , paired-like homeodomain 2 ( PITX2 ) and olfactory receptor family 2 subfamily K member 2 ( OR2K2 ) ( Table S28 ) . Interestingly , APOBEC3G codes for a well-established cellular antiviral protein and a specific inhibitor of human immunodeficiency virus-1 ( HIV-1 ) infectivity [49] , [50] . The MT1G gene also has a role in HIV-1 infection because it upregulates MT1G expression in immature dendritic cells , which in turn facilitates the expansion of HIV-1 infection [51] . Although the prevalence of HIV was not surveyed in our fieldwork , HIV/AIDS is known to be a major health problem in Ethiopia [52] , [53] . While these functions for APOBEC3G and MT1G point to a role for methylation in defense against pathogens , MT1G also plays a role in the response to hypoxia as its promoter is induced by vascular endothelial growth factor ( VEGF ) , which in turn contributes to the prosurvival and angiogenic functions of VEGF [54] . Likewise , expression of PITX2 is required for normal hematopoiesis [55] , [56] , raising the interesting scenario that methylation of this gene may influence beneficial phenotypes in the response to hypoxia . Some of the CpG sites with nominally significant ( p<6 . 7×10−5 ) differences in methylation between HA and LA are close to genes that are differentially expressed in response to hypoxia; these genes include: toll-like receptor 6 ( TLR6 ) [57] , mif two 3 homolog 1 ( SUMO1 ) [27]; phosphodiesterase 4A ( PDE4A ) [58] and human immunodeficiency virus type I enhancer binding protein 2 ( HIVEP2 ) [42] . In Amhara , no CpG site showed a methylation difference that reached significance after multiple test correction ( Table S29 ) . However , we note that the 3rd most significant differentially methylated CpG site ( p = 8 . 03×10−05 ) was closest to Glutathione-S-Transferase ( GSTP1 ) whose expression is increased by prolonged hypoxia [59] and whose loss of expression correlates with methylation in prostate cancer [60] . Hypoxia also regulates the expression of genes close to other differentially methylated CpG sites in Amhara ( p<1 . 5×10−3 ) : protein regulator of cytokinesis 1 ( PRC1 ) [42] , protein tyrosine phosphatase receptor type O ( PTPRO ) [41] , ring finger protein 146 ( RNF146 ) [27] and Ras-related GTP binding D ( RRAGD ) [41] . Finally , no significant excess of methylation differences between LA and HA populations was observed at the genome-wide level in Oromo or Amhara ( Figure S19 ) nor did we find a significant enrichment of methylation differences between LA and HA populations for gene sets defined by BioCarta or KEGG pathways and by Gene Ontology categories ( data not shown ) .
HA human populations across the world allow studying independent realizations of the adaptive process in response to the same selective pressure , i . e . hypoxia , thus providing an excellent opportunity to investigate how natural selection shapes the genetic architecture of adaptive traits . To make progress on these enduring questions , we have sampled two closely related ethnic groups in the Ethiopian highlands that include both HA and LA residents , thus allowing comparisons across altitudes within and between ethnic groups . Of these two groups , the Oromo have moved to HA only 500 years ago [61] , [62] , thus making it unlikely that genetic adaptations evolved in this group . In contrast , the Amhara have a history of HA residence of at least 5 ky and possibly as far as 70 ky [24] , [25] . Because previously identified selection signals [63] , [64] , [65] , [66] occurred within a similar period of time , including HA adaptations [18] , we conclude that enough time has elapsed since the Amhara moved to HA for genetic adaptations to have taken place . Consistent with this idea , we observe significant phenotypic differences between Amhara highlanders and the more recent HA residents , i . e . the Oromo . While HA Amhara are characterized by mildly elevated Hb levels ( similar to Tibetans ) and no or mildly reduced O2 sat [13] , the HA Oromo sample resembles acclimatized lowlanders with a response characterized by elevated Hb concentration and marked reduction in O2 sat . Our data indicates that Amhara and Oromo are very similar at the genome-wide level , therefore , the observed phenotypic differences are likely due to the different histories of HA occupation . In addition to these phenotypic comparisons , our genomic analyses of these two ethnic groups resulted in several important observations that shed new light on the biology of HA adaptations and that are discussed in detail below . First , in a GWAS of Hb levels in Amhara , we find a genome-wide significant signal of association as well as an excess of low p-values . In addition , the second most strongly associated SNP is found within the RORA gene , which belongs to HIF1 pathway and is , therefore , an excellent candidate gene for hypoxia response phenotypes . Additional strong associations were observed in other candidate hypoxia genes , such as COL6A1 , SLC30A9 , and HGF . We looked at the Amhara data of Scheinfeldt et al [21] to test for replication of the association signal at these genes . Two of them showed suggestive associations with Hb levels ( p = 0 . 06 and p = 0 . 15 for SLC30A9 and RORA , respectively ) , with β values ( −0 . 60 and 1 . 31 , respectively ) consistent with ours ( −0 . 67 and 0 . 92 , respectively ) . It should be noted that the replication test was performed in only 21 Amhara samples in Scheinfeldt et al [21] for which age and BMI data were available and who had Hb levels within the normal range; thus , the lack of replication may well be due to the very low power of the replication sample . We note that , though variation in EPAS1 and EGLN1 has been consistently associated with Hb levels in Tibetans , no genome-wide significant association signal and no excess of low p-values were observed ( see Figure S5 in Simonson et al [16] , [17] ) . While the signals we detected await replication , it is interesting to note that their effect sizes are as high as those found in Tibetans for SNPs in EPAS1 , thus raising the interesting scenario that selection may have favored alleles with similar effect sizes on Hb levels even though the specific loci contributing to the trait are different . Some interesting patterns are beginning to emerge with regard to the genetic contribution to variation in Hb levels and O2 sat , the two phenotypes that have been most widely studied in highlander populations . No evidence of a genetic contribution to O2 sat in Amhara , Oromo , and the combined Ethiopian sample could be detected ( Figures S7 , S8 , S9 , S10 , S11 , S12 , S13 , S14 , S15 ) . This is true also in the Tibetans , even though segregation analysis detected a major O2 sat locus , which is also associated with reproductive success [15] . Therefore , the data so far suggest that while genetic factors contribute to variation in Hb levels , their importance in O2 sat is lower . This is consistent with studies in Tibetans and Andeans showing a markedly lower heritability for O2 sat compared to Hb levels; indeed , the O2 sat heritability in Andeans was not significantly different from zero [67] , [68] , [69] . More data , especially at the genome-wide level , are needed to elucidate the contribution of genetic factors to these two phenotypes . Second , to compare the genetic bases of Hb variation with the Tibetans , we tested for an association between SNP genotypes and Hb levels within the Ethiopians . Although we had appropriate power , none of the SNPs within 10 kb of the EPAS1 and EGLN1 genes or the genes in the hypoxia pathway , including SNPs previously associated with Hb variation ( and signatures of natural selection ) in Tibetans , associated with Hb . Therefore , we can rule out that the SNPs and loci contributing to Hb variation and showing selection signals in Tibetans affect the same trait in Ethiopians even though Tibetans and Amhara have lower Hb levels compared to all other highlanders . Alternatively , if the same variant affects Hb levels in both populations , their effect sizes in Amhara must be markedly lower than those reported for the Tibetans . Third , by using approaches based on allele frequency divergence , we find that outlier SNPs in the HA Amhara sample are not in hypoxia response genes , but in loci known to play a role in immune defense . These include variation in the HLA-DRA locus and the null allele at the Duffy blood group locus; none of these variants is associated with Hb levels or O2 sat . Interestingly , malaria and schistosomiasis were prevalent in the LA , but not in the HA Amhara communities sampled in this study ( Text S1 ) , reflecting important differences in pathogens between HA versus LA environments . Indeed , epidemiological studies in the areas near the LA sampling sites for the Amhara and Oromo reported malaria prevalence of 39 . 6% and as much as 25% of malaria morbidity is due to P . vivax [70] , [71] , [72] . Therefore , the immune defense variants with extreme frequency divergence represent excellent candidates as selection targets . These findings are important in several respects . First , they indicate that , because HA and LA habitats differ by multiple environmental stresses , signals of allele frequency divergence cannot be unambiguously attributed to hypoxia without additional information about the gene function or the specific phenotypic effects of the alleles . Second , they further corroborate that the Amhara populations are indeed adapted to spatially-varying selective pressures , despite likely high levels of gene flow between HA and LA communities . Third , the fact that variants in hypoxia response genes are not outliers in these analyses suggests that adaptations to pathogens and to hypoxia have a different genetic architecture or that the intensity of pathogen-related selective pressures is stronger than those due to hypoxia . Overall , these findings highlight the opportunities and challenges of ecological genomic studies and point to the power of approaches that use environmental information combined with phenotypic data collected in the field . Although selection has not created extreme HA versus LA frequency shifts in hypoxia genes , we find the SNPs within candidate genes for response to hypoxia as a group show an excess of allele frequency differentiation based on the PBS analysis performed in the Amhara ( Table 2 ) . This approach , however , requires specification of a set of three populations to be tested . When we used our MR approach , which uses information from all populations simultaneously , the response to hypoxia genes did not show a significant excess ( Table 2 ) . In contrast , both approaches detected a significant excess of allele frequency divergence in genes involved in cell cycle control , DNA repair , DNA damage , chromatin structure and modification , consistent with the known role of oxygen sensing in the regulation of cell proliferation [73] . Therefore , our findings raise the possibility that genetic variation in these pathways can contribute to adaptations to HA . Fourth , though we did not observe a genome-wide excess of methylation differences between HA and LA samples , we found genome-wide significant signals in the Oromo in genes with a known function in pathogen defense or in the biology of hypoxia , i . e . VEGF signaling and hematopoiesis . Interestingly , no genome-wide significant differential methylation was observed in the Amhara; this may be due to the fact that DNA from different tissues was analyzed in the two ethnic groups . However , given the difference in the history of HA residence between Amhara and Oromo , it is also possible to speculate that epigenetic modifications play a role in the early phases of adaptations to new environments and that this role is replaced over time by adaptations at the genetic level . It has been proposed that epigenetic modifications are important in ecological adaptations [74] and methylation is known to play a crucial role in the cellular response to hypoxia [48] . A previous study showed that gene expression differences observed between Moroccan populations were not explained by methylation differences on the tested 1 , 505 CpG sites [75] . Our study is more comprehensive having interrogated 27 , 578 CpG sites in a larger sample size . Though our results do not unambiguously point to a major role for methylation in the adaptations to HA , they suggest that further studies using DNA extracted from different tissues and including additional epigenetic modifications , in addition to methylation , are warranted . In conclusion , studies of genetic variation in indigenous populations with long-time residence at HA are giving rise to a composite picture regarding the genes contributing to and the genetic architecture of HA adaptations . Clearly , different loci contribute to Hb levels in Ethiopians and Tibetans . Moreover , while some aspects ( i . e . phenotypic effect sizes ) of the genetic architecture of Hb levels may be similar , others ( i . e . the allele frequency shifts due to selection ) are different . Additional examples of environmental pressures that acted on different human populations include malarial endemia , low UVB radiation levels , and an adult diet rich in dairy products . Detailed genome-wide studies of parallel adaptations to these selective pressures are needed to elucidate the impact of natural selection on the genetic architecture of complex adaptive traits .
All participants in the study gave informed consent . The studies were approved by the Institutional Review Boards of Case Western Reserve University and of the University of Chicago and by the Ethiopian Science and Technology Council Ethics Review Board . Samples were drawn from native residents of the Semien Mountains area of northern Ethiopia inhabited mainly by Christian Amhara and the Bale Mountains area of southern Ethiopia inhabited mainly by Muslim Oromo ( also referred to as Galla , Boran and Gabbra ) . For each ethnic group , we sampled individuals at HA and LA . The LA samples were chosen to achieve the maximum altitude contrast , yet avoid confounding due to the presence of endemic malaria; all LA samples were from agropastoral people reporting the same ethnicity as the HA samples and no visits to altitudes above 2500 m in the past 6 months . For details on the sampled populations and their ecology , see Text S1 . DNA was extracted from blood samples provided by 192 Amhara individuals living at 3700 m in the Simien Mountains National Park or at 1200 m in the town of Zarima . Forty-seven of these individuals were sampled in 1995 and previously described [13] , [76] , the remaining 145 individuals were sampled in a separate expedition in 2005 . DNA was extracted from , saliva samples provided by 118 Oromo individuals and collected using Oragene DNA sample collection kits; 79 individuals lived at 4000 m in the Bale Mountains National Park while 39 individuals lived at 1560 m in the town of Melkibuta . The study participants were healthy ( refer to Text S3 for details ) . Hb and O2 sat of Hb were measured in all individuals . Hb was determined in duplicate using the cyanmethemoglobin technique ( Hemocue Hemoglobinometer , Hemocue AB , Angelholm , Sweden ) , immediately after drawing a venous blood sample . O2 sat was determined by pulse oximetry ( Criticare Models 503 and SpO2 ) as the average of six readings taken 10 seconds apart . The samples were genotyped using Hap650Yv3 ( n = 46 ) , Human1M-duoV3 ( n = 112 ) , and Human Omni-Quad1 ( n = 160 ) Illumina arrays at Southern California Genotyping Consortium . Nineteen samples with less than 93% genotype call rate were omitted from the analysis . We used the program RELPAIR 2 . 0 . [77] to test for hidden relatedness in the samples by using 3 independent sets of 1800 autosomal and 200 X-linked SNPs; individuals with relationships closer than first cousins to any other individual in the sample were omitted from the analysis . After applying these filters , 260 unrelated samples remained: 102 HA Amhara , 60 LA Amhara , 63 HA Oromo and 35 LA Oromo . Principal component analysis ( PCA ) of the genotype data did not detect any major outlier in either population ( see Text S2 and Figure S3 . Due to the incomplete overlap between SNPs on the three genotyping arrays , genotypes for 1 , 819 , 369 HapMap3 SNPs were imputed by using the program IMPUTE2 [78] and the HapMap populations as a reference panel: Utah residents with Northern and Western European ancestry from the CEPH collection ( CEU ) , Han Chinese in Beijing , China ( CHB ) , Japanese in Tokyo , Japan ( JPT ) , Luhya in Webuye , Kenya ( LWK ) , Maasai in Kinyawa , Kenya ( MKK ) , Toscani in Italy ( TSI ) and Yoruba in Ibadan , Nigeria ( YRI ) samples . A total of 1 , 297 , 134 autosomal SNPs with minor allele frequency ( MAF ) higher than 0 and imputation accuracy higher than 90% were used in the downstream analyses ( Figure S20 ) . Phenotype-genotype associations were tested at each SNP with MAF>0 . 1 by linear regression using the whole-genome association analysis toolset in PLINK [79] . To determine whether there was an excess of low p-values taking into account the large number of tests performed , the distribution of observed p-values from the linear regression tests were compared to a null distribution obtained by permuting 100 times the phenotype ( and corresponding covariate variables – see below ) value across individuals and running the same linear model . The results were visualized by means of quantile-quantile ( QQ ) plots and the 95% confidence interval ( CI ) was estimated by permutations . Gender , body mass index ( BMI ) , altitude , ethnic group and year of sampling were used as covariates in the linear regression . As an alternative approach , we grouped the samples based on their gender , altitude , ethnic group and year of sampling and we quantile normalized the observed phenotypes within each group; these phenotypes were then pooled across groups before testing for an association with genotype by linear regression . This latter approach preserves information about the individual ranks without introducing a bias due to the effects of covariates . However , it does not preserve information about the phenotype values , thus leading to a reduction in power . The results of these two approaches correlated highly ( r2>0 . 92 ) ; therefore , only the results for the linear regression with the covariates are shown . The population branch statistic ( PBSA_BC ) was used to summarize the amount of allele frequency change in the history of population A since its divergence from two related populations , B and C [18] . In a complementary approach based on multiple regression ( MR ) , we used the allele frequencies for the LA Oromo and world-wide population samples ( i . e . , Human Genome Diversity Project ( HGDP ) panel populations [80] and 4 HapMap Phase III populations -LWK , MKK , TSI and Gujarati Indians in Houston , Texas ( GIH ) - ( www . hapmap . org ) ) to detect SNPs whose HA Amhara frequencies deviate most from estimated frequencies . Briefly , denoting the observed allele frequencies in the p populations by X1 , X2 , … , Xp and the expected allele frequency within the HA Amhara by y , we can predict the allele frequency for the ith SNP using the linear model yi = β0+β1Xi1+ β2Xi2+…+βpXip+εi . Using the data from all genotyped SNPs that overlap among the p populations and applying multivariate linear regression , we found b0 , b1 , b2 , … , bp , which are estimates of parameters β0 , β1 , β2 , … , βp and estimated the magnitude of the residual for the ith SNP as |εi| = |yi− ( b0+b1X1+b2X2+…+bpXp ) | . We refer to this residual as the MR score . To test for an enrichment of allele frequency differentiation in candidate hypoxia genes , we focused on the 28 genes belonging to the “Response to hypoxia” category in the Gene Ontology database ( GO:0001666 ) . We then compared the proportion of SNPs within 10 kb of each gene in this category to that of SNPs within 10 kb of all other genes in the tail of the PBS and MR score distributions . Given the arbitrary nature of choosing a single cutoff for the tail of the distribution , we set three cutoffs ( 5% , 1% and 0 . 5% ) . In other words , we looked at the top 5% , 1% and 0 . 5% of all PBS and MR values and asked whether there is an enrichment of hypoxia gene SNPs relative to all other genic SNPs for each tail cut-off . A value of 1 represents no excess and a value greater than 1 represents enrichment in the tail of the distribution . SNPs are likely to cluster along the genome due to linkage disequilibrium , thus reducing the number of independent signals contributing to an observed enrichment . To account for this possibility , we found the confidence interval for the enrichment using a bootstrap approach described in Hancock et al [81] . An enrichment of hypoxia SNPs was considered significant ( with a one-tailed test ) if at least 95% of the 1000 bootstrap replicates were enriched ( i . e . , had a ratio above 1 ) . Methylation levels were measured at 27 , 578 CpG sites in 17 HA and 17 LA Amhara and Oromo DNA samples , for a total of 68 individuals , using six Infinium HumanMethylation27 arrays at Southern California Genotyping Consortium . Two LA Amhara sample data were discarded due to low data quality . In each ethnic group , the HA methylation level of each CpG site was compared to corresponding LA levels using the following linear model correcting for age , gender , and the methylation array: lm ( % methylation ∼ altitude + gender + age + [array] ) . Two comparisons were performed: ( 1 ) HA versus LA Amhara and ( 2 ) HA versus LA Oromo . As for the genotype-phenotype association , excess of differential methylation between HA and LA was tested by comparing the observed p-value distribution to the null distribution obtained by permuting 100 times the altitude label across individuals and running the same linear model . Permutation based 95% CI was estimated . | Although hypoxia is a major stress on physiological processes , several human populations have survived for millennia at high altitudes , suggesting that they have adapted to hypoxic conditions . Consistent with this idea , previous studies have identified genetic variants in Tibetan highlanders associated with reduction in hemoglobin levels , an advantageous phenotype at high altitude . To compare the genetic bases of adaptations to high altitude , we collected genetic and epigenetic data in Ethiopians living at high and low altitude , respectively . We find that variants associated with hemoglobin variation among Tibetans or other variants at the same loci do not influence the trait in Ethiopians . However , we find a different variant that is significantly associated with hemoglobin levels in Ethiopians . Approaches based on the difference in allele frequency between high- and lowlanders detected strong signals in genes with a clear role in defense from pathogens , consistent with known differences in pathogens between altitudes . Finally , we found a few genome-wide significant epigenetic differences between altitudes . These results taken together imply that Ethiopian and Tibetan highlanders adapted to the same environmental stress through different variants and genetic loci . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"medicine",
"population",
"genetics",
"global",
"health",
"epigenetics",
"biology",
"evolutionary",
"genetics",
"adaptation",
"natural",
"selection",
"genetics",
"genomics",
"evolutionary",
"biology",
"genomic",
"evolution",
"evolutionary",
"processes",
"genetics",
"and",
"genomics",
"human",
"genetics"
] | 2012 | The Genetic Architecture of Adaptations to High Altitude in Ethiopia |
Miltefosine ( MIL ) is an oral antileishmanial drug used for treatment of visceral leishmaniasis ( VL ) in the Indian subcontinent . Recent reports indicate a significant decline in its efficacy with a high rate of relapse in VL as well as post kala-azar dermal leishmaniasis ( PKDL ) . We investigated the parasitic factors apparently involved in miltefosine unresponsiveness in clinical isolates of Leishmania donovani . L . donovani isolated from patients of VL and PKDL at pretreatment stage ( LdPreTx , n = 9 ) , patients that relapsed after MIL treatment ( LdRelapse , n = 7 ) and parasites made experimentally resistant to MIL ( LdM30 ) were included in this study . MIL uptake was estimated using liquid chromatography coupled mass spectrometry . Reactive oxygen species and intracellular thiol content were measured fluorometrically . Q-PCR was used to assess the differential expression of genes associated with MIL resistance . LdRelapse parasites exhibited higher IC50 both at promastigote level ( 7 . 92 ± 1 . 30 μM ) and at intracellular amastigote level ( 11 . 35 ± 6 . 48 μM ) when compared with LdPreTx parasites ( 3 . 27 ± 1 . 52 μM ) and ( 3 . 85 ± 3 . 11 μM ) , respectively . The percent infectivity ( 72 hrs post infection ) of LdRelapse parasites was significantly higher ( 80 . 71 ± 5 . 67% , P<0 . 001 ) in comparison to LdPreTx ( 60 . 44 ± 2 . 80% ) . MIL accumulation was significantly lower in LdRelapse parasites ( 1 . 7 fold , P<0 . 001 ) and in LdM30 parasites ( 2 . 4 fold , P<0 . 001 ) when compared with LdPreTx parasites . MIL induced ROS levels were significantly lower ( p<0 . 05 ) in macrophages infected with LdRelapse while intracellular thiol content were significantly higher in LdRelapse compared to LdPreTx , indicating a better tolerance for oxidative stress in LdRelapse isolates . Genes associated with oxidative stress , metabolic processes and transporters showed modulated expression in LdRelapse and LdM30 parasites in comparison with LdPreTx parasites . The present study highlights the parasitic factors and pathways responsible for miltefosine unresponsiveness in VL and PKDL .
Visceral leishmaniasis ( VL ) is the most severe form of leishmaniasis , being fatal if left untreated . More than 90% of the global burden of VL occurs in just six countries: India , Bangladesh , Sudan , South Sudan , Brazil and Ethiopia [1] . India alone shares almost 50% of the world’s VL burden with Bihar , Uttar Pradesh , West Bengal and Jharkhand as the endemic states . Post kala-azar dermal leishmaniasis ( PKDL ) , a dermal sequel reported in 5–15% of VL treated cases in the Indian subcontinent is reported to serve as a major reservoir of Leishmania [2 , 3] . Miltefosine ( MIL ) was introduced in the Indian subcontinent for VL treatment with cure rate of more than 94% [4] . However , the relapse rate has increased to 10% in India [5] and even more in Nepal where it ranged from 10% to 20% during 6 month and 12 month follow up , respectively [6] . The relapse rate in PKDL has also increased substantially ( from 4% to 15% ) [3] . The mean MIL IC50 value of the parasites obtained from pretreatment cases ( often referred to as MIL sensitive cases ) is lower in comparison to mean MIL IC50 value of the parasites obtained from the relapse cases in both Indian ( VL and PKDL ) and Nepalese ( VL ) patients [3 , 5–8] . Increased infectivity and metacyclogenesis of parasites have been linked with high relapse rate in MIL treated Nepalese patients [6] . Emergence of MIL tolerant parasites due to prolonged exposure to MIL may be associated with decline in drug efficacy , as observed in case of antimony resistance [3 , 9] . Experimental resistance to MIL was shown to be readily induced in vitro [10–12] . Impairment in drug uptake machinery involving amino-phospholipid translocase miltefosine transporter ( LdMT ) and an accessory protein ( LdRos3 ) was proposed to be the most likely mechanism of resistance [10–13] . Recent study from our group on comparative transcriptome profiling of MIL sensitive and laboratory generated resistant parasites revealed altered expression of genes involved in thiol metabolism , drug transport , protein translation and folding , DNA repair and replication machinery [12] . Miltefosine tolerant parasites of L . donovani exhibited greater ability to resist reactive oxygen species than the sensitive parasites [14] . Suppression of oxidative stress induced apoptotic cell death has been reported in MIL resistant L . donovani parasites [15] . Majority of evidence regarding parasitic factors associated with the development of miltefosine resistance is restricted to laboratory adapted resistant parasites and limited information is available with respect to clinical isolates of L . donovani [10–12] . Understanding the parasitic factors and molecular mechanisms involved in miltefosine unresponsiveness in clinical isolates of L . donovani is crucial to monitor drug efficacy and its longevity . In the present study , we have investigated several parasitic factors including in vitro MIL susceptibility , infectivity to macrophages , metacyclogenesis , drug accumulation , intracellular thiol content , reactive oxygen species ( ROS ) and targeted transcript profile of genes implicated in MIL resistance , in parasites obtained from the cases of VL and PKDL that relapsed ( LdRelapse ) in comparison to the parasites from pretreatment cases ( LdPreTx ) , and laboratory adapted MIL resistant parasites ( LdM30 ) .
Clinical isolates of L . donovani used here were reported in our previous studies [3 , 7 , 12] . Parasite isolates were prepared from splenic aspirates of VL patients or from dermal lesions of PKDL patients reporting to KAMRC , Muzaffarpur , Bihar or Safdarjung Hospital ( SJH ) , New Delhi , under the guidelines of the Ethics Committee of the respective Institutes . Parasites were isolated from each of the relapse cases at the time of reported relapse ( after completion of MIL treatment at four , six and seven months VL and at 12 and 18 months in PKDL ) and were designated as MHOM/IN/year/code/month ( at which relapse occurred ) . Experimental MIL resistant parasites were prepared as described earlier [12] . Briefly , wild type L . donovani promastigotes were adapted to grow under high MIL pressure by in vitro passage with stepwise increase in the MIL concentration ( 2 . 5 , 5 , 7 . 5 , 10 , 20 and 30 μg/ml ) in medium M199 to generate MIL resistant parasite ( LdM30 ) . At each step , parasites were cultured for at least 4–6 passages to attain steady growth comparable to the wild type counterpart . The susceptibility towards MIL of adapting parasites was assessed at each step during adaptation . Adapted parasites were maintained under miltefosine pressure ( 30 μg/ml ) in M199 medium . LdAG83 parasite was used as standard reference strain . Parasites were routinely grown in Medium 199 ( Sigma , St . Louis , MO USA ) supplemented with 10% heat-inactivated foetal bovine serum ( HI FBS , Gibco , USA ) , 100 IU/ml penicillin G , and 100 μg/ml streptomycin at 25°C . Mouse derived peritoneal macrophages were cultured in RPMI1640 medium supplemented with 10% FCS , 100 IU/ml penicillin G , and 100 μg/ml streptomycin at 37°C in 5% CO2 . Drug susceptibility of the parasites towards MIL was estimated at the promastigote stage as described previously [16] . Briefly , late log phase L . donovani promastigotes ( 105 promastigotes/well ) were seeded into 96 well culture plate containing MIL concentration ranging from 0 . 4 μM to 390 μM . After 72 h incubation at 25°C , 50 μl resazurin [0 . 0125% ( w/v in PBS ) ] was added , and plates were incubated for a further 24 h . Cell viability was measured fluorometrically ( λex 550 nm; λem 590 nm ) on Infinite M200 multimode reader ( Tecan ) . The results were expressed as percentage reduction in the parasite viability compared to untreated control wells . 50% inhibitory concentration ( IC50 ) was calculated by sigmoidal regression analysis . All experiments were performed at least thrice in quadruplicate . Similarly , in vitro MIL susceptibility was assessed at intracellular amastigote level by following macrophage-amastigote model described elsewhere [17] with modifications . Briefly , the mice peritoneal exudates derived macrophages were infected with late log phase parasites at a ratio of 10 parasites per macrophage , in a volume of 200 μl complete RPMI 1640 medium into 8-well chamber slides and incubated for 16 h at 37°C in 5% CO2 . Excess , non-adhered promastigotes were removed by washing and infected cells were re-incubated for 48 h , with MIL ( 1 , 5 , 10 , 20 and 40 μM ) . Macrophages were then examined for intracellular amastigotes after staining with Diff-Quik solutions . The number of L . donovani amastigotes was counted in 100 macrophages , at 1000x magnification . The IC50 was calculated using sigmoidal regression analysis . All experiments were performed at least thrice in duplicate . The percentage of L . donovani infected macrophages was assessed following macrophage-amastigote model as described elsewhere with modifications [17–19] . Primary peritoneal macrophages were extracted from female BALB/c mice , plated at 2x105 cells per well in RPMI 1640 medium supplemented with 10% FBS in 8-well chamber slides , and incubated at 37°C in 5% CO2 . Twenty-four hours later , the medium was gently removed and the macrophages were infected with late log phase promastigotes at a ratio of 10 parasites per macrophage , in a volume of 200 μl complete RPMI 1640 medium without MIL . After 6 h of infection , non-internalized parasites were washed off , and after addition of fresh complete RPMI medium the slide was further incubated for 18 h ( group 1 ) , 42 h ( group 2 ) and 66 h ( group 3 ) . Slides were fixed with methanol and stained with Diff-Quik solutions . A total of 500 macrophages were counted in randomly selected fields for each group at 1000x magnification . The percentage of infected macrophages was calculated by counting the number of infected cells out of 100 macrophages . The experiment was repeated thrice . Percent metacyclogenesis was assessed following the methods described elsewhere [20 , 21] . Briefly , stationary phase promastigotes from LdRelapse ( n = 7 ) and LdPreTx ( n = 9 ) were harvested at 2000g X 10 min and resuspended at a cell density of 2X108 cells/ml in 10 ml of complete M199 medium containing 50 μg/ml peanut agglutinin ( PNA ) ( Sigma , St . Louis , MO USA ) . Promastigotes were allowed to agglutinate at room temperature for 30 min and the sediment and the supernatant were recovered . The sediment was diluted to the initial volume in fresh complete medium containing 50 μg/ml PNA . Both fractions were centrifuged at 200g for 10 min and the supernatant obtained were centrifuged at 2000g to obtain PNA- ( metacyclic ) promastigotes . All steps were checked under a light microscope ( Nikon ECLIPSE TS100 ) . Percent metacyclic population was calculated by counting PNA- population out of the total promastigote cell density . MIL accumulation was studied according to the protocol described earlier , with slight modifications [22 , 23] . Briefly , log phase promastigotes ( 108 cells/ml ) were incubated with 100 μM MIL for 90 min , washed twice with PBS and digested by overnight incubation in 200 μl of 2 N HNO3 . The lysate was made up to 1 ml with PBS and analyzed for MIL content using liquid chromatography-mass spectrometry ( LC-MS ) . Chromatographic separation of MIL was carried out using an HP1290 liquid chromatograph system ( Agilent Technologies; Palo Alto , CA , USA ) consisting of a binary pump , degasser and auto-sampler . Chromatography was performed on a ZORBAX Eclipse XDB C8 column , maintained at 45°C . A mobile phase of 0 . 1% formic acid in methanol and water ( 95:5; v/v ) at a flow rate 0 . 45 ml/min was employed . The mass spectrometer detection system consisted of an AB sciex API 3000 LCMS with electron spray ionization ( ESI ) positive mode . The mass spectrometer was operated in MRM mode . The setting of the mass spectrometer was as follows: the spray voltage 5000 V , nebulizer gas flow 12 . 0 L/min , curtain gas flow 8 . 0 L/min , collision gas ( Nitrogen ) flow 6 . 0 L/min . The source temperature was 450°C . Declustering potential , focussing potential , entrance potential were 40 , 150 and 10 V , respectively . The transition of 408 . 5 to 125 . 0 was optimised for MIL with 40 V collision energy . Data were processed using Analyst TM software ( version 1 . 6 . 2; Sciex ) . The final concentration was calculated after multiplying the value obtained with dilution factor and presented in ng/ml . All stock and working solutions were prepared in methanol–water ( 1:1 , v/v ) . MIL stock solution was prepared at a concentration of 0 . 1 mg/ml for the preparation of calibration standards . This solution was further diluted to obtain working solutions with concentrations of 0 . 5 , 1 . 0 , 2 . 5 , 5 . 0 , 10 . 0 , 25 . 0 , and 50 ng/ml . Calibration standards were prepared freshly before each analytical run . MIL induced oxidative stress response was studied both at the promastigote and the intracellular amastigote stage following the macrophage-amastigote model in terms of accumulation of reactive oxygen species ( ROS ) described elsewhere with modifications [24] . All the experiments were carried out in triplicate . Promastigotes were exposed to MIL ( 5 , 10 , 15 and 20 μM ) for 48 h and centrifuged at 3000 rpm for 10 min . Cells ( 2x107 parasites/ml ) were resuspended in Hepes–NaCl ( 21 mM Hepes , 137 mM NaCl , 5 mM KCl , 0 . 7 mM Na2HPO4 . 7H2O , 6 mM glucose , pH 7 . 4 ) and incubated with 40 nM of cell permeable fluorescent probe 2’ 7’-Dichlorodihydrofluorescin Diacetate ( H2DCF-DA ) ( molecular probe ) for 45 minutes in dark at 26°C . Although the H2DCF-DA has limitation as a probe for direct measurement of ROS [25] , it has been widely used for assessment of ROS levels in Leishmania [24 , 26] . Fluorescence was measured at 495 nm excitation and 535 nm emission wavelength using cytofluorimeter ( Infinite M200 , Tecan , Switzerland ) . The mice peritoneal exudates derived macrophages were infected with LdPreTx , LdRelapse or LdM30 parasites in a ratio of 1:10 in 96 well tissue culture plates and incubated with 5% CO2−95% air mixture at 37°C in a CO2 incubator . After overnight infection , cells were washed with incomplete RPMI ( without FBS ) medium to get rid of the uninternalized promastigotes followed by addition of MIL at 20 μM . After 48 h of drug exposure , the cells were washed with incomplete RPMI , 30 μM of H2DCF-DA was added in 200 μl volume of incomplete RPMI and the plate was incubated further for 30 min . Uninfected macrophages with MIL exposure were taken as control to measure background fluorescence that was subtracted from the fluorescence measured at 495nm excitation and 535 nm emission wavelength ( Infinite M200 , Tecan , Switzerland ) . The fluorimetric measurements were expressed as mean fluorescent intensity units ( MFI ) that represented level of ROS . The intracellular thiol concentration was measured using thiol detection assay kit ( Cayman Chemical Company , MI , USA ) following manufacturer’s instructions . Briefly , 1x107 late log phase promastigotes were harvested by centrifugation at 2000 g at 4°C for 10 min . After washing with 1x PBS ( pH 7 . 4 ) cell pellet was homogenised in 0 . 5 ml cold buffer ( 100 mM Tris-HCI , pH 7 . 5 , containing 1mM EDTA ) and centrifuged at 10 , 000g for 15 min at 4°C . Supernatant was collected for the fluorometric estimation of thiol at 390 nm excitation and 520 nm emission wavelengths . Thiol concentration was estimated in 100 μl of supernatant distributed into 96 well plate from three independent preparations and calculated by using the equation obtained from the linear regression of the standard curve , substituting adjusted fluorescence value for each sample as follows Thiol concentration ( nmol/107promastigotes ) =Adjusted sample florescence- ( y intercept ) × sample dilutionslope Total RNA was extracted from 108 late log phase promastigotes using Trizol reagent according to manufacturer’s instructions . RNA clean up was performed using RNeasy Plus mini kit ( Qiagen , Gaithersburg , MD , USA ) as described by the manufacturer . The purified RNA was quantified using Nanodrop by estimating the absorbance at 260 and 280 nm . Briefly , first strand cDNA was synthesized from 5 μg of total RNA of L . donovani parasites using the Superscript II RNAse H reverse transcriptase enzyme ( Invitrogen , Carlsbad , CA , USA ) and Oligo dT primers ( Fermentas , USA ) according to the manufacturer’s protocol . Three independent RNA preparations were used for each Q-PCR experiment . Equal amounts of cDNA were run in triplicate and amplified in 25 μl reactions containing 1x Fast SYBR Green Mastermix ( Applied Biosystems , USA ) , 100 ng/ml forward and reverse primers . The sequences of the primers for all the genes ( n = 11; including 2 genes as internal control ) amplified in the study are given in Table 1 . Reactions were performed in triplicate in an ABI Prism 7500 ( Applied Biosystems , CA , USA ) . The analysis of gene expression was performed using the 2-ΔΔCT method . The data was presented as the fold change in the target gene expression in L . donovani parasites normalized to the internal control genes [27 , 28] ( GAPDH and α-Tubulin ) and relative to the LdAG83 reference strain of L . donovani . Statistical analysis of data was performed using Graph Pad Prism 5 software ( GraphPad Software Inc . , San Diego , CA , USA ) . Statistical significance was determined by Student’s t-test . Correlation was determined by Spearman’s rank test . P values <0 . 05 were considered significant . The study was approved by the Ethical Committee of the Institute of Medical Sciences , Banaras Hindu University ( Dean 2007-08/42 , dated 15-05-2008 ) , Varanasi and Institutional Ethics committee of Safdarjung Hospital & VMMC ( VMMC/SJH/PROJECT/22-10-2012/7 ) , New Delhi , India . Written informed consent was obtained from patients . Clinical isolates of L . donovani were analyzed anonymously . In vitro susceptibility experiments and ROS tolerance at amastigote level were carried out using mice peritoneal derived macrophages after approval from Institutional Animal Ethics Committee ( IAEC-3/2010 ) of National Institute of Pathology , New Delhi following guidelines for animal care and handling protocols recommended by committee for the purpose of control and supervision of experiments on animals ( CPCSEA ) .
MIL susceptibility profile of parasites isolated from pretreatment group LdPreTx ( n = 9 ) , from cases that relapsed after MIL treatment [VL ( n = 3 ) and PKDL ( n = 4 ) ] along with experimental MIL resistant isolates LdM30 ( n = 2 ) is given in Table 2 . The mean IC50 at promastigote stage of LdRelapse group parasites ( 7 . 92 ± 1 . 30 μM ) was significantly higher ( P<0 . 001 ) than LdPreTx ( 3 . 27 ± 1 . 52 μM ) while that of LdM30 parasites was the highest ( 76 . 50 ± 2 . 89μM ) ( Fig 1A ) . Likewise , the mean IC50 at intracellular amastigote stage of LdRelapse group parasites ( 11 . 35 ± 6 . 48 μM ) was significantly higher ( P<0 . 01 ) than LdPreTx ( 3 . 85 ± 3 . 11 μM ) while that of LdM30 parasites was the highest ( 77 . 98 ± 2 . 00 μM ) ( Fig 1B ) . The percentage of infected macrophages with parasites from LdRelapse group was significantly higher in comparison to the LdPreTx group at both 48 h ( LdRelapse = 77 . 85 ± 5 . 17 , LdPreTx = 60 . 11 ± 5 . 68 , P<0 . 001 ) and 72 h post infection ( LdRelapse = 80 . 71 ± 5 . 67 , LdPreTx = 60 . 44 ± 2 . 80 , P<0 . 001 ) , although infectivity was comparable at 24h post infection ( Fig 1C ) . LdM30 parasites exhibited similar infectivity as the wild type . This is as expected since these parasites were subjected to numerous in vitro passages during adaptation and continuous axenic cultivation of Leishmania promastigotes is known to lead to decreased infectivity [29] . The percent meatcyclogenesis was evaluated based on negative selection with PNA in culture . LdRelapse parasites exhibited significant increase ( P <0 . 001 ) in metacyclic promastigotes ( PNA- ) ( 44 . 81 ± 7 . 20% ) when compared with LdPreTx ( 20 . 27 ± 3 . 19% ) ( Fig 1D ) . The increase in metacyclic population in LdRelapse correlated strongly ( r = 0 . 92 ) with their infectivity to macrophages . The mean MIL accumulation ( ng/108 promastigotes ) in LdRelapse group parasites ( 102 . 6 ± 13 . 5 ) was significantly lower ( 1 . 7 fold , P<0 . 001 ) than that in LdPreTx ( 174 . 6 ± 28 . 9 ) , while it was the lowest ( 2 . 4 fold ) in LdM30 parasites ( 72 . 8 ± 2 . 8 ) . We found a decline in MIL accumulation with increasing IC50 value ( r = -0 . 78 ) ( Table 2 and Fig 1E ) . At promastigote stage both LdPreTx and LdRelapse showed dose dependent increase in ROS level which LdM30 parasites did not . At 20 μM of MIL , the ROS level was significantly lower ( P<0 . 001 ) in LdRelapse ( MFI 195 . 28 ) compared to LdPreTx group ( MFI 307 . 16 ) . ROS level in LdM30 parasites ( MFI 54 . 00 ) was significantly lower than both LdPreTx ( P<0 . 001 ) and LdRelapse , ( P<0 . 001 ) ( Fig 2A ) . Subsequent to MIL exposure , macrophages infected with LdRelapse parasites or LdM30 showed no increase in ROS accumulation , while macrophages infected with LdPreTx group showed significant increase ( ≥2 fold , P<0 . 01 ) in ROS levels ( Fig 2B ) . The tolerance to ROS was significantly ( P<0 . 05 ) higher in MIL exposed macrophages infected with LdRelapse and LdM30 compared to LdPreTx . The mean thiol content ( nmol/107 promastigotes ) of LdRelapse ( 120 . 0 ± 1 . 1 nmol ) was significantly ( P<0 . 001 ) higher than that of LdPreTx ( 82 . 3 ± 9 . 3 ) . The mean thiol content was the highest in LdM30 isolates ( 140 . 0 ± 5 . 5 ) ( Fig 2C ) . Thiol content of L . donovani isolates strongly correlated ( r = 0 . 78 ) with the IC50 value of the respective isolate . The genes ( n = 9 ) for expression profiling in clinical isolates were selected based on previous transcriptome studies on MIL sensitive and resistant L . donovani parasites ( Table 1 ) [12] . The expression levels of selected genes with respect to LdAG83 in clinical isolates ( LdPreTx n = 8 and LdRelapse n = 7 ) and LdM30 ( n = 1 ) are depicted in Fig 3 . Based on the fold changes , gene expression levels were designated as increased ( N-fold ≥1 . 5; p<0 . 05 ) , comparable ( N-fold ranging from -1 . 49 to 1 . 49 ) or decreased ( N-fold ≤-1 . 5; p<0 . 05 ) . Out of 9 genes evaluated , 6 genes ( TRYP , TSH , LPP , Cytb5Red , PGMPUT and MDRP ) showed upregulated expression ( 1 . 8 to 3 . 6 fold ) , while 3 genes ( ABCF2 , AAT and TCP20 ) were down regulated ( 1 . 6 to 1 . 9 fold ) in LdRelapse group parasites ( Fig 3 ) . The expression pattern of the selected genes in LdM30 parasites was consistent with the expression in LdRelapse group parasites . We observed significant upregulation in the mean expression level ± SD in LdRelapse vs LdPreTx parasite for TRYP ( 1 . 78 ± 0 . 44 vs -1 . 67 ± 0 . 46; P<0 . 01 ) ; TSH ( 2 . 40 ± 0 . 76 vs -2 . 90 ± 2 . 32; P<0 . 001 ) ; Cytb5Red ( 3 . 09 ± 1 . 16 vs -1 . 82 ± 0 . 41; P<0 . 001 ) ; LPP ( 2 . 01 ± 0 . 76 vs -3 . 20 ± 2 . 95; P<0 . 001 ) ; PGMPUT ( 2 . 30 ± 0 . 36 vs -1 . 68 ± 0 . 39; P<0 . 01 ) ; MDRP ( 3 . 61 ± 1 . 30 vs -1 . 58 ± 0 . 53; P<0 . 001 ) . There was significant down regulation in the mean expression level ± SD in LdRelapse vs LdPreTx parasite for ABCF2 ( -1 . 90 ± 0 . 72 vs 4 . 22 ± 3 . 26; P<0 . 01 ) , AAT ( -1 . 85 ± 0 . 56 vs 2 . 70 ± 1 . 45; P<0 . 001 ) and TCP20 ( -1 . 68 ± 0 . 34 vs 4 . 70 ± 2 . 90; P<0 . 001 ) .
This study revealed modulation in several parasitic factors in LdRelapse isolates in comparison with LdPreTx including ( i ) increased metacyclogenesis and infectivity ( ii ) reduced MIL uptake ( iii ) decreased MIL induced ROS accumulation ( iv ) increased intracellular thiol content . Analysis of the targeted expression of genes associated with these factors reiterated the above findings in MIL unresponsive clinical isolates of L . donovani . It is remarkable that the isolates from relapsed cases behaved very similar to the laboratory adapted MIL resistant parasites , however , the magnitude of modulation in gene expression , ROS and thiol levels , drug accumulation was lower in clinical isolates while the infectivity to macrophages was higher . The novelty of the study lies with the investigation of parasitic factors in parasites isolated from PKDL and VL cases at pre treatment and from patients that relapsed after miltefosine treatment . The laboratory generated MIL resistant parasite was used as the reference strain , since well defined miltefosine resistant clinical isolate of L . donovani are not available . We observed increased infectivity of LdRelapse parasites to macrophages when compared to LdPreTx group as an important parasitic factor associated with MIL unresponsiveness , unlike the in vitro adapted MIL resistant parasite LdM30 . High infectivity and metacyclogenesis have been shown to be linked with the high relapse rate in post MIL treated VL cases in Nepal [8] . The reduced uptake of drug , increased efflux and faster metabolism of drug linked with altered plasma membrane permeability have been cited as the most likely mechanisms responsible for development of resistance against miltefosine in Leishmania [10–12 , 30 , 31] . LdRelapse and LdM30 parasites showed significantly lower accumulation of MIL compared to LdPreTx . Further , MIL accumulation in L . donovani isolates was negatively correlated ( r = -0 . 78 ) with IC50 value . Down regulated expression of phospholipid translocase machinery LdMT and its beta subunit LdRos3 at plasma membrane surface of LdM30 parasites and mutation at these loci result in defective translocation of the drug [12] . Recently , whole genome sequencing approach in clinical isolates of L . infantum , revealed mutation in MIL transporter genes LiMT and its accessory protein LiRos3 [31] . In our earlier study , we did not observe polymorphism in nucleotide sequence of LdMT-LdRos in LdRelapse , unlike LdM30 which showed two point mutations in LdMT gene sequence [7 , 12] . This suggests that the reduced uptake of MIL may not be necessarily associated with point mutation in LdMT-LdRos complex in clinical isolates of L . donovani . The differential accumulation of MIL may be due to low influx or higher efflux as we observed upregulated expression of multidrug resistance like protein ( MDRP ) and downregulated expression of transporter ABCF2 at mRNA level in miltefosine unresponsive isolates . ROS mediated apoptosis like cell death in Leishmania induced by MIL is widely accepted as one of the mechanisms associated with antileishmanial activity of MIL [15 , 32 , 33] . To encounter host defence , parasites adopt strategies to overcome the oxidative environment and maintain redox homeostasis . The MIL induced ROS generation pattern was found to be comparable before and after MIL exposure in macrophages infected with LdRelapse and LdM30 parasites . In macrophages infected with LdPreTx parasites the level of ROS was significantly higher ( ≥2 fold , P<0 . 05 ) after MIL exposure . Thus , MIL unresponsive parasites ( LdRelapse and LdM30 ) had survival advantage over LdPreTx . Thiol metabolism plays an important role in combating drug pressure by suppressing oxidative stress . Although , the individual levels of glutathione and/or trypanothione were not determined , the total intracellular thiol content was found elevated in LdRelapse ( 1 . 5 fold ) and LdM30 isolates ( 1 . 7 fold ) compared to LdPreTx . Trypanothione synthetase ( TSH ) , cytosolic tryparedoxin peroxidase ( TRYP ) and cytochrome b5 reductase ( Cytb5Red ) have established role in antioxidant defence and in combating drug induced oxidative stress . We observed increased expression of all the 3 genes in LdRelapse and LdM30 compared to LdPreTx group . Increased intracellular thiol would help parasites to maintain redox homeostasis during MIL pressure . L . donovani parasites adopt metabolic reconfiguration by shift in primary carbon metabolism to subvert oxidative stress posed by antileishmanial drug [34] . The key enzymes of glycolytic pathway [12] and glucose uptake remain unaltered during stress in Leishmania [34] . We observed increased expression of phosphoglucomutase putative ( PGMPUT ) in LdRelapse and LdM30 parasites . This increased expression of PGMPUT would increase conversion of glucose-1-phosphate to glucose-6-phosphate , an important metabolic intermediate in glycolysis and pentose phosphate pathway , leading to generation of ATP molecules to meet energy demands during oxidative stress posed by miltefosine exposure . Lipases play important role in acquisition of host resources for energy metabolism and building blocks for the synthesis of complex parasite lipids important for membrane remodelling [35] . We observed increased expression of gene encoding lipase precursor like protein ( LPP ) in LdRelapse and LdM30 parasites [12] . This could be another strategy adopted by miltefosine unresponsive L . donovani parasites to exploit lipid catabolism and use of free fatty acid as an alternate energy source during miltefosine pressure . Elevated level of amino acids help drug resistant parasite in surviving within parasitophorous vacuoles [36] . A recent study has shown downregulated expression of amino acid permease in MIL resistant L . donovani parasites [37] . We found downregulated expression of a gene coding for amino acid transporter in MIL unresponsive phenotype ( LdRelapse and LdM30 ) of L . donovani . A recent study highlighted the downregulated expression of mitochondrial HSP70 in MIL resistant L . donovani parasites [38] . The chaperones TCP20 responsible for protein folding showed downregulated expression in LdRelapse and in LdM30 parasites as reported earlier [12] . The development of miltefosine unresponsiveness in L . donovani parasites is a pleotropic phenomenon . The present study revealed that parasites isolated from the cases that relapsed exhibited high infectivity , increased metacyclogenesis , reduced drug accumulation and reconfigured metabolism to overcome the oxidative stress induced during MIL exposure , factors contributing to high relapse rate observed in MIL treated VL and PKDL patients . This study highlighted that the overall changes in parasitic factors investigated here are similar in clinical and laboratory adapted parasites , but possibly , different mechanisms are operative for adaptation in the field . | Increasing rate of relapse against miltefosine ( MIL ) and decline in its efficacy prompted us to study the parasitic factors associated with MIL unresponsiveness in clinical isolates of Leishmania donovani . Studies to explore the mechanism of MIL resistance in L . donovani are largely restricted to experimentally induced resistant parasites . In the present study , parasites isolated from MIL treated patients that relapsed ( LdRelapse ) were found to exhibit increased metacyclogenesis and infectivity to macrophages , decreased miltefosine accumulation and increased tolerance towards MIL induced oxidative stress in comparison to isolates from pretreatment cases ( LdPreTx ) . Reduction in drug accumulation and increase in intracellular thiol content as well as tolerance to MIL induced oxidative stress were the highest in experimentally induced MIL resistant parasites ( LdM30 ) . Both LdRelapse and LdM30 parasites showed differential expression of genes associated with oxidative stress , metabolic processes and transport activity in comparison with LdPreTx parasites . The present study revealed that the parasites isolated from the cases that relapsed exhibited high infectivity , increased metacyclogenesis , reduced drug accumulation and reconfigured metabolism to overcome the oxidative stress induced during MIL exposure . These factors may be contributing to the high relapse rate observed in MIL treated VL and PKDL patients . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"chemical",
"compounds",
"oxidative",
"stress",
"immunology",
"tropical",
"diseases",
"microbiology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"protozoan",
"life",
"cycles",
"organic",
"compounds",
"developmental",
"biology",
"protozoans",
"leishmania",
"neglected",
"tropical",
"diseases",
"promastigotes",
"infectious",
"diseases",
"white",
"blood",
"cells",
"zoonoses",
"animal",
"cells",
"gene",
"expression",
"life",
"cycles",
"chemistry",
"thiols",
"protozoan",
"infections",
"leishmania",
"donovani",
"cell",
"biology",
"organic",
"chemistry",
"leishmaniasis",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"protozoology",
"macrophages",
"physical",
"sciences",
"organisms"
] | 2017 | Increased miltefosine tolerance in clinical isolates of Leishmania donovani is associated with reduced drug accumulation, increased infectivity and resistance to oxidative stress |
Epilepsy and progressively worsening severe chronic headaches ( WSCH ) are the two most common clinical manifestations of neurocysticercosis , a form of cysticercosis . Most community-based studies in sub-Saharan Africa ( SSA ) use a two-step approach ( questionnaire and confirmation ) to estimate the prevalence of these neurological disorders and neurocysticercosis . Few validate the questionnaire in the field or account for the imperfect nature of the screening questionnaire and the fact that only those who screen positive have the opportunity to be confirmed . This study aims to obtain community-based validity estimates of a screening questionnaire , and to assess the impact of verification bias and misclassification error on prevalence estimates of epilepsy and WSCH . Baseline screening questionnaire followed by neurological examination data from a cluster randomized controlled trial collected between February 2011 and January 2012 were used . Bayesian latent-class models were applied to obtain verification bias adjusted validity estimates for the screening questionnaire . These models were also used to compare the adjusted prevalence estimates of epilepsy and WSCH to those directly obtained from the data ( i . e . unadjusted prevalence estimates ) . Different priors were used and their corresponding posterior inference was compared for both WSCH and epilepsy . Screening data were available for 4768 individuals . For epilepsy , posterior estimates for the sensitivity varied with the priors used but remained robust for the specificity , with the highest estimates at 66 . 1% ( 95%BCI: 56 . 4%;75 . 3% ) for sensitivity and 88 . 9% ( 88 . 0%;89 . 8% ) for specificity . For WSCH , the sensitivity and specificity estimates remained robust , with the highest at 59 . 6% ( 49 . 7%;69 . 1% ) and 88 . 6% ( 87 . 6%;89 . 6% ) , respectively . The unadjusted prevalence estimates were consistently lower than the adjusted prevalence estimates for both epilepsy and WSCH . This study demonstrates that in some settings , the prevalence of epilepsy and WSCH can be considerably underestimated when using the two-step approach . We provide an analytic solution to obtain more valid prevalence estimates of these neurological disorders , although more community-based validity studies are needed to reduce the uncertainty of the estimates . Valid estimates of these two neurological disorders are essential to obtain accurate burden values for neglected tropical diseases such as neurocysticercosis that manifest as epilepsy or WSCH . ClinicalTrials . gov NCT03095339 .
The Global Burden of Disease Study estimates epilepsy and migraines to be among the 30 leading causes of years of life lost due to disability [1] . The prevalence of epilepsy and headaches varies globally , likely due to different risk factors across populations or epidemiological biases [2 , 3] . A meta-analysis estimated the median prevalence of lifetime epilepsy in developing countries at 15 . 4‰ in rural and 10 . 3‰ in urban areas , higher than the estimated 5 . 8‰ in developed countries [4] . The estimated prevalence of lifetime epilepsy shows substantial variation within sub-Saharan Africa ( SSA ) [4–8] , ranging from 7 . 3‰ to 29 . 5‰ . In contrast , the prevalence of migraine in adults is reported to be higher in developed countries than that in developing countries , with an estimate of 15% in Europe and 5% in Africa [3] . The limited data collected on headaches in SSA suggest that the one-year period prevalence ranges from 3 . 0% to 5 . 4% for migraines , and 1 . 7% to 7 . 0% for tension-type headaches [9–11] . Reasons for the opposite trends in the frequency of epilepsy and headaches may be due to the distribution of socio-economical , cultural , and infectious risk factors and genetic susceptibilities , as well as important methodological differences between studies [1 , 3 , 12] . In particular , whereas prevalence data on epilepsy and headaches often originate from national surveys or nation-wide electronic health records in high income countries , it is not the case in SSA , where data originate from research projects conducted in a small number of communities . The way these neurological disorders are measured in different settings could result in important misclassification error bias which , in turn , could explain some of this variation in estimates . Rural community-based studies in SSA often use a two-step approach to identify neurological disorder cases for prevalence estimates [13 , 14] . In step one , a screening test , often in the form of a questionnaire , is used to identify positive cases for the neurological disorder ( s ) of interest in the study population . The participants in this step can be randomly selected or identified via door-to-door visits . In step two , a physician or sometimes a neurologist confirms the given neurological disorder ( s ) through a medical examination , often only among those screened positive [15–19] . Diagnosis through electroencephalography is often not possible in rural community-based settings due to limited resources [13] , and therefore , the physician/neurologist’s diagnostic is often considered as the gold standard . The two-step approach is frequently used for studies in low-resource settings due to its convenience and cost-effectiveness . Despite its frequent use , the prevalence estimates from the two-step approach can be seriously biased from failure to account for the imperfect validity of the tests , referred to as misclassification error hereinafter , employed either at step one or step two , or both . The validity of a test is typically assessed by its sensitivity and specificity , and the correct sensitivity and specificity must be used to obtain unbiased prevalence estimates . However , for questionnaires used as the screening tool for epilepsy and headaches , especially those used in rural areas of SSA , little is known about their validity . In most studies , standardized screening questionnaires are used , but their validity is not determined in the study population [8–10 , 19] . To our knowledge , only two studies have reported validity estimates of epilepsy screening questionnaires in the general population; however , neither addresses potential verification bias in their validity estimates [20 , 21] . Verification bias occurs when the participants from step one have different probabilities of being selected for step two [22] . For example , individuals screened positive at step one have a much higher chance of being confirmed by a physician or neurologist , compared to those screened negative , who are rarely , if at all , examined at step two . As a result , the selected individuals at step two are not representative of the entire study population , which may lead to biased prevalence estimates . These biases can have important consequences in evaluating the global burden of neglected tropical diseases . For example , epilepsy and progressively worsening severe chronic headaches ( WSCH ) are the most frequently observed clinical signs of neurocysticercosis ( NCC ) , a preventable infection with the eggs of Taenia solium [23] , which is present in communities with poor sanitation and free roaming pigs [24 , 25] . Most epidemiological studies evaluating the prevalence of NCC in communities will first identify people with epilepsy , and rarely with headaches , and invite them to obtain brain imaging to diagnose NCC . The proportion of NCC among people with epilepsy ( or rarely headaches ) is then used to estimate the prevalence of epilepsy-associated ( or headaches-associated ) NCC in the study population . Such estimates may then be combined to the prevalence of epilepsy or headaches in the population to estimate Disability Adjusted Life Years ( DALYs ) associated with NCC [26–28] . The global burden of disease initiative used this approach to estimate DALYs associated with cysticercosis [29] . Therefore , to obtain accurate estimates of the global burden of NCC , or of any neglected tropical disease causing neurological signs , we first need valid prevalence estimates of epilepsy and WSCH . From there , we may obtain more reliable assessments of the relative burden of NCC , or other neglected tropical diseases manifesting as epilepsy or headaches , compared to other infections or chronic diseases . In this paper , we aimed to quantify the bias introduced in the prevalence estimates when failing to account for the verification bias and the imperfect validity of the screening tool using data collected from 60 villages in Burkina Faso . We also investigated the validity of screening questionnaire by estimating the sensitivity and specificity to detect epilepsy and WSCH .
Baseline cross-sectional data collected from February 2011 to January 2012 for a cluster randomized controlled trial were used . The aim of the parent study was to estimate the effectiveness of a community-based educational intervention to reduce the cumulative incidence of human and porcine cysticercosis in 60 villages of Burkina Faso [30] . From 70 to 80 individuals aged 5 years or above were sampled in each village using a cluster random sampling approach described elsewhere [30] . The University of Oklahoma Health Sciences Center Institutional Review Board and the Centre MURAZ ethical review panel ( Burkina Faso ) approved this study . The field staff read the written consent forms to participants and answered all their questions . Consent forms were signed , marked with a cross or a fingerprint by those who were unable to write . All consent forms were signed by a witness . Parents of individuals 5 to 16 years of age gave consent for their children . Individuals aged 10 to 15 years were invited to give their assent . All participants were given a bar of soap as incentive . The parent study was registered through clinicaltrials . gov ( NCT03095339 ) . In step one of the two-step approach , each participant was screened for epilepsy and WSCH using a screening questionnaire ( S1 Questionnaire ) . Questions related to epilepsy were based on the International League Against Epilepsy screening of epilepsy questionnaire developed by Preux et al . [31] , and was previously used in three villages in the same study area [32] . Questions related to headaches were designed to capture NCC-related headaches [33] . In step two , all individuals screened positive for either epilepsy or WSCH from the screening questionnaire were invited to be examined by a study physician . In addition , 231 screened-negative individuals were randomly selected to be examined by the physician . The medical examination results were collected on a medical examination questionnaire ( S2 Questionnaire ) . Two medical examination rounds took place with the second round aimed at capturing individuals who were absent during the first round and to examine patients from the more remote province of Nayala . The physicians discussed any uncertain diagnosis with the neurologist on the phone at the time of the medical examination . At the end of the study , the neurologist reviewed all diagnoses and his final diagnosis was considered as the gold standard . The diagnostic result that confirmed whether the individual had epilepsy and/or WSCH was used to assess the validity of the screening questionnaire . Epilepsy was defined as having more than one seizure of central nervous system origin without apparent cause [34] . Individuals not meeting the epilepsy case definition were considered as epilepsy-free ( i . e . , screened negative for epilepsy ) . Six individuals diagnosed with single epileptic seizures were excluded from all analyses . WSCH was defined as having symptoms arising more than weekly for two weeks or more , with each episode lasting at least 3 hours , and progressively worsening in severity with time . Headaches had to be severe enough to require analgesic or to prohibit working , playing , attending school , or partaking in daily activities [35] . Individuals not meeting the WSCH definition were considered as WSCH-free ( i . e . , screened negative for WSCH ) . The statistical analyses aimed at assessing the bias from failing to correct for both verification bias and misclassification error , when estimating the prevalence of epilepsy and WSCH in the study population . To assess the degree of the bias , we first calculated the unadjusted estimates , where no correction was made for verification bias and misclassification error . The unadjusted estimates were obtained by running a Bayesian binomial model , where the number of confirmed cases of either epilepsy or WSCH was assumed to result from a binomial distribution with a probability corresponding to the unadjusted prevalence and the number of individuals screened . The prior choice for the prevalence parameter is discussed below . The adjusted prevalence estimate was obtained by running a Bayesian latent-class model [22] . In this model , the probabilities that participants were selected for the confirmation test at step two were specified by a set of conditional distributions . Participants with different selection probabilities had different conditional probability distributions . This way , the selection probabilities were correctly accounted for , eliminating verification bias . To obtain the specificity and sensitivity estimates of the screening questionnaire while correcting for verification bias , we provided prior information on the model parameters of sensitivity and specificity . Different priors , including both informative and non-informative ( i . e . , vague ) priors , were investigated in modeling either epilepsy or WSCH . Specifically , for epilepsy , one set of informative priors based on the sensitivity and specificity estimates for similar epilepsy screening questionnaires from two previous community-based studies [25 , 26] were used . In these studies , sensitivity and specificity were estimated as 92 . 9% and 79 . 3% , and 99 . 6% and 72 . 4% , respectively . To allow some variability , the prior sensitivity and specificity values in our analysis were assumed to follow beta distributions with mean based on these validity estimates and a standard deviation of 0 . 05 . This led to a Beta ( 54 . 8 , 17 . 5 ) prior for sensitivity , and a Beta ( 12 . 9 , 0 . 5 ) for specificity . We also considered the vague priors of Unif ( 0 . 5 , 1 ) for both parameters . Since epilepsy was highly stigmatized in SSA [6] and some forms of partial epilepsy may be difficult to identify , we also used a vague prior of Unif ( 0 . 3 , 1 ) for the sensitivity of epilepsy screening . Due to the lack of validity studies for WSCH screening , the same informative priors as epilepsy were adopted for WSCH . We also ran the model with vague priors of Unif ( 0 . 5 , 1 ) for both the specificity and sensitivity , and results were compared with those obtained using informative priors . When modeling each epilepsy and WSCH , we used a vague Unif ( 0 , 0 . 3 ) for the unadjusted prevalence estimate . For the adjusted prevalence parameter , a vague prior of N ( 0 , 10 ) was used ( on logit scale ) . We used WinBUGS [36] for all the Bayesian analyses and reported the posterior mean and 95% credible intervals . The bias was then evaluated by the ratio of the adjusted to the unadjusted estimates , with the Bayesian credible interval for the ratio excluding 1 indicating the existence of bias . We examined three scenarios reflecting different screening strategies often used in community-based studies conducted in low-resource areas . In Scenario 1 , we estimated the unadjusted prevalence using the screening information for only one neurological disorder instead of for epilepsy and WSCH , to reflect a common situation in the existing literature where only one neurological disorder was studied at a time instead of both disorders simultaneously . In Scenario 2 , we estimated the unadjusted estimate using the information for participants that were only examined in the first medical round . This scenario was considered to reflect situations where there were insufficient personnel and monetary resources to find people absent from the village during the initial visit . Scenarios 1 and 2 are fairly common in field studies conducted in resource-poor contexts . In Scenario 3 , we only used the screening information for the unadjusted prevalence estimate , reflecting a situation where the validity of the screening questionnaire cannot be assessed . We calculated the adjusted estimates resulting from each scenario , and evaluated the bias estimating the ratio of the verification bias and misclassification error-adjusted to the unadjusted estimates . In this exercise , the impact of verification bias and misclassification error on community-based estimates of NCC-associated epilepsy and WSCH prevalence and number of cases was evaluated . To obtain the estimated prevalence of NCC-associated epilepsy and WSCH for our study population , the adjusted and unadjusted estimates of epilepsy and WSCH prevalence were multiplied by the proportion of NCC among people with epilepsy reported in a meta-analysis [37] and the proportion of NCC among people with headaches reported in a case-control study [38] . We then investigated the difference between the adjusted and unadjusted prevalence estimates of NCC-associated epilepsy and WSCH . To obtain the difference between the number of NCC-associated epilepsy and WSCH cases in our study population , the estimated adjusted and unadjusted prevalence estimates were multiplied by the study population size . We assumed that the mean number of NCC-associated cases of epilepsy and WSCH would follow beta distributions with parameters based on the estimated means in the published studies ( 29% for epilepsy and 4 . 7% for headaches ) [37 , 38] and a standard deviation of 2 . 75% . All estimates were run using the set of informative priors described earlier . The adjusted prevalence and number of cases of NCC-associated epilepsy and WSCH were also estimated under the three screening strategies described above .
A total of 4794 individuals were sampled at baseline , including the analytical sample of 4768 with complete screening data . Of these , 669 ( 14 . 0% ) screened positive for epilepsy ( 7 . 1% ) , WSCH ( 2 . 8% ) or both ( 4 . 2% ) ( Table 1 ) . A physician examined 609 ( 91 . 0% ) screened-positive and 231 ( 5 . 6% ) screened-negative individuals . The higher proportion of screened-positive examined by a physician compared to those screened-negative showed evidence for the need to adjust for potential verification bias ( Fig 1 ) . Of those examined in step two , 748 , 57 and 35 were seen at the first , second and both medical rounds , respectively . Perfect agreement was observed for those examined at both rounds . Table 1 describes the characteristics of the analytical sample , and the screening and medical examination results . The majority of participants were either a farmer or housewife and a high proportion did not complete primary school . Table 2 provides the posterior estimates of the unadjusted prevalence and the corresponding adjusted prevalence with the associated sensitivity and specificity estimates , using the different sets of priors . The epilepsy screening questionnaire showed posterior sensitivity and specificity estimates of moderate variation with different priors used , while the specificity estimates remained robust . The posterior sensitivity estimate was 44 . 7% ( 95%BCI: 33 . 0%;60 . 0% ) with the most vague prior and increased to 66 . 1% ( 95%BCI: 56 . 4% , 75 . 3% ) with the informative priors based on previous literature . Posterior sensitivity and specificity estimates of the WSCH screening questionnaire were less affected by the priors . Despite the variation in the estimated sensitivity and specificity , the unadjusted prevalence estimates were consistently lower than the adjusted ones for both epilepsy and WSCH , as indicated by the posterior bias distribution lying under the value of 1 . Somewhat less bias was observed using the informative priors . Bias ( i . e . , where the unadjusted estimate was smaller than the adjusted estimate ) was introduced for both epilepsy and WSCH ( Figs 2 and 3 ) under the two most common screening strategies found in the literature , namely Scenarios 1 ( i . e . when we assumed that the study would only screen for one neurological disorder ) and 2 ( i . e . when resources would not allow for returning to communities to examine those absent during a first visit ) . The magnitude of the bias was similar to that observed in the main analyses for epilepsy , but more marked for WSCH . For Scenario 3 ( i . e . when only screening results were used ) , an important positive bias was present where the unadjusted prevalence estimate of 11 . 2% ( 95%BCI: 10 . 4%; 12 . 2% ) was considerably larger than the adjusted prevalence for epilepsy when more informative priors were used . For WSCH , Scenario 3 resulted in a negligible positive bias . The magnitude of the bias did not vary extensively by the priors used for epilepsy and WSCH , except for epilepsy in Scenario 3 . Using the study data , the differences between the adjusted and unadjusted prevalence and number of NCC-associated epilepsy were 0 . 6% ( 95%BCI: 0 . 3%; 1 . 0% ) and 29 cases ( 95%BCI: 13; 49 ) , respectively .
This was the first study to estimate the sensitivity and specificity of a screening questionnaire for epilepsy and WSCH in the community while adjusting for both verification and misclassification error bias . This was also the first study to quantify the bias from failing to account for verification and misclassification error bias when estimating the prevalence of epilepsy and WSCH in a community-based study . Our sampling strategy was not designed to provide population prevalence estimates of epilepsy and WSCH in Burkina Faso as a whole , in the three study provinces , or even the villages selected for the parent study . Indeed , our sampling strategy , which favored concessions with pigs , was likely to result in higher estimates of prevalences than what would have been observed if a simple random sampling strategy had been adopted . Nonetheless , we provided below prevalence estimates from other community-based studies conducted in resource-poor settings using the two-step approach . In our study sample , the unadjusted prevalence of epilepsy was higher than that of 1 . 6% ( 95%CI: 1 . 2%;2 . 0% ) , 1 . 1% ( 95%CI: 0 . 9%; 1 . 4% ) and 0 . 5% ( 95%CI: 0 . 2%;0 . 8% ) found in Benin , Tanzania and Nigeria , respectively [7 , 15 , 39] . Our estimate was also higher than that of 0 . 6% ( 95%CI: 0 . 5%;0 . 7% ) found in Cambodia where a screening questionnaire similar to ours was used [40] . Similarly , our unadjusted estimate was higher than the first study conducted in rural Burkina Faso , which estimated an epilepsy prevalence of 1 . 1% in 18 villages [41] . Our prevalence estimate was most similar to a recent Burkinabé study , conducted by the same research group , which estimated a lifetime epilepsy prevalence of 4 . 5% ( 95%CI: 3 . 3%;6 . 0% ) in three villages purposely sampled to have a high prevalence of epilepsy [32] . This supported the suspicion that our study sample might represent people at higher risk of epilepsy than the general population . Our unadjusted estimate of lifetime WSCH prevalence was similar to the lifetime migraine prevalence estimate of 3 . 3% ( 95%CI: 2 . 4%;4 . 6% ) in Benin [42] , and lower than the lifetime migraine prevalence estimate of 5 . 3% ( 95%CI: 5 . 0%; 5 . 6% ) in Nigeria [43] . Lifetime prevalence estimates of tension-type headaches in SSA were unavailable for comparison . Compared to published one-year prevalence estimates of tension-type headaches , ours was higher than the 1 . 7% ( 95%CI: 1 . 5%; 1 . 9% ) prevalence in Ethiopia among adults 20 years and older , and lower than the 7 . 0% ( 95%CI: 6 . 5%;7 . 6% ) prevalence in Tanzania among all ages [9 , 11] . This suggested that our sampling strategy might have not influenced the estimated frequency of WSCH as much as it did for epilepsy . The posterior estimates of the specificity for the neurological screening questionnaire were similar for epilepsy and WSCH , and were consistently around 88% . These estimates were slightly lower than those previously reported [20 , 21] , possibly because our study population had a higher proportion of false negatives compared to the previously published validity studies . Study participants in our study may also have been less likely to report their symptoms compared to those in Ecuador and Bolivia , perhaps due to the stigmatizing effects of epilepsy in SSA [6] . The posterior estimates of sensitivity varied depending on the priors used , particularly for epilepsy . The sensitivity posterior median was higher when prior knowledge [20 , 21] information was used and lower for vague priors; although the credible intervals overlapped . A similar observation where the sensitivity parameter was more affected to the prior choice was also found in a study assessing the validity measures of a screening test for human papillomavirus [44] . Our sensitivity estimates were lower and specificity estimates higher for epilepsy than that found in validation studies conducted in clinical settings with estimates of sensitivity between 91% and 100% and specificity between 51% and 85% [19 , 20 , 45] . Placencia et al . found lower sensitivity and similar specificity estimates when the same screening questionnaire was used in the community as compared to the clinic [20] . Such discrepancies may be due to spectrum bias where populations in clinical settings have more severe disease and acknowledge their symptoms more , thereby increasing the test sensitivity . The observed estimates from clinical settings may also result in part from verification bias , which typically leads to an overestimation of the sensitivity and underestimation of the specificity [22] . Our sensitivity estimates for epilepsy were similar to the two other community-based studies with sensitivities of 72 . 4% ( 95% CI: 52 . 8–87 . 3 ) and 79 . 3% [20 , 21] . To our knowledge , prevalence studies using the two-step approach for WSCH have not reported validity estimates . Even in a situation where 90% of those screened positive were examined by a physician , significant biases were observed . The unadjusted prevalence estimates were consistently lower than the adjusted prevalence estimates regardless of the priors used . However , the adjusted prevalence estimates were highly dependent on priors , particularly for epilepsy , which introduced considerable uncertainty . Such uncertainty could be reduced by conducting more community studies assessing and reporting validity estimates of screening . The validity estimates were expected to vary from one community to the next , due to how epilepsy and WSCH were reported by patients and to the way interviewers were trained to use the questionnaire . This would also result in varied bias estimates . Therefore , the reported sensitivity and specificity estimates of the screening questionnaire may not be applicable to other communities . We chose the Bayesian framework [46 , 47] to simultaneously correct for verification bias and misclassification error in the two-step approach for two reasons . First , verification bias can be treated as a missing data problem , and therefore be corrected in a straightforward manner . Second , verification bias and misclassification error can be addressed simultaneously by including an additional level in the Bayesian model estimating the specificity and sensitivity of the invalid test ( s ) . The capture-recapture method is an alternative approach for obtaining corrected prevalence estimates [13] . This method combines multiple sources of information independently , such as medical records and non-medical interviews with community members , with the two-step approach . This method yielded higher prevalence estimates than the two-step approach alone in two studies in Benin [7 , 42] . Our bias estimation method using their data yielded similar results ( between 0 . 5 and 0 . 3 ) . However , the capture-recapture method has multiple difficult-to-meet assumptions: closed population , statistical independence between sources , identical case definitions across multiple sources and requires more personnel resources [7 , 42] . Our study was less resource-intensive and provided an alternative to the capture-recapture method . When screening strategies commonly used in community-based studies were explored , the unadjusted prevalence was lower than the adjusted prevalence in the two most frequently encountered scenarios , and the bias estimates were similar to that observed under the screening strategy used in the main analysis of this study . In our study , we had resources to capture those missed through the first medical round and we screened for two neurological outcomes , which resulted in more individuals being confirmed in step two compared to most community-based studies . Despite this more complete examination of individuals screened positive , the level of bias was similar to that estimated for screening strategies commonly used in community-based studies . In the last scenario explored , the prevalence relied entirely on the screening questionnaire , which led to a large number of false positive cases of epilepsy . This was especially true because the prevalence of epilepsy was relatively low , and hence , there were relatively more people without epilepsy who were false positives than people with epilepsy who were false negatives , even if the screening test’s specificity was much better than its sensitivity . As opposed to the other scenarios , failure to use physician confirmation led to an overestimate of the adjusted prevalence of epilepsy . This scenario could occur when economical and personnel resources are too limited for physician confirmation . For WSCH , the estimated negative bias ( i . e . when the unadjusted estimate was lower than the adjusted estimate ) was more marked in the two scenarios most often encountered in community-based studies than when using the screening strategy in the main analysis of this study . This suggested that by only screening for WSCH with or without two rounds , more WSCH cases were missed . There is rising evidence that seizures might increase the risk of headaches [48 , 49] . Hence , it is possible that screening for both outcomes improved the detection of WSCH through the neurological examination of people screening positive for epilepsy . The opposite ( i . e . the screening of people with WSCH increases the detection of epilepsy ) might not be as marked in our study because a lot less participants screened positive for WSCH than for epilepsy or both epilepsy and WSCH . In the last scenario where the prevalence estimate relied solely on the screening questionnaire , we observed negligible positive bias for WSCH . This finding illustrated that screening for both epilepsy and WSCH led to a less biased estimate of the prevalence compared to the more commonly used approach of only screening for one neurological condition . This was not observed for the last scenario for epilepsy because screening for WSCH along with epilepsy might not increase the detection of epilepsy . When we examined the impact of verification and misclassification bias on the estimation of the prevalence and frequency of NCC-associated epilepsy and WSCH , we found that the prevalence and number of NCC were underestimated when using two-step approach regardless of the scenario . In the scenario without confirmation by a physician or neurologist , we found that the number of NCC-associated epilepsy cases were over-estimated , while we did not observe a difference between the unadjusted and adjusted prevalence and number of NCC-associated WSCH cases . These results could have important consequences in the estimation of the monetary and non-monetary burden of NCC locally and globally . For example , the estimated global DALYs of NCC-associated epilepsy were estimated to be 2 , 788 , 426 ( 95% Uncertainty Estimates ( UI ) : 2 , 137 , 613–3 , 606 , 582 ) by the Foodborne Epidemiology Research Group in 2010 [50] and to be 468 , 100 ( 95% UI: 322 , 900–625 , 800 ) in 2016 by the Global Burden of Disease 2016 Collaborative [51] . Moreover , the 2015 DALYs for migraine were estimated to be 32 , 899 , 000 ( 95%UI: 20 , 295 , 000–48 , 945 , 000 ) [52] . If these estimates used underlying NCC-associated epilepsy and headaches frequencies which were underestimated to a similar level as in our study , it could have important consequences on how NCC would rank among all diseases locally and globally and on policy making . Indeed , if the burden of NCC were higher than currently believed , more resources should be allocated to control it . Our study has several limitations . First , the physicians examined a small proportion of individuals screened negative . Increasing this proportion would have reduced the variance of our posterior sensitivity estimates . Second , our study used two medical rounds with different physicians . However , perfect agreement between the two physicians was observed and all diagnoses were reviewed and confirmed by the neurologist , minimizing the possibility for bias . Third , we could not conclude that our prevalence estimates for the study population were reflective of the prevalence for the villages due to the sampling scheme . Since the aim of our study was to quantify the bias that rose from failure to account for misclassification error and verification biases by comparing the unadjusted estimates to the adjusted estimates , we believe our findings were still of importance . Our results suggest that the burden of epilepsy and WSCH in low-resource settings might be much higher than previously reported . Future studies should consider using the statistical models presented here to account for the imperfect nature of screening questionnaires . Bias-adjusted prevalence estimates of these two neurological disorders will improve our understanding of the burden of these conditions and help identify where cysticercosis may be present . More valid prevalence estimates will allow for the development of targeted cysticercosis control programs in those communities . | Epilepsy and progressively worsening severe chronic headaches are the two most common clinical manifestations of neurocysticercosis , a form of cysticercosis . To understand where the prevalence of these neurological disorders is highest for targeted infection control , valid prevalence estimates are needed . Most neuroepidemiological studies conducted in low-resource settings use a two-step approach to identify cases ( screening questionnaire followed by physician examination among screened positives ) to obtain prevalence and burden estimates . We found that this most commonly-used two-step approach in community-based studies leads to an underestimation of the prevalence in our study . Our paper provides an analytic solution to reduce errors in estimating the prevalence of neurological disorders in community-based studies when using the two-step approach . Our proposed approach provides more valid estimates of the prevalence of these neurological disorders and could be used to better reflect the consequences that neglected tropical diseases manifesting as epilepsy or headaches have on people’s health and disabilities . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"physicians",
"medical",
"doctors",
"pathology",
"and",
"laboratory",
"medicine",
"neurocysticercosis",
"medical",
"personnel",
"tropical",
"diseases",
"parasitic",
"diseases",
"migraine",
"health",
"care",
"headaches",
"research",
"design",
"health",
"care",
"providers",
"signs",
"and",
"symptoms",
"neglected",
"tropical",
"diseases",
"questionnaires",
"research",
"and",
"analysis",
"methods",
"epilepsy",
"people",
"and",
"places",
"helminth",
"infections",
"professions",
"diagnostic",
"medicine",
"survey",
"research",
"neurology",
"cysticercosis",
"population",
"groupings"
] | 2019 | The impact of imperfect screening tools on measuring the prevalence of epilepsy and headaches in Burkina Faso |
DNA sequences capable of adopting non-canonical secondary structures have been associated with gross-chromosomal rearrangements in humans and model organisms . Previously , we have shown that long inverted repeats that form hairpin and cruciform structures and triplex-forming GAA/TTC repeats induce the formation of double-strand breaks which trigger genome instability in yeast . In this study , we demonstrate that breakage at both inverted repeats and GAA/TTC repeats is augmented by defects in DNA replication . Increased fragility is associated with increased mutation levels in the reporter genes located as far as 8 kb from both sides of the repeats . The increase in mutations was dependent on the presence of inverted or GAA/TTC repeats and activity of the translesion polymerase Polζ . Mutagenesis induced by inverted repeats also required Sae2 which opens hairpin-capped breaks and initiates end resection . The amount of breakage at the repeats is an important determinant of mutations as a perfect palindromic sequence with inherently increased fragility was also found to elevate mutation rates even in replication-proficient strains . We hypothesize that the underlying mechanism for mutagenesis induced by fragile motifs involves the formation of long single-stranded regions in the broken chromosome , invasion of the undamaged sister chromatid for repair , and faulty DNA synthesis employing Polζ . These data demonstrate that repeat-mediated breaks pose a dual threat to eukaryotic genome integrity by inducing chromosomal aberrations as well as mutations in flanking genes .
Chromosomal instability and mutagenesis are two fundamental processes that alter prokaryotic and eukaryotic genomes . The deleterious consequences of excessive DNA perturbations are hereditary diseases and cancer in humans ( reviewed in [1] , [2] , [3] ) . At the same time , a fine balance between acquiring genetic changes and restoring original DNA content is paramount for organismal development , adaptation , polymorphism and evolution ( for example [4] , [5] , [6] ) . Double-strand breaks ( DSBs ) in DNA are a driving force for both chromosomal instability and accumulation of mutations . DSBs are a well-established source of a variety of chromosomal aberrations including translocations and copy number variations [7] , [8] . It has also become evident from studies in bacteria and yeast that DSB formation and repair are associated with an increased level of mutations , even during homologous recombination which was considered to be an error-free process . In E . coli , the role of DSB formation in the induction of mutagenesis was first inferred based on the requirement of RecA and RecBCD for the occurrence of adaptive mutations in the LacZ gene [9] and was later directly demonstrated by using I-SceI endonuclease-induced breaks [10] . In yeast , elevated levels of base substitutions and frame shift mutations were shown to be due to DSB repair in meiosis [11] and as a result of induction of DSBs in mitotically-dividing cells as shown in gene conversion ( GC ) [12] , [13] , break-induced replication ( BIR ) [14] and single-strand annealing ( SSA ) assays [15] . The proposed mechanism for break-induced mutagenesis , surmised from these studies , involves the formation of long regions of single-stranded DNA ( ssDNA ) as a result of DSB end resection . Mutations arise during error-prone synthesis either across the damaged ssDNA template or during synthesis following invasion into the undamaged donor strand . There are two lines of evidence supporting this mechanism . First , Yang et al . [15] have shown in yeast , that single-stranded DNA is drastically more prone to the accumulation of mutations with and without treatment with DNA damaging agents than double-stranded DNA . Second , in several studies , mutagenesis was shown to be fully or partially dependent on highly inaccurate translesion polymerases ( TLS ) . The bacterial TLS polymerase , DinB is responsible for 85% of the mutations triggered by DSB repair during adaptive mutagenesis [16] . In yeast , depending on the assay and nature of mutations , DSB-induced mutagenesis is either completely ( SSA [15] ) , partially ( GC next to the DSB site and BIR [12] , [13] , [14] ) or not ( classical GC assay [17] ) attributed to the activity of the error-prone TLS polymerase , Polζ . Problems encountered by DNA replication machinery are a major source of spontaneous chromosomal breakage in eukaryotes , estimated to be approximately 10 DSBs per cell cycle in human cells ( reviewed in [18] ) . Certain chromosomal regions , the fragile sites , often containing secondary structure-forming repeats , are susceptible to breakage especially under conditions of replication stress [19] . The mutagenic potential of replication-associated breaks has not been studied in detail . It is also unknown what the level of breaks during replication should be for mutagenesis to be manifested . The latter is important considering the fact that in previous studies mutagenesis was detected under conditions of extremely high level of DSBs , reaching up to 100% as seen in the case of site-specific endonucleases . Whether fragile motifs on their own or under conditions of replication stress could be a potent endogenous source of mutations remains to be established . In this study , we investigate the mutagenic potential of two sequence motifs , inverted repeats and GAA/TTC tracts , which are natural chromosomal fragile sites [20] , [21] under conditions of unperturbed and compromised replication . Long inverted repeats can adopt non-B DNA secondary structures such as hairpins and cruciforms owing to their internal symmetry [22] . They are a potent source of genome rearrangements in both prokaryotes and eukaryotes including humans [23]–[26] . We have previously demonstrated that in yeast a 320 bp Alu-quasi-palindrome triggers gross chromosomal rearrangements by inducing special type of DSBs that have hairpin-capped termini [21] , [26] . The hairpin ends are a substrate for opening and processing by Sae2 and the Mre11/Rad50/Xrs2 ( MRX ) complex . In Δmre11 , Δrad50 , Δxrs2 , or Δsae2 mutants , the resection of broken ends is completely blocked , giving rise to inverted dimers . GAA/TTC tracts adopt another kind of non-canonical DNA structure , namely , H-DNA or triplex DNA ( reviewed in [27] ) . The triplex secondary structure is a driving force for the expansions of GAA tracts , a phenomenon responsible for Friedreich's ataxia in humans [28] . Triplex-adopting sequences , including GAA/TTC repeats , are also responsible for breakage and induction of recombination and rearrangements in bacteria , yeast and humans [20] , [29]–[34] . Using yeast as an experimental system , we previously demonstrated that triplex structure-imposed replication problems can contribute to breakage at long GAA/TTC tracts [20] . At the same time , GAA-mediated breaks can occur in non-dividing cells where transcription is an important determinant of DSBs [35] , [36] . H-DNA forming sequences are mutagenic in yeast and mammalian systems [34] , [37]–[39] , albeit , direct evidence that repeat-induced fragility is the reason for mutagenesis in the vicinity of the repeats remains to be found . In this work , we demonstrate that increased break formation at the location of inverted repeats causes mutagenesis at distances up to 8 kb away from the DSB site . The accumulation of mutations requires the Sae2 protein , indicating that resection and generation of long ssDNA is a critical parameter for this phenomenon . We have found that error–prone synthesis involving the translesion polymerase Polζ during repair is primarily responsible for the observed mutagenesis . We also show that in replication-deficient strains the triplex-adopting GAA/TTC repeats are associated with hypermutability at distant loci , suggesting that a similar mechanism of mutagenesis can operate at repeat-associated chromosomal break sites under conditions of replication stress . These data demonstrate that secondary structure-mediated breaks pose a dual threat to eukaryotic genome integrity by inducing chromosomal aberrations and mutations extending to distant chromosomal sites . It is conceivable that the mechanisms of DSB-induced mutagenesis uncovered in this study are also relevant to human evolution , polymorphism and tumorigenesis .
The experimental system used to assess the mutagenic potential of fragile inverted and GAA/TTC repeats in this study is based on the GCR assay described in [20] , [26] . Briefly , the LYS2 gene containing the fragile motifs was inserted 43 kb from the telomere on the left arm of chromosome V in haploid yeast strains ( Figure 1 ) . There are no essential genes between the left telomere and the LYS2 gene . CAN1 is located 8 kb telomere-proximal to the LYS2 gene . The insertion of ADE2 between CAN1 and LYS2 allows for the differentiation between two types of events on media containing canavanine and low amounts of adenine . Breakage at the location of structure-forming repeats leads to the loss of the terminal 43 kb of the chromosomal arm containing both CAN1 and ADE2 , resulting in canavanine-resistant red-colored colonies ( CanRAde− ) . On the other hand , mutations in CAN1 are manifested as white-colored canavanine-resistant colonies ( CanRAde+ ) ( Figure 1 ) . The correlation between colony color and the requirement of adenine for growth was verified by replica plating the CanR colonies to media lacking adenine . The three fragile motifs inserted into LYS2 were 100% homologous inverted Alu repeats , 320 bp each with a 12 bp spacer ( Alu-IRs ) ; 100% homologous IS50 palindromic repeats ( IS50-PAL ) , 1 . 3 kb each; and 230 repeats of GAA/TTC in the orientation wherein the GAA sequence is the template for the lagging strand synthesis . To estimate how far mutagenesis can extend from the break site , URA3 was inserted into chromosome V telomere-proximal ( TP ) 0 . 6 kb and 30 kb away from the repeats , and telomere-distal ( TD ) 0 . 4 kb , 8 kb and 30 kb away from the repeats ( Figure 1 ) . Mutations in URA3 were measured on 5-fluoroorotic acid-containing media lacking adenine ( 5-FOAR Ade+ ) , allowing us to preferably select these events in contrast to GCRs that give rise to 5-FOAR Ade− colonies . Mutation levels in the CAN1 locus in wild-type strains with inverted Alu-quasipalindrome are not different from strains that lack the sequence motif . We wanted to determine whether addition of replication stress will enhance the fragility potential of these repeats and increase mutagenesis . In a screen for mutants that exhibit an increased level of hairpin-capped DSBs we identified the pol3-P664L allele that affects the functions of replicative polymerase δ responsible for synthesis of the lagging strand [40] . The P664L mutation is located in the polymerase domain of Polδ [41] and the yeast strains carrying this mutant allele exhibit temperature-sensitive growth at 37°C ( data not shown ) . The rate of CAN1 region loss in strains containing Alu-quasipalindrome was 40-fold higher in pol3-P664L mutants than in wild-type ( Table 1 ) . Moreover , pol3-P664L strains with Alu-IRs exhibited elevated levels of mutagenesis in CAN1 loci located 8 kb away from the DSB site . Notably , the mutagenesis was completely dependent on the presence of fragile motifs , suggesting that the mutator phenotype is not a feature of the pol3 allele but rather is a consequence of increased breakage . It is important to note that in pol3-P664L strains without the Alu-quasi-palindrome , the relative rate of arm loss was nearly 3-fold higher than in wild-type strains . However , the fragility due to deficiency in Polδ is not high enough to induce mutagenesis . A similar increase in Alu-IR-dependent fragility and mutagenesis in CAN1 gene was observed in strains where the POL3 expression was under the control of a tetracycline-repressible promoter ( tetO7 ) [42] . Belli et al . , 1998 [42] showed that tetO7–driven expression of genes in the presence of the antibiotic leads to a reduction in protein levels in comparison to conditions when genes were expressed from their native promoters . Western blotting analysis of c-Myc-tagged Pol3 revealed that upon treatment of cells with doxycycline the protein level was indeed ∼10 fold decreased in comparison with the wild-type level ( Figure 2 ) . Hence , we refer to TET-POL3 as a mutant allele and all further tests were carried out in the presence of doxycycline ( see Material and Methods ) . We also replaced the native promoter of another replication gene , RFA2 , that encodes one of the subunits of the single-stranded DNA-binding protein participating in DNA replication and repair , with the tetO7 promoter . Upon downregulation with doxycycline , the expression of Rfa2 was ∼4-fold lower than the wild-type level ( Figure 2 ) . Similar to the TET-POL3 strain , the TET-RFA2 strain exhibited increased levels of arm loss and mutagenesis ( Table 1 ) . It is important to note that neither the pol3-P664L strain nor the TET-POL3 and TET-RFA2 strains grown in the presence of doxycycline at chosen concentrations showed sensitivity to DNA damaging agents such as MMS and camptotechin , indicating that they are proficient in DNA repair ( Figure S1 ) . Overall , these data show that mutations at distant loci require the presence of fragile motifs and are dependent on the amount of replication-associated breaks . We addressed directly whether a fragile site can induce mutations at distant loci in replication-proficient strains . This experiment also helps to distinguish which is the key factor in mutagenesis , the level of breakage or repair of broken molecules by faulty replication proteins . The 1 . 3 kb long IS50 palindrome was found to induce mutagenesis in CAN1 when replication was unimpaired ( 3-fold ) . The increased length of the interacting arms and the lack of a spacer between them likely create a problem even for intact replication machinery and render this motif highly fragile with a 14-fold increase in GCR rates as compared to the Alu-IR strain ( Table 2 ) . Consistently , using Southern hybridization , we estimated the level of breakage at this palindrome to be 4 . 8% ( average of 4 . 6% , 5% and 4 . 9% ) which is ∼3-times higher than in strains carrying the Alu-quasi-palindrome ( 1 . 6% , average of 1 . 4% , 1 . 5% and 1 . 9% ) ( Figure 3 ) . Taking into account that a deficiency in Pol3 causes a 7-fold increase in Alu-IR-mediated breakage ( 11% , average of 11% , 10% and 11% ) and a 12-fold increase in mutagenesis , it is evident that the levels of DSB formation and not DSB repair by defective replication proteins are the important determinant of mutagenesis . Previously , we have shown that Alu-IRs induce DSBs that have hairpin-capped termini [21] . The resection and repair of these DSBs requires the hairpin-opening activity of Sae2 and the Mre11 nuclease [43] . To test if ssDNA generated as a result of 5′-3′DSB end resection is a critical requirement for repeat-induced mutagenesis , the SAE2 gene was disrupted in pol3-P664L , TET-POL3 and TET-RFA2 Alu-IR strains . The level of CAN1 mutagenesis in pol3-P664LΔsae2 , TET-POL3Δsae2 and TET-RFA2Δsae2 mutants was reduced to levels observed in strains without inverted Alus ( Table 1 ) , indicating that mutations are indeed a consequence of DSB resection . Similarly , in the strains carrying IS50 repeats , mutation rates declined upon deletion of SAE2 ( Table 2 ) . To determine to what distance the DSB-associated mutagenesis can spread on either side of the fragile site , we inserted the URA3 reporter 0 . 4 , 8 and 30 kb telomere-distal ( TD ) and 0 . 6 and 30 kb telomere-proximal ( TP ) to Alu-IRs in pol3-P664L strains ( Figure 1 ) . The average length of ssDNA generated via DSB end resection in yeast varies from 2 kb to 10 kb [44] . This predicts that mutations in URA3 situated past 10 kb should diminish . Consistently , although mutation rates at 0 . 4 kb , 0 . 6 kb and 8 kb were approximately the same ( 10–15-fold higher than in wild-type strain ) , at 30 kb the rate of ura3 mutations significantly decreased in TP and TD constructs ( Table 3 ) . The dependence of the efficiency of mutagenesis on the activity of Sae2 and the distance of the reporter from DSB site demonstrates that ssDNA is an intermediate for the occurrence of mutations . Holbeck and Strathern , [45] and Rattray et al . [12] showed that Polζ translesion synthesis activity is required for the generation of base substitutions in a reporter located 0 . 3 kb from the site of an HO-endonuclease-induced break . To assess if mutagenesis induced by fragile motifs depends on translesion synthesis , we disrupted the REV3 gene encoding the catalytic subunit of Polζ [46] in wild-type , pol3-P664L , TET-POL3 and TET-RFA2 strains with Alu-IRs ( Table 1 ) . REV3 disruption in replication-defective strains brought the mutation level in the CAN1 reporter to almost the level observed in the wild-type strain . In replication-proficient strains carrying Δrev3 , only a modest 2-fold decrease in CAN1 mutation rate was observed . Augmented mutation rates in the strains with IS50 repeats were also dependent on the activity of Polζ ( Table 2 ) . To gain further insight into the spectrum of mutations generated at distant loci as a result of DSB formation by inverted repeats , we sequenced 22–31 independent CanRAde+ isolates from wild-type , pol3-P664L and pol3-P664LΔrev3 strains , respectively . In the wild-type strain with Alu-IRs , 85% of the mutations were base substitutions and 15% were single base deletions ( Table 4 and Table S1 . ) . A similar mutation spectrum was also observed in other studies [47] , [48] . This correlates with the lack of increase of CAN1 mutagenesis in the wild-type Alu-IR strain ( Table 1 ) , indicating that the observed mutations in replication-proficient strains were a result of spontaneous mutagenesis rather than secondary structure-induced DSBs . In the pol3-P664L strain the mutation spectrum was changed . There was a significant increase in the frequencies of base substitutions , particularly G∶C→T∶A and G∶C→C∶G transversions characteristic of Polζ errors during spontaneous mutagenesis [49] ( Table 4 and Table S2 ) . Increases in deletions ranging from 1 to 5 bp and complex mutations ( two or more changes in a run of 10 bp ) were also observed . These types of changes were also previously attributed to the TLS activity of Polζ [48] , [50] . A similar mutation spectrum was also seen for CanRAde+ clones from strains containing the IS50-perfect palindrome ( Table 4 and Table S5 ) . Since mutagenesis observed in these strains requires the activity of Polζ , it is likely that error-prone synthesis by the TLS polymerase during DSB repair causes base substitutions as well as deletions and complex mutations . Consistently , errors that could be assigned to the activity of Polζ were suppressed in pol3-P664LΔrev3 strains ( Table 4 and Table S3 ) . We also uncovered large deletions ( up to 39 bp ) and a duplication of 27 bp flanked by short direct repeats in pol3 mutants with or without Alu-IRs . This is most probably attributed to the defective Polδ . Notably , pol3-P664L strains that lack fragile motifs also exhibited complex mutations ( Table 4 and Table S4 ) . Taking into account that fragility in pol3-P664L without Alu-IRs is low , it can be inferred that these changes reflect mutations arising during DNA replication carried out by a faulty DNA polymerase ( a process that also might require TLS polymerases [51] ) rather than a consequence of error-prone synthesis during DSB repair . Overall , analysis of mutation spectra in wild-type and replication-deficient strains is in agreement with genetic analysis and supports the conclusion that repeat-mediated mutations are generated by error-prone Polζ and do not occur due to faulty synthesis by replicative polymerases . To determine if DSB-induced mutagenesis can be observed at another fragile motif , we assessed CAN1 mutation rate in strains carrying 230 repeats of the triplex-adopting GAA/TTC ( Figure 1 ) . Although the rate of CanR mutations was unaltered in wild-type strains , a 4-fold increase in mutagenesis was detected in pol3-P664L and TET-POL3 strains ( Table 5 ) . The level of DSB formation at GAA/TTC repeats in the TET-POL3 strain was estimated to be 3 . 3% ( average of 3 . 1% , 3 . 3% and 3 . 6% , Figure 3 ) . This is a minimal estimation of GAA/TTC-mediated DSBs since , unlike the situation with palindromic sequences , resection of the broken fragments cannot be prevented by SAE2 disruption and a proportion of degraded DSBs are excluded from detection . Similar to Alu-IR-mediated mutagenesis , Polζ plays a role in the induction of mutations by GAA/TTC repeats . There was a mild but statistically significant reduction ( 2-fold ) in mutagenesis in pol3-P664LΔrev3 versus pol3-P664L ( p<0 . 05 ) and TET-POL3Δrev3 versus TET-POL3 ( p<0 . 05 ) strains as determined using an unpaired t-test . Although it is difficult to evaluate the contribution of resection and long ssDNA to GAA/TTC-associated mutagenesis , the involvement of REV3 suggests that the mechanism underlying mutagenesis in the case of inverted repeats and GAA/TTC fragile sites can be similar .
The induction of DSBs using site-specific endonucleases has been shown to drive mutagenesis [12]–[15] . This study demonstrates that natural chromosomal fragile sites comprising of sequence motifs that can adopt non-B DNA structures are also mutagenic . Under condition of replication stress , the mutagenesis can reach up to the levels caused by deficiency in the mismatch repair system [52] . We also show that the mutations are a consequence of error-prone repair of repeat-induced DSBs . Overall , we establish secondary structure-forming motifs as a potent source of endogenous mutagenesis and reveal the mechanism underlying this phenomenon . In this study we found that when replication is compromised , Alu-quasi-palindrome promotes chromosomal fragility and mutagenesis at CAN1 and URA3 reporters located 8 kb from the break site . Mutations were also increased in strains with a perfect IS50-palindrome with inherently higher fragility even in replication-proficient strains . We have previously shown that inverted repeats induce hairpin-capped DSBs in replication-proficient strains [21] . We have found that in replication-defective mutants the DSBs mediated by the Alu-quasi-palindrome also have hairpin-capped termini ( Y . Zhang , N . Saini , Z . Sheng , K . S . Lobachev , in preparation ) . The opening of the hairpins necessitates the nuclease activity of the MRX complex and Sae2 . The requirement of Sae2 for mutagenesis at distant loci unequivocally demonstrates that mutations are a consequence of DSB formation ( Table 1 and Table 2 ) . Moreover , these data also implicate the formation of long ssDNA upon resection of DSB ends as the second step in repeat-mediated mutagenesis . ssDNA has been shown to be prone to accumulation of mutations during SSA or in a situation where the telomeres become uncapped [15] . Therefore , it is possible that hairpin-processing generates damaged ssDNA that can serve as a faulty template for synthesis during SSA or GC . Alternatively , the undamaged ssDNA can be involved in strand invasion and mutations could arise due to error-prone synthesis during homologous recombination as suggested in other studies [9] , [13] , [14] . Error-prone synthesis of the undamaged DNA template in replication deficient strains by Polζ was observed by Northam et al . [51] . Although we cannot determine whether mutagenesis is due to accumulation of damage in resected DNA or error-prone synthesis on undamaged template , our data point towards synthesis-dependent strand annealing ( SDSA ) as the underlying mechanism for mutagenesis ( Figure 4 ) . None of the analyzed CanR clones contained interstitial deletions and all of the clones retained intact Alu-IRs or IS50-palindrome ( data not shown ) . This suggests that SSA is unlikely to operate during Alu-IR-mediated mutagenesis and alludes to a template-dependent repair process that involves the undamaged sister chromatid . Thus , we favor a scenario wherein hairpin-capped DSBs are induced in late S or G2 stage of the cell cycle . Upon hairpin opening by Sae2 and MRX , the 3′ end of the resected DSB invades the intact sister chromatid template . The requirement for invasion in mutagenesis is a likely step but ultimately cannot be proven by using rad51 or rad52 mutants for two reasons: these strains exhibit a mutator phenotype on their own [49] and Rad51 and Rad52 proteins are required for DSB formation at the Alu-IRs in replication-defective strains ( Y . Zhang , N . Saini , Z . Sheng , K . S . Lobachev , in preparation ) . It is conceivable that the invasion event can proceed either as a BIR or as an SDSA event . SDSA is the most probable mechanism owing to the fact that mutations were observed in both TP and TD reporters and that reduced mutation rates were measured at reporters 30 kb from the break site ( Table 3 ) . It is important to note that SDSA preserves the original inverted repeats that can trigger additional rounds of breakage and associated mutagenesis . If extrapolated to humans , these observations identify secondary structure-forming repeats as a potent source of mutagenesis that can change the expression of flanking genes during the lifetime of healthy individuals even in the absence of exogenous damage . Mutations generated in the reporter 8 kb away from DSB site were also strongly dependent on the activity of Polζ ( Table 1 and Table 2 ) indicating that the error-prone translesion synthesis operates during SDSA . This is consistent with the Hirano and Sugimoto , 2006 study that showed that Mec1 kinase is needed to recruit the Polζ-Rev1 complex to the DSB site [53] and other studies where DSB-induced mutagenesis required Polζ [12]–[15] . Analysis of mutations in the CAN1 locus of the hyper-fragile strains revealed an increase in G∶C→T∶A and G∶C→C∶G transversions , frameshift and complex mutations that are signatures of Polζ ( Table 4 ) [48]–[50] . In this study we also show that DSB-triggering long GAA/TTC repeats induce mutagenesis at distant loci , indicating that a similar underlying mechanism of mutagenesis described above for inverted repeats can operate for triplex-forming motifs . The requirement of Rev3 for mutagenesis is more evident for inverted repeats than for GAA/TTC repeats . It would be interesting to see if other TLS polymerases , Rev1 and Polη , besides Polζ operate in GAA/TTC-associated mutagenesis and to determine if the mutation spectra in GAA/TTC- and inverted repeat-containing strains differ . It is also important to note that in our experimental system , we observe mutagenesis by GAA/TTC tracts only under conditions of compromised replication wherein the repeat-mediated fragility is further increased . In other studies , mutagenesis is induced by GAA/TTC repeats in replication-proficient strains [37] , [38] . These discrepancies might reflect the distance of the used reporter from the fragile motif . It is possible that at closer distances , SSA might be the predominant pathway for mutagenesis where mutations introduced by Polζ can be scored above the spontaneous level of mutagenesis , while SDSA requires higher frequencies of breakage and longer ssDNA . This can be checked experimentally in future studies . Our data are also in agreement with the recent study by Shah et al . , 2012 wherein GAA/TTC-induced mutations in a closely juxtaposed reporter in Polδ mutants were dependent on Polζ [39] . Overall , this study demonstrates that fragile sequence motifs that are found in eukaryotic genomes , including humans , can be potent inducers of mutagenesis . Thus , secondary structure-adopting repeats can represent a dual threat to DNA stability by changing the structural organization of the genome and causing mutations . Recent studies linking the occurrence of mutations near chromosomal rearrangement break-points in primates and humans suggest that error-prone repair of DSBs can operate during speciation , evolution and tumorigenesis [54]–[57] . Thus it is likely that fragile and mutagenic non B-DNA-forming motifs are contributing factors to these processes .
The yeast strains used for the analysis of the inverted repeat-induced mutagenesis were derivatives of the KT19 strain ( MATa , bar1-Δ , his7-2 , trp1-Δ , ura3-Δ , leu2-3 , 112 , ade2-Δ , lys2-Δ , cup1-Δ , yhro54c-Δ , cup2-Δ , V34205::ADE2lys2::Alu-IRs , V29616::CUP1 ) . GAA/TTC-mediated mutagenesis was measured in strains that were derivative of YKL36 ( MATa , bar1-Δ , his3-Δ , trp1-Δ , ura3-Δ , leu2-Δ , ade2-Δ , lys2-Δ , V34205::ADE2lys2:: ( GAA ) 230 ) . Alu-IRs and GAA repeats were inserted into the BamHI site and the IS50 palindrome was inserted into HpaI site in LYS2 . The strains without repeats had an intact LYS2 gene . For measuring the distance dependence of repeat-induced mutagenesis , URA3 was amplified from pRS306 with flanking regions for the points of insertion into chromosome V . URA3 was inserted close to the repeats in lys2 ( 586 bp TP and 352 bp TD ) , ∼8 kb TD of the repeat locus between SGD coordinates 42096 and 42097 and ∼30 kb TP between SGD coordinates 11910 and 11911 and TD between coordinates 64686 and 64687 ( Figure 1 , Table S6 ) . The pol3-P664L allele was created via site-directed mutagenesis using p170 [58] . The mutation P664L results in the appearance of the AseI site . The plasmid was digested with HpaI and the mutation was obtained using pop-in pop-out methodology . The mutant shows mild temperature sensitivity at 37°C . The tetracycline promoter construct was obtained from Euroscarf ( pCM225 ) . PCR was performed with primers carrying overhangs for RFA2 and POL3 promoter regions and one-step integration was used to replace the promoters for RFA2 and POL3 ( Table S6 ) . REV3 was replaced with the kanMX cassette in wild-type and pol3-P664L strain and with the hphMX cassette amplified from pAG32 in the TET-POL3 and TET-RFA2 strains [59] . SAE2 was disrupted with the kanMX cassette in the wild-type strains and with TRP1 in pol3-P664L , TET-POL3 and TET-RFA2 strains . Fluctuation tests were carried out to estimate mutation and GCR rates . The strains were allowed to grow on YPD agar for 3 days at 30°C . The TET-POL3 and TET-RFA2 strains were grown on YPD containing 2 µg/ml and 0 . 1 µg/ml doxycycline , respectively . At these chosen concentrations of doxycycline the proteins are downregulated leading to an increase in fragility without significantly affecting viability of the strains . 14 individual colonies were diluted in 0 . 25 ml water each and serial dilutions were made to approximately 1∶10000 . The cultures were plated on YPD and on L-canavanine ( 60 mg/L ) low adenine ( 5 mg/L ) containing synthetic media in order to obtain approximately several hundred colonies per plate after incubating for 3 days at 30°C . White colonies on canavanine-containing media are indicative of mutations in CAN1 while red colonies depict GCR events . For mutation rate estimation at URA3 , the cultures were appropriately diluted and plated on 5-FOA ( 1 g/L ) containing synthetic media lacking adenine . Mutation rates and 95% confidence intervals were calculated as previously described [21] . Yeast cells were embedded into agarose plugs at a concentration of 2×109 cells/ml for detection of inverted-repeat-mediated DSBs and at a concentration of 8×109 cells/ml for detection of GAA/TTC-induced DSBs . The chromosomes were separated using contour-clamped homogeneous electric field gel electrophoresis as described previously [36] and transferred onto a nylon membrane . Southern hybridization was carried out using a probe specific to HPA3 that is telomere-proximal to the repeats . Densitometry analysis was performed using ImageJ ( NIH ) and the intensity of the broken product was normalized against the intact chromosome V . CAN1 mutants were obtained by plating approximately 30 individual colonies from two independent isolates for each strain on L-canavanine low adenine containing synthetic media . The CanRAde+ isolates were then streaked out on YPD to obtain single colonies from which DNA was extracted . PCR was carried out using primers 60 bp upstream and 158 bp downstream of CAN1 . The PCR product was sequenced using 4 internal primers for CAN1 such that the entire gene would be covered at least twice during sequencing . The primers used for sequencing CAN1 are can1-o1: 5′CATCTACTGGTGGTGACAAAG3′; can1-s1: 5′GCCACGGTATTTCAAAGCTTGC3′; can1-s2: 5′GGCTCTTGGAACGGATTTTC3′; can1-s3: 5′TGTAGCCATTTCACCCAAGG3′ . The sequencing results are depicted in Table 4 and Tables S1 , S2 , S3 , S4 , S5 . To quantify the changes in protein expression POL3 was tagged with 13 copies of c-Myc-epitope tag and RFA2 was tagged with 9 copies of the c-Myc-epitope tag at the C-terminus in both wild-type and tetracycline downregulatable strains . TET-POL3 was grown in the presence of 2 µg/ml doxycycline overnight , while TET-RFA2 was grown with 0 . 1 µg/ml doxycycline overnight . Total protein was extracted as previously described [60] and separated using SDS-polyacrylamide gel electrophoresis on an 8% gel . After electrophoresis the gel was blotted on a nitrocellulose membrane and probed with an antibody specific to c-Myc ( Genscript ) and an antibody for Pgk1 ( Invitrogen ) . The membrane was further treated with anti-mouse secondary antibody ( Genscript ) and chemiluminescent detection was carried out using the protocol described by GE Healthcare . Densitometry analysis was performed using ImageJ ( NIH ) and the intensity of Pol3 and Rfa2 were normalized against the loading control Pgk1 . | Eukaryotic chromosomes include regions that are susceptible for breakage and rearrangements . Repeats that can adopt non-B form DNA secondary structure are often found to be responsible for the induction of rearrangements . Here , we demonstrate that inverted repeats and GAA/TTC breakage sites are also sources of point mutagenesis that can spread to the genes located at long distances from the repeats . Remarkably , repair of the break involving error-prone synthesis restores inverted repeats making them a long-term resource of mutations . These results demonstrate that chromosomal regions with breakage motifs have a high potential for structural rearrangements as well as a tendency to accumulate nucleotide polymorphisms . Increased genetic changes in such regions may alter the rate of changes at the evolutionary scale and contribute to the development of diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"gene",
"identification",
"and",
"analysis",
"genetics",
"molecular",
"genetics",
"biology"
] | 2013 | Fragile DNA Motifs Trigger Mutagenesis at Distant Chromosomal Loci in Saccharomyces cerevisiae |
Urinary tract infection ( UTI ) is one of the most common bacterial infections with frequent recurrence being a major medical challenge . Development of effective therapies has been impeded by the lack of knowledge of events leading to adaptive immunity . Here , we establish conclusive evidence that an adaptive immune response is generated during UTI , yet this response does not establish sterilizing immunity . To investigate the underlying deficiency , we delineated the naïve bladder immune cell compartment , identifying resident macrophages as the most populous immune cell . To evaluate their impact on the establishment of adaptive immune responses following infection , we measured bacterial clearance in mice depleted of either circulating monocytes , which give rise to macrophages , or bladder resident macrophages . Surprisingly , mice depleted of resident macrophages , prior to primary infection , exhibited a nearly 2-log reduction in bacterial burden following secondary challenge compared to untreated animals . This increased bacterial clearance , in the context of a challenge infection , was dependent on lymphocytes . Macrophages were the predominant antigen presenting cell to acquire bacteria post-infection and in their absence , bacterial uptake by dendritic cells was increased almost 2-fold . These data suggest that bacterial uptake by tissue macrophages impedes development of adaptive immune responses during UTI , revealing a novel target for enhancing host responses to bacterial infection of the bladder .
Urinary tract infection ( UTI ) is one of the most common bacterial infections , impacting more than 130 million people annually worldwide [1 , 2] . The principal causative agent is uropathogenic Escherichia coli ( UPEC ) , accounting for more than 75% of all community acquired infections , particularly among a seemingly healthy population ( e . g . , premenopausal women ) [1] . In uncomplicated UTI ( i . e . , cystitis ) , nearly half of all women infected will experience recurrence [3] . Currently , there is little consensus in the field regarding the underlying causes of the high rate of recurrence . Mechanisms previously proposed to explain this phenomenon include that UPEC forms protected reservoirs in the bladder , remerging at later time points after initial infection [4 , 5]; that UPEC strains colonize the gut and periodically migrate to the urinary tract [6]; or that the immune response to infection is suppressed by mast cell-derived IL-10 in the bladder [7] . The innate immune response to UPEC infection is characterized by robust cytokine and chemokine expression , leading to rapid neutrophil and monocyte infiltration and subsequent bacterial clearance [8–12] . Depletion of both neutrophils and monocytes , by Gr1 antibody treatment [13] , leads to increased bacterial burden , whereas a reduction in circulating neutrophils alone decreases bacterial burden , suggesting that monocytes help eliminate bacteria in the bladder [10 , 11] . A recent study demonstrated that innate immune cell crosstalk is necessary for a coordinated innate response , whereby resident macrophages , responding to signals from infiltrating monocytes , induce MMP9 expression in neutrophils , in turn facilitating their trans-urothelial migration [14] . This mechanism likely works in concert with cytokine and chemokine expression from infected urothelium that mediates neutrophil recruitment and trans-urothelial migration [9] . The mechanisms involved in the initiation of adaptive immunity , and indeed the full nature of the response generated from the bladder during UTI , remain unclear [15] . The majority of studies have focused on innate immunity to UTI , such as neutrophil or monocyte infiltration , while only a limited number of studies have focused on adaptive immune mechanisms [16] . For example , UPEC-specific antibodies arise during UTI in mice , non-human primates , and human patients , and can inhibit UPEC binding to urothelial cells in vitro [17–19] . With respect to the role of effector cells , only one study has examined the induction of antigen-specific antibody and T cell responses after UPEC infection , demonstrating that transfer of serum or T cells from infected animals limits infection in naïve mice [19] . In this study , we investigated the initiation of adaptive immunity to UPEC to determine whether defects exist preventing the induction of sterilizing immunity . We conclusively demonstrated that adaptive immune responses are generated in response to UPEC infection; however , they are insufficient to prevent reinfection . We performed the first systematic analysis of the tissue-resident immune cell compartment in the steady state bladder of mice and investigated the role of macrophages , and their precursors , in the adaptive immune response during UTI . Strikingly , macrophage depletion , prior to primary infection , improved adaptive immune responses to challenge infection in a macrophage-replete environment . We observed that upon infection , macrophages were the principal population , among the antigen presenting cells , to acquire UPEC early in infection , and in their absence , bacterial uptake by dendritic cells ( DCs ) was increased . These data support a model in which bladder-resident macrophages sequester bacteria , consequently limiting adaptive immune responses , and provides an explanation for the failure of the immune system to respond effectively to UPEC infection .
Surprisingly , no study has directly tested the necessity of an adaptive immune response to limit UPEC reinfection or the role of specific components of the adaptive immune system in generating these responses . We employed a model of UPEC-induced cystitis in which 107 colony-forming units ( CFU ) of UPEC isolate UTI89 , made resistant to either ampicillin or kanamycin , were instilled intravesically into 7–8 week old female wildtype C57Bl/6 or C57Bl/6 RAG2-/- mice [20] . Animals were sacrificed at 24 hours post-infection ( P . I . ) to assess bacterial burden or monitored for bacteriuria to evaluate the resolution of acute infection , defined by the absence of bacteria in the urine . Three to four weeks later , when the mice had resolved the primary infection , animals were challenged with 107 CFU of an isogenic UPEC strain , resistant to the antibiotic not employed for primary infection , and sacrificed 24 hours P . I . to evaluate bacterial clearance ( Fig 1A ) . Importantly , the use of isogenic UPEC strains , differing only by antibiotic resistance and fluorescent marker , permitted differentiation between quiescent bacteria residing in reservoirs established during primary UPEC infection [5] and the challenge strain . Of note , this distinction has not been made in previous reports , and thus it has remained unclear whether bacteria measured in the bladder after challenge infection derive from the primary or challenge infection , or represent a mixture of both infections [19] . After UPEC challenge in wildtype mice , we observed a >2 log reduction in CFU of the challenge UPEC strain compared to the bacterial burden 24 hours after primary infection ( Fig 1B ) . By contrast , the bacterial burden after challenge infection in RAG2-/- mice was similar to that after primary infection ( Fig 1B ) . As DCs are the principal cells to present antigen to lymphocytes , we investigated their role in inducing an adaptive immune response following a primary infection . Utilizing CD11c-DTR chimeric mice , we depleted CD11c-expressing DCs , prior to primary infection , by administration of two doses of diphtheria toxin ( S1 Fig ) . As previously reported , diphtheria toxin treatment also impacted the number of tissue resident macrophages ( S1 Fig and [21] ) ; however , this reduction was minimal and the number of macrophages present in toxin-treated mice was within the range of normal variance ( Table 1 ) . Twenty-four hours after depletion , we infected mice as described in Fig 1A , and followed resolution of infection by assessing bacteriuria . Upon resolution of the primary infection , mice were challenged with an isogenic UTI89 strain , as described above . To measure the bacterial burden following primary infection in the chimeric mice , an additional cohort of naïve , untreated chimeric mice received a primary infection at the same time as the infected animals received the challenge infection ( Fig 1C , 1° group ) . We assessed bacterial burden at 24 hours following primary or challenge infection and observed that animals treated with PBS were better able to clear UPEC after challenge compared to DC-depleted animals ( Fig 1C ) . Together , these results suggest that the reduction in CFU observed after a challenge infection is mediated by an adaptive immune response , dependent upon DCs and lymphocytes . Interestingly , this response to UPEC reduces bacterial burden but does not prevent re-infection after challenge ( Fig 1 and [19] ) . Finally , in our model , approximately 10% of the total CFU measured after a challenge infection arose from the primary infecting strain , representing the reservoir formed during infection . These bacteria appeared to be protected from host clearance mechanisms , as no differences in the number of bacteria present in the reservoir between wildtype and RAG2-/- mice ( Fig 1D ) , or PBS and DT-treated CD11c-DTR mice were observed ( Fig 1E ) . To understand how adaptive immune responses are initiated in the bladder , we began by investigating the resident immune cell compartment of the bladder . Earlier studies have described bladder-resident DCs [22–25] , however , a comprehensive analysis of all immune cell populations has not been previously performed . Thus , we executed a systematic analysis of the bladder-resident CD45+ immune cell compartment . Bladders from naïve C57Bl/6 mice were enzymatically digested and immunostained ( Materials and Methods , Table 2 ) . As the bladder’s broad autofluorescent signal interfered with immune cell detection , we developed a gating strategy to exclude nonhematopoietic autofluorescent cells during analysis ( S2A–S2C Fig ) . Antigen presenting cells ( APCs ) , defined as MHC II+ , comprised the majority of CD45+ cells ( 69% ± 7 . 5 of CD45+ cells , Fig 2A and 2E ) . Macrophages , delineated by CD64 [26 , 27] and F4/80 co-expression , were by far the largest APC population ( ~40% of CD45+ cells ) ( Fig 2B and 2E ) . The CD11b+ and CD103+ dendritic cell ( DC ) subsets represented 15% and 5% of CD45+ cells , respectively ( Fig 2B and 2E ) . Within the MHC II- CD11b- gate , we identified NK1 . 1+ NK cells , CD11blo-int cKit+ IgE+ mast cells [28] , CD3+ CD4+ , and CD3+ γδ+ T cells , but never observed CD8+ T cells in naïve bladders ( Fig 2C and 2E ) . Recently , it was reported that classical Ly6C+ monocytes constitutively traffic into naïve nonlymphoid tissues , such as skin , in the steady state [29 , 30] . Accordingly , in the MHC II- CD11b+ gate , we identified resident Ly6C+ monocytes as well as SiglecF+ eosinophils [31] ( Fig 2D and 2E ) . Notably , no neutrophils were observed in naïve bladders . In addition to neutrophils , monocytes infiltrate the bladder upon UPEC infection [8 , 11] . To determine the fate of infiltrating monocytes , we employed in vivo labeling of circulating monocytes to monitor their entry into infected bladders [8 , 32] . Mice were infected 24 hours after labeling circulating classical or nonclassical monocytes and sacrificed at 4 , 24 , and 48 hours P . I . , for analysis by flow cytometry . In line with observations from other infection models [33 , 34] , a greater number of classical monocytes infiltrated the bladder over time than nonclassical monocytes ( Fig 3A , note the scales of the y-axes ) . In the classical labeling protocol , a majority of infiltrated bead+ cells upregulated CD11c , MHC II , and CD64 expression , and downregulated CD11b and Ly6C from 24 to 48 hours P . I . , phenotypically resembling resident macrophages ( gray histograms ) ( Fig 3B ) . The percentage of bead+ cells that were identified as macrophages increased from 24 to 48 hours , while only 10% of all bead+ cells had a DC phenotype ( Fig 3C ) , supporting the conclusion that infiltrating classical monocytes predominantly differentiate into macrophages during UTI . During infection , infiltrating monocytes increased the already substantial macrophage compartment , thus we hypothesized that monocyte-derived and/or resident macrophages might play an important role during UPEC infection . To test this hypothesis , we depleted each population separately to determine the impact on bacterial burden . Circulating monocytes , but not bladder-resident macrophages , were depleted with clodronate liposomes [35] ( S3A and S3B Fig ) , and mice were infected 15–18 hours later with UPEC . Mice were sacrificed 24 hours P . I . to determine CFU . Monocyte-depleted animals had a small ( <1-log ) but significant improvement in bacterial elimination 24 hours P . I . ( Fig 4A ) . This difference , however , was lost if mice were infected 24 hours post-clodronate treatment , when monocytes have begun to repopulate the circulation , suggesting that monocyte depletion has a transient impact on bacterial burden . Supporting this conclusion , there were no differences in bacterial burden 24 hours P . I . in CCR2-/- mice ( S4 Fig ) , which have greatly reduced numbers of circulating monocytes [33] , as compared to wildtype mice . Bladder-resident macrophages were eliminated by administration of anti-CSF1R depleting antibody 24 hours prior to infection ( S3C Fig ) [36] . Importantly , the anti-CFS1R antibody also targeted monocytes , however these cells were not completely eliminated from circulation at the time of infection ( compare S3A Fig to S3D Fig ) . Mice depleted of macrophages had a similar bacterial burden at 24 hours P . I . compared to non-depleted mice ( Fig 4B ) suggesting that the absence of macrophages does not impact UPEC clearance at early time points P . I . As macrophage depletion had no impact on the primary infection , we considered whether their absence might influence the generation of an adaptive immune response . Indeed , macrophages can impact adaptive immunity via cytokine secretion or antigen sequestration . To directly test the influence of monocytes and macrophages on the generation of adaptive immunity during UTI , we depleted each of these cell types , as above . To address the role of monocytes , mice were depleted by clodronate treatment , infected , and subsequently challenged with an isogenic UTI89 strain and sacrificed at 24 hours P . I . to determine CFU , as in ( Fig 1A ) . We did not observe a difference in bacterial burden at 24 hours post-challenge ( Fig 4C ) or in reservoir formation ( S5 Fig ) , suggesting that the absence of monocytes , before primary infection , did not influence bacterial clearance after challenge infection . Further , these data demonstrate that the small improvement observed in bacterial clearance at 24 hours post-primary infection after clodronate treatment ( Fig 4A ) , did not influence the development of an adaptive immune response to the bacteria . To determine whether resident macrophages influence bacterial clearance after challenge , we treated mice with anti-CSF1R and infected with UPEC . Upon resolution of the primary infection , mice were challenged with an isogenic UTI89 strain and bacterial burden determined 24 hours post-challenge . In the course of the experiment , macrophage depletion did not impact reservoir formation ( S5 Fig ) . Despite similar clearance during the primary infection , we observed a surprising reduction of nearly 2 orders of magnitude in CFU after challenge infection of mice depleted of macrophages prior to the primary infection as compared to control isotype-treated and untreated ( infected and challenged , but not receiving antibody injection ) mice ( Fig 4D ) . Importantly , at the time of the challenge infection , macrophages had repopulated the bladder in depleted animals , ruling out the possibility that bacterial burden was influenced by the absence of macrophages during challenge infection ( S3E Fig ) . To test whether this improvement in bacterial clearance in the absence of macrophages was dependent upon components of the adaptive immune system , we depleted macrophages in RAG2-/- mice . We observed no difference in the CFU per bladder after UPEC challenge between macrophage-depleted or control treated RAG2-/- mice ( Fig 4E ) . To specifically assess the necessity of T cells , we depleted macrophages in mice that had been treated with CD4 and CD8 depleting antibodies prior to primary infection . We observed fewer UPEC post-challenge only in macrophage-depleted mice that were replete of their T cells . However , mice depleted of CD4+ and CD8+ T cells did not demonstrate improved bacterial clearance after challenge independently of macrophage depletion ( Fig 4F ) . Together , these results support the conclusion that macrophage depletion at the time of primary infection positively impacts the capacity to generate a T cell-dependent adaptive immune response against UPEC . To investigate potential mechanisms mediating improved clearance after macrophage depletion in a challenge infection , we first focused on events following the challenge . We depleted macrophages and infected mice as above . When all mice had resolved the primary infection , we challenged the mice and 24 hours post-challenge , mice were sacrificed . We assessed immune cell infiltration into the bladder by flow cytometry and surprisingly , we did not observe differences in the number of T cells , B cells , neutrophils , or monocytes infiltrating the bladder after challenge infection ( Fig 5A and 5B ) . These data suggest that improved bacterial clearance is not mediated by increased numbers of innate or effector cells . In addition to cell infiltration , we evaluated UPEC-specific IgA in the urine over time , however , the levels of UPEC-specific IgA were at the limit of detection of the assay , as has been reported in other studies [7] . Macrophages can influence the generation of an adaptive immune response through the modulation of the cytokine microenvironment [37 , 38] . Thus , we evaluated cytokine expression in the bladder 24 hours after a primary infection in control or antibody-treated mice by luminex technology . Notably , we did not observe significant differences between isotype-treated and antibody-depleted mice in any of the 32 cytokines evaluated ( Fig 5C and S6 Fig ) , suggesting that cytokine expression is not significantly impacted at this timepoint in infection . Given that we found no differences in effector cell infiltration or cytokine expression , we considered whether macrophages sequester bacteria during infection . To test which immune cells acquire UPEC during infection , we utilized our kanamycin-resistant UTI89 strain , which also expresses MARS red fluorescent protein ( UTI89-RFP ) ( S7A and S7B Fig ) . Importantly , when we infected mice with our ampicillin-resistant UTI89 strain expressing GFP , we could not clearly differentiate between UTI89-GFP-containing cells and the background autofluorescence of bladder macrophages ( S7A Fig ) or urothelial cells ( S2 Fig ) . Importantly , UTI89-RFP had similar infectivity in vitro and in vivo compared to the parental UTI89 strain and UTI89-GFP ( S7C and S7D Fig ) . Mice were instilled with 107 CFU of UTI89-RFP and sacrificed at 4 , 24 , and 48 hours P . I . to analyze their bladders by flow cytometry . UTI89-RFP-containing cells were identified by gating CD45+ cells with RFP fluorescence levels greater than those in bladders infected with the non-fluorescent parental UTI89 strain ( Fig 6A ) and their phenotypes were determined based on expression of cell-specific markers as in Fig 2 . At 4 hours P . I . , the majority of UTI89-RFP+ cells were in the MHC II+ cell compartment . At 24 and 48 hours P . I . , UTI89-RFP+ cells were evenly distributed between the MHC II+ and II- gates ( Fig 6B ) . The MHC II- cells containing bacteria at 24 and 48 hours were primarily CD11b+ Ly6G+ Ly6C+ neutrophils and CD11b+ Ly6G- Ly6C+ monocytes . Among the UPEC+ MHC II+ cells , the majority exhibited a macrophage phenotype ( Fig 6C ) . Notably , two subpopulations were distinguishable in the macrophage gate at 24 and 48 hours P . I . , representing resident ( CD64hi F4/80hi Ly6C- ) and monocyte-derived macrophages ( CD64int F4/80int Ly6C+ ) ( Fig 6C ) . Indeed , while the number of DCs harboring UTI89-RFP changed very little , the number of macrophages containing bacteria increased more than 7-fold at 24 hours and remained elevated at 48 hours ( Fig 6D ) . At all timepoints analyzed , macrophages harbored approximately 60–80% of the bacteria found within the MHC II+ APC compartment , demonstrating that macrophages were the primary cell type to phagocytose bacteria at early timepoints post-infection ( Fig 6E ) . Macrophages can influence the generation of an adaptive immune response by antigen sequestration , as has been observed in the lung , hindering antigen presentation and subsequent T cell priming by DCs [39–42] . More specifically , it has been proposed that DC migration and the induction of adaptive immunity can only occur in the lung after the phagocytic capacity of macrophages has been saturated and excess antigen is available to DCs [42 , 43] . To test the hypothesis that antigen availability during primary infection impacted initiation of an adaptive response during UTI , we infected untreated mice with 107 , 108 , or 109 CFU of UPEC . When the animals had resolved the infection , they were challenged with 107 CFU of an isogenic UPEC strain and 24 hours post-challenge , sacrificed to determine bacterial burden . Notably , we did not observe a difference in the number of bacteria per bladder after a primary challenge despite increasing the inoculum 100-fold ( S8A Fig ) . Furthermore , we did not observe any differences in bacterial burden after challenge among the three groups ( S8B Fig ) . Thus , increasing the number of bacteria during primary infection did not improve the generation of an adaptive immune response during UTI , however it is possible that we did not saturate bladder resident macrophages with bacteria . To directly test whether macrophages were physically sequestering UPEC in the bladder during UTI , we evaluated bacteria uptake in the bladder in the context of macrophage depletion . Mice were depleted or not of macrophages by anti-CSF1R antibody , infected with UTI89-RFP and sacrificed 24 hours P . I . The distribution of UPEC was altered in mice depleted of macrophages as compared to the control group . More bacteria localized to MHC II- cells ( Fig 7A and 7B ) , potentially explaining why macrophage depletion did not impact bacterial clearance after primary infection . The percentage of neutrophils containing bacteria was specifically increased in macrophage-depleted mice , likely compensating for the lack of monocytes and resident macrophages ( Fig 7B ) . In MHC II+ cell populations , while the total number of DCs in infected bladders was unchanged ( Fig 7C ) , we observed a significantly greater percentage of DCs had taken up UPEC in macrophage-depleted animals ( Fig 7D ) . Notably , the percentage of UPEC-containing resident and MHC II+ monocyte-derived macrophages was not different between the isotype and depleting antibody groups ( Fig 7E ) .
UTI is unusual in that it is a common infection that recurs with high frequency , particularly in otherwise healthy adult women [1] , suggesting a defect exists in the ability to mount an adaptive immune response to UPEC . We observed that the absence of B and/or T cells or DCs impaired the host’s capacity to clear bacteria after a challenge infection , confirming that adaptive immune responses are primed during UTI . Although this was an expected result , surprisingly , it has never been formally demonstrated in the literature until now . We further demonstrated that while immune responses are primed , they neither prevent reinfection nor eliminate the bacterial reservoir established during primary infection . To shed light on potential mechanisms preventing the development of effective adaptive immunity to UPEC infection , we focused on the role of MHC II+ cells as they are the key initiators of adaptive immunity . Unexpectedly , in the context of a challenge infection , resident macrophage depletion improved the host’s ability to eliminate bacterial load . Importantly , macrophages were depleted prior to the first infection; however , they were present in normal numbers at the time of challenge infection . Notably , this improvement was dependent on the adaptive immune system as the phenotype was lost when macrophages were depleted in RAG2-/- mice or in mice depleted of T cells . Given that macrophages were the principal APC to acquire UPEC early in infection , these data suggest that macrophages subvert initiation of a robust adaptive immune response during UTI . To understand the mechanism of macrophage subversion of adaptive immunity during UTI , we evaluated events post-challenge and post-primary infection . We observed a significant infiltration of T cells and a smaller infiltration of B cells post-challenge but no major differences in cell numbers between the control and treated groups . At this time , we cannot rule out potential qualitative differences in the activation or specificity of the infiltrating effector cells , including T cells or possibly NK T cells , which may play a role in kidney infection [44] . As macrophages repopulated the bladder before challenge infection , we hypothesized that the impact on adaptive immunity occurred in the first few hours or days following primary infection . To explore the possibility of antigen sequestration , we evaluated which immune cell populations acquired UPEC in the bladder in the absence of macrophages . We observed an increase in the percentage of DCs containing bacteria . DCs are key players in initiating adaptive immune responses , and we found that they can do so from the bladder mucosa in the context of UTI . Indeed , even a partial depletion of bladder-resident DCs , prior to primary infection , rendered animals less capable of clearing bacteria after challenge infection . The intermediate clearance phenotype observed in DT-treated mice may have been mediated by DCs remaining after depletion or by DCs repopulating the tissues before the primary infection was resolved , permitting delayed antigen presentation . Thus , our data suggest that the more efficacious adaptive immune response against UPEC observed during macrophage depletion may be mediated by the increase in the percentage and number of DCs carrying bacteria . Notably , however , the proportion of DCs containing UPEC in infected bladders was less than 5% of total DCs present in mice , suggesting that only a very small number of antigen-carrying DCs are required to mount an adaptive immune response during UTI . Macrophages outnumbered DCs in both naïve and infected bladders and were the principal cell to acquire UPEC . Our findings contradict a recent study in which Schiwon et al . suggest that bladder-resident macrophages sense UPEC infection , but do not phagocytose bacteria [14] . As an explanation for this apparent discrepancy , we found that MHC II+ cells containing GFP-expressing bacteria were indistinguishable from autofluorescent but uninfected cells . We engineered UTI89 to express a red fluorescent protein to specifically overcome the challenge of distinguishing naturally autofluorescent cells in the bladder ( e . g . , macrophages and urothelial cells ) from those containing UPEC . Lending credence to this interpretation , the authors also did not detect GFP-expressing UPEC in urothelial cells [14] , which are invaded during the course of UTI , as their autofluorescence also likely masked the GFP signal [45 , 46] . We evaluated macrophages because of their prominent role in bacterial acquisition . However , even when the majority of UPEC was captured by macrophages at early timepoints P . I . , their depletion did not impact bacterial clearance after the primary infection . This apparent contradiction may be explained by the increased bacterial uptake by neutrophils observed in the absence of macrophages . Our data support a model in which macrophages sequester bacteria from DCs early in infection , however we cannot rule out that depletion of macrophages alters the microenvironment during infection , despite our negative findings in bladder homogenates . Indeed , a recent study suggests that IL-10 expression from mast cells suppresses adaptive immunity to UPEC [7] . However , in the course of our study , we found few mast cells in naïve bladder tissue . Furthermore , multi-analyte cytokine analysis revealed no striking differences between control and depleted mice and we could not detect IL-10 expression in this or a prior study [11] . The reasons for this are unclear , however may be due to the significant variation that exists in the genomes of commonly used strains such as cystitis strain UTI89 , pyelonephritis strains J96 , 563 , CFT073 , and clinical isolates ( see phylogenetic tree in [47] ) . Having defined an early role for resident macrophages and DCs during UTI , our work significantly advances the understanding of how adaptive responses to UPEC are achieved . However , we still do not completely understand how the adaptive immune system eliminates UPEC . Though we were not able to detect UPEC-specific antibodies above the limit of detection of our assay , others have identified that antibodies against the bacteria are generated during infection [17–19] . Thus , we hypothesize that the protection induced during UTI following a primary infection is mediated by an antibody response; however T cells may also play a critical role in the killing of infected cells . With respect to the role of the bladder macrophage , our data point to a barrier in the immune system that must be overcome , particularly for patients with recurrent UTI . Although macrophage sequestration of particulate antigen in the lung has been described , this is , to the best of our knowledge , the first study to propose a role for the physical sequestration of antigen during live bacterial infection . Strategies that increase DC number or migration may overcome the subversion imposed by macrophages , providing a viable solution to treat patients with recurrent UTI .
At Mount Sinai School of Medicine , mouse experiments were conducted in accordance with approval of protocol number LA11-00003 by the Institutional Animal Care and Use Committee at Mount Sinai School of Medicine , which adheres to the guidelines put forth by the Animal Welfare Act and the Public Health Service policy on Humane Care and Use of Laboratory Animals . At Institut Pasteur , mouse experiments were conducted in accordance with approval of protocol number 2012–0024 by the Comité d’éthique en expérimentation animale Paris Centre et Sud ( the ethics committee for animal experimentation ) , in application of the European Directive 2010/63 EU . In our experiments , mice were anesthetized either by inhalation of isoflurance ( 3–4% ) or by injection of 100 mg/kg ketamine and 5 mg/kg xylazine and sacrificed either by cervical dislocation or carbon dioxide inhalation . The human UPEC cystitis isolate , UTI89 ( kind gift from Scott Hultgren ) [46] , and the fluorescent protein-expressing strains UTI89-RFP and UTI89-GFP were used for infection . Briefly , fluorescent bacteria were engineered using lambda red recombination [48] to introduce an aphA-marsRFP or bla-GFP cassette in the UTI89 chromosome at the attB lambda phage integration site . UTI89 is sensitive to antibiotics , UTI89-RFP is resistant to kanamycin , and UTI89-GFP is resistant to ampicillin . Bacteria were grown overnight in static cultures at 37°C in Luria-Bertani broth ( LB ) in the presence of antibiotics ( kanamycin 50 μg/mL or ampicillin 100 μg/mL ) where appropriate . The mouse urothelial cell line NUC-1 [49] was used to evaluate the in vitro invasion efficacy of each UTI89 strain . Fifty thousand cells were infected with UTI89 , UTI89-GFP , or UTI89-RFP at increasing MOIs . Thirty minutes post-infection , cells were washed , lysed , and serial dilutions were plated . Percent invasion was calculated by dividing the number of bacteria inside the cells by the inoculum x 100 . Female C57BL/6 mice between 6 and 8 weeks old were from The Jackson Laboratory or Charles River . CD11c-DTR mice were a kind gift from Marc Lecuit and Claude LeClerc ( Institut Pasteur ) . RAG2-/- mice were a kind gift from Antonio Freitas ( Institut Pasteur ) . Briefly , mice anesthetized with isoflurance ( 4% ) or 100 mg/kg ketamine and 5 mg/kg xylazine were infected with 107 colony-forming units ( CFU ) of one of two UTI89 strains in 50 μL PBS via a catheter introduced into the urethra [20] except in the inoculum escalating experiment , where mice received 107 , 108 or 109 CFU in 50 μL PBS . To calculate CFU , bladders were aseptically removed and homogenized in 1 mL PBS . Serial dilutions were plated on LB agar , with or without antibiotics , as required . For challenge infection experiments , mice were infected with one of the two fluorescent strains of UTI89 , expressing antibiotic resistance ( kanamycin or ampicillin ) ( See Fig 1A ) . Once the primary infection cleared , 3 to 4 weeks , mice were infected with 107 CFU of an isogenic UTI89 strain with a different antibiotic resistance . The strain used for the challenge infection was determined by that used in the primary infection , such that the antibiotic resistances were different between the primary and challenge infection , e . g . , UTI89-GFP for the primary and UTI89-RFP for the challenge infection . Importantly , both strains were used as the primary or the challenge strain in different experiments . Resolution of infection was monitored by plating urine every 5–6 days on antibiotic-containing plates . C57Bl/6 mice were irradiated with a single dose of 5–6 gray in an x-ray irradiator at 6 weeks of age . Animals were reconstituted with 1 . 6–3 . 2 x 106 total bone marrow cells from CD11c-DTR mice 6 hours after irradiation . Mice were allowed to reconstitute for a minimum of 12 weeks and reconstitution was evaluated by flow cytometry of congenic markers . 24 and 48 hours prior to infection , mice were administered 4 ng/g of diphtheria toxin I . V . Depletion efficiency was tested in each batch of chimeric mice prior to experimentation . At indicated timepoints , bladders were removed and minced with dissection scissors into tubes containing digestion buffer kept at 4°C . Minced tissue was then incubated at 37°C in 1 mL of digestion buffer containing 0 . 34 U/mL of Liberase TM ( Roche ) and 100 μg/mL of DNase in PBS . Tubes were vigorously shaken by hand every 15 minutes . 45 minutes to one hour post-incubation , when the tissue had a glassy , transparent appearance and was almost entirely digested , digestion was stopped by adding several mL of PBS supplemented with 2% FBS and 0 . 2 μM EDTA . The entire bladder digest was passed through a 100 μM cell strainer to obtain a single cell suspension . Gentle pressure was applied to any tissue remaining in the strainer . Samples were washed , Fc receptors blocked , and stained with antibodies listed in Table 2 . Total cell counts in the bladder were determined by the addition of AccuCheck Counting beads ( Invitrogen ) to a known volume of sample after staining , just prior to cytometer acquisition . Gating strategies for all cell populations except for neutrophils are depicted in Fig 2 . Neutrophils were identified as MHC II- , CD11b+ , Ly6G+ , Ly6C- , SiglecF- , and F4/80- . To identify cell populations in the circulation , whole blood was incubated with BD PharmLyse , ( BD Bioscience ) and subsequently stained with antibodies indicated in the Table 2 . Samples were acquired on a BD LSRFortessa using DIVA software and data were analyzed by FlowJo ( Treestar ) software . Total cell counts in the blood were determined by the addition of AccuCheck Counting beads ( Invitrogen ) to 10 μL of whole blood diluted in 1-step Fix/Lyse Solution ( eBioscience ) . In vivo bead labeling of classical and nonclassical monocytes was performed as previously described [32] . Briefly , classical monocytes were labeled by I . V . administration of 200 μL clodronate liposomes to transiently deplete all monocytes and then by I . V . injection of 1 μM nondegradable fluorescent particles 24 hours later . Nonclassical monocytes were labeled by injection of 1 μM nondegradable fluorescent particles without prior monocyte depletion . Labeling efficiency was confirmed by flow cytometry . To deplete monocytes , wildtype C57BL/6 mice received I . V . injection of 200 μL of clodronate liposomes ( or PBS control liposomes ) 15–18 hours prior to infection [35] . Anti-CSF1R antibody ( 2 mg/mL , clone AFS98 , eBioscience ) was used to deplete bladder-resident macrophages . Animals received two I . V . injections , on consecutive days , of anti-CSF1R antibody or isotype control ( clone eBR2a , eBioscience ) . We administered 400 μg/mouse on day 1 and 200 μg/mouse on day 2 , to decrease the impact on circulating monocytes . To deplete T cells , 100 μg of CD4 ( clone GK1 . 5 , Bio X Cell ) and 100 μg of CD8 ( clone YTS 169 . 4 , Bio X Cell ) per mouse were injected together intraperitoneally 24 hours prior to primary infection . 200 μg of isotype control ( clone LTF-2 , Bio X Cell ) per mouse was injected intraperitoneally . The depletion was repeated 5 days post-infection , and once a week to maintain the depletion until challenge infection . Mice were infected with UTI89 and bladders removed 24 hours P . I . Bladders were homogenized with a handheld tissue grinder in 1 mL PBS on ice . After removal of a 100 μL aliquot to calculate CFU by serial dilution , bladder homogenates were clarified by microcentrifugation ( 13K , 4 , 5 minutes ) and stored at -80°C until assessment by Luminex Milliplex MAP Mouse Cytokine/Chemokine Magnetic Bead Panel , Premixed 32-Plex , according to the manufacturer’s recommendations ( Merck Millipore ) [11] . All samples were assessed together to avoid inter-assay variability . Just prior to analysis , after thawing , samples were centrifuged a second time to remove any cell debris . GraphPad Prism was used to evaluate statistical significant . Graphs depict medians and statistical significance was determined by the nonparametric Mann-Whitney test . | Urinary tract infection is a common infection with a high propensity for recurrence . The majority of infections are caused by uropathogenic E . coli , a growing public health concern with increasing prevalence of antibiotic resistant strains . Finding therapeutic options that circumvent the need for antibiotics , while boosting patients’ immune response to infection is desirable to counteract further increases in antibiotic resistance and to provide long-lasting resistance to infection . Currently , little is known about how adaptive immune responses , which typically prevent recurrent infection in other organs , arise from the bladder during urinary tract infection . Here , we investigated the initial interactions between immune cell populations of the bladder and uropathogenic E . coli , finding that macrophages are the principal cell population to engulf bacteria . Interestingly , these same cells appear to inhibit the development of adaptive immunity to the bacteria , as their depletion , prior to primary infection , results in a stronger immune response during bacterial challenge . We found that in the absence of macrophages , dendritic cells , which are the most potent initiators of adaptive immunity , are able to take up more bacteria for presentation . Our study has revealed a mechanism in which specific immune cells may act in a manner detrimental to host immunity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Macrophages Subvert Adaptive Immunity to Urinary Tract Infection |
The continuously prolonged human lifespan is accompanied by increase in neurodegenerative diseases incidence , calling for the development of inexpensive blood-based diagnostics . Analyzing blood cell transcripts by RNA-Seq is a robust means to identify novel biomarkers that rapidly becomes a commonplace . However , there is lack of tools to discover novel exons , junctions and splicing events and to precisely and sensitively assess differential splicing through RNA-Seq data analysis and across RNA-Seq platforms . Here , we present a new and comprehensive computational workflow for whole-transcriptome RNA-Seq analysis , using an updated version of the software AltAnalyze , to identify both known and novel high-confidence alternative splicing events , and to integrate them with both protein-domains and microRNA binding annotations . We applied the novel workflow on RNA-Seq data from Parkinson's disease ( PD ) patients' leukocytes pre- and post- Deep Brain Stimulation ( DBS ) treatment and compared to healthy controls . Disease-mediated changes included decreased usage of alternative promoters and N-termini , 5′-end variations and mutually-exclusive exons . The PD regulated FUS and HNRNP A/B included prion-like domains regulated regions . We also present here a workflow to identify and analyze long non-coding RNAs ( lncRNAs ) via RNA-Seq data . We identified reduced lncRNA expression and selective PD-induced changes in 13 of over 6 , 000 detected leukocyte lncRNAs , four of which were inversely altered post-DBS . These included the U1 spliceosomal lncRNA and RP11-462G22 . 1 , each entailing sequence complementarity to numerous microRNAs . Analysis of RNA-Seq from PD and unaffected controls brains revealed over 7 , 000 brain-expressed lncRNAs , of which 3 , 495 were co-expressed in the leukocytes including U1 , which showed both leukocyte and brain increases . Furthermore , qRT-PCR validations confirmed these co-increases in PD leukocytes and two brain regions , the amygdala and substantia-nigra , compared to controls . This novel workflow allows deep multi-level inspection of RNA-Seq datasets and provides a comprehensive new resource for understanding disease transcriptome modifications in PD and other neurodegenerative diseases .
Recent studies have identified conspicuous diversity in large intergenic long non-coding RNAs ( lncRNAs ) found across many species [1] [2] . LncRNAs are currently defined as transcripts of over 200 nucleotides [3] . Nonetheless , the GENCODE non-coding RNA set , the largest currently lncRNA database , contains currently as much as 136 spliced transcript shorter than 200 bp , and the general and structural annotation of lncRNA overall is still ongoing [4] . LncRNAs may contain open reading frames ( ORF ) , and are often transcribed by RNA polymerase II , spliced and polyadenylated – but do not code for any protein product . LncRNAs are the least well studied among thousands non-coding eukaryotic RNAs that have been discovered so far . While genome-wide expression and evolutionary analyses suggest that some of them play functional roles , their cellular mechanisms of action are still largely unknown [5] . Nonetheless , accumulating evidence suggests that in the nervous system , lncRNA functions span regulating brain evolution and neural development [6] and mediate behavioral and cognitive processes [7] . In Drosophila , the neuronal-expressed CRG lncRNA is involved in regulating locomotion by recruiting RNA polymerase II to the adjacent promoter of the movement-related protein-coding gene CASK , thereby increasing CASK expression [8] . In humans , lncRNAs are involved in neurogenesis , neuropsychiatric disorders [9] , cancer ( for example , HT19 which is involved in tumor growth ) [10] ) and in Autism [11] as well as in the neurodegenerative Huntington's [12] and Alzheimer's ( AD ) diseases [13] . However , the involvement of lncRNAs in the leading neurodegenerative motor disorder worldwide , Parkinson's disease ( PD ) , is still unknown . PD is the second most common neurodegenerative disease worldwide ( after AD ) [14] , [15] , with age being the leading risk factor currently known and no known cure . It affects 1–2% of the population above 65 years of age [16] , [17] , [18] and is characterized by four cardinal motor symptoms ( resting tremor , bradykinesia ( “slow movement” ) , postural instability and akinesia ( “lack of movement” ) [19] [20] . These appear when most of the brain's dopamine-producing neurons have already been diminished . Most cases are defined as ‘sporadic’ and treatment is aimed at replacing lost DA through adjusting the declining levels of the precursor L-Dopa . The alternative , deep brain stimulation ( DBS ) treatment allows a significant reduction in the medication dosage while drastically improving motor function in patients . DBS presumably alleviates the disease symptoms by targeting the basal ganglia Sub-Thalamic Nucleus ( STN ) brain region through yet undefined mechanisms [21] . While the underlying aetiology of sporadic PD remained elusive , genomics studies have implicated several genes in the loss of DA neurons . Mutations in α-synuclein ( SNCA ) , the first gene identified as linked to PD [22] , cause early-onset PD , and the SNCA protein product has been identified as a major component of Lewy bodies [23] , a morphological pathological hallmark of PD [24] . Mutations in the ubiquitin ligase Parkin ( PARK2 ) that targets proteins for degradation in the proteasome through linkage of ubiquitin molecules cause DA neuron pathology [25] and autosomal recessive Parkinsonism [26] . The DJ-1 ( PARK7 ) protein regulates oxidation–reduction signalling pathways via inducing gene expression [27] , inhibiting the formation of SNCA aggregates [28] and limiting dopaminergic cell death in cellular and animal PD models [29] . Mutations in the putative serine threonine kinase LRRK2 ( PARK8 ) cause uncoupling of mitochondria in fibroblast and neuroblastoma cells [30]; and the PTEN-induced protein kinase PINK1 ( PARK6 ) , mutations in which cause early-onset PD [31] is believed to be involved in mitophagy [32] . Taken together , these observations suggest that inherited , and possibly acquired impairments in the pathways regulating protein metabolism , oxidative stress and mitochondrial functioning are causally involved in PD emergence . Yet , current medications only improve the disease motor symptoms – but do not provide a cure . Furthermore , identification of the disease in its early stages , before the majority of the dopaminergic neuron population have diminished , is currently impossible . Transcriptome analysis of peripheral blood is of great interest for clinical research , as differences between samples obtained in a minimally invasive and cost-effective manner can be translated into gene signatures of disease , as well as disease stage , drug response and toxicity [7] . Blood cells interact with most tissues and organs in the human body and their cellular composition provides a reflection of both physiological and pathogenic stimuli , including brain treatment effects [33] . Furthermore , 80% of the genes expressed in peripheral blood cells are shared with other central tissues [12] . While nucleated white blood cells make up the minority of blood cells , they are the most informative . Correspondingly , gene expression differences in peripheral whole blood have been used to determine gene signatures related to both acute myeloid leukemia [8] and neuropsychiatric disorders and Huntington's disease , where significant correlation between blood and brain transcripts was identified [9] , [10] . Other neurological diseases for which peripheral blood-based biomarkers have been identified include multiple sclerosis , schizophrenia and Alzheimer's disease [34] , [35] , [36] . These effects have been specifically attributed to neuronal death , neuronal cell-free RNA expression and well-described neuro-immune modulatory effects [34] , [35] , [36] . Likewise , we have recently observed parallel changes in microRNAs ( miRNAs ) and genes predicted as their targets that further underwent splicing changes in PD leukocytes and in PD-relevant brain regions , including the substantia nigra ( SN ) as well as the frontal lobe ) through coupled analysis of small RNA-Seq data and splice junction arrays [37] . These spanned immune , mitochondrial and oxidative stress changes , supporting our microarray identification of interleukin-4 ( IL4 ) related processes in whole blood data from a large early PD cohort [38] . Since the first report of miRNA involvement in PD [39] , new findings provide ample evidence for involvement of differentially expressed miRNAs in the PD brain [40] [41] [42] [43] [44] [45] [46] . In differential expression studies of PD patients' leukocytes , we found expression changes that were partially reversed following DBS treatment [47] , [48] . Parallel changes were also detected in both the frontal cortex and the caudate-putamen brain areas from PD model mice treated with 1-methyl-4-phenyl-1 , 2 , 3 , 6-tetrahydropyridine ( MPTP ) neurotoxin . Thus , leukocyte datasets provide useful resource for identifying possible disease biomarkers ( which are urgently needed for PD ) , and for studying PD-related processes in an easily accessible tissue . With the advent of current genomic technologies , new tools are required for linking between different datasets produced from various technologies including different types of microarrays , and RNA-Seq of both long and short molecules . Other potentially involved regulators of other transcripts are microRNAs ( miRNAs ) , ∼22 nucleotides small non-coding RNAs processed by Drosha and Dicer from larger pri-miRNA molecules ( the initial miRNA transcript ) and pre-miRNA ( i . e the 65–70 nucleotide hairpin ) [49] . Binding of mature miRNAs to RNA-induced silencing complexes ( RISC ) is followed by guidance to target cognate protein-coding mRNAs , largely through identification of ‘seed’ matches ( sequences complementary to positions 2–8 of the miRNA ) in the 3′ un-translated region . The miRNA-RISC complex then initiates a program for mRNA degradation or block of translation . Of note , miRNAs can operate in tandem , cooperatively , or without an apparent seed sequence match . Since the first report of miRNA involvement in PD [39] , new findings provide ample evidence for involvement of differentially expressed miRNAs in the PD brain [40] [41] [42] [43] [44] [45] [46] . Much of these changes are reflected in PD leukocytes , where we recently observed by microarray analyses coupled miRNA-Alternative Splicing ( AS ) modifications that were modulated by DBS [37] . Thus , leukocyte datasets provide useful resources for studying PD-related processes , and new tools are required for linking between these different datasets . In our current study , we employed whole-transcriptome RNA sequencing and a newly developed workflow that includes comprehensive RNA-Seq analysis approaches to characterize all of the leukocyte-expressed protein coding transcripts as well as the non-coding class of lncRNAs in PD patients and control volunteers . We identified splicing changes , as well as novel exons and junctions by implementation widespread gene- , exon- and junction-level analyses . We implemented a variety of differential expression and splicing analysis methods including linear regression , Splicing Index ( SI ) , ASPIRE and FIRMA . This enabled an integrated analysis of exons and junctions ( both separately and combined ) , transcript structural variations and functional processes , and allowed the conduction of integrated RNA-Seq analysis of whole transcriptome data . We also implemented a new module that enables identification of known and novel Poly-A sites . Our comprehensive novel RNA-Seq analysis workflow further enabled the identification of specific protein domains translated from sequence regions detected as spliced , as well as potential miRNA binding sites within the detected regions . We applied this vast range of analysis tools on PD leukocytes from PD patients ( pre-DBS ) and post treatment ( post-DBS ) while being either on- or off-electrical stimulation , all as compared to matched controls the Amygdala and SN . The experimental and analysis workflow is illustrated under Figure 1 . Additionally , we used a publically available junction array datasets to characterize knock out effects of two PD leukocyte modified genes that are involved in additional neuropathologies , both include prion-like domains: FUS and HNRNP A/B [50] . We further analyzed an additional independent RNA-Seq dataset from PD brain samples and compared the PD blood and leukocyte modified lncRNAs to the brain modified ones following the full characterization of PD brain expressed lncRNAs in the external dataset . Experimental quantitative reverse-transcription polymerase chain reaction ( qRT-PCR ) tests validated exemplary findings both in patients' leukocytes and two brain regions from an additional set of PD brain samples as compared with unaffected control brain samples .
So far , only a few RNA-Seq studies have detected or analyzed lncRNAs [55] . We took advent of our next generation RNA-Seq data to identify all of the leukocyte-expressed lncRNAs and to search for differentially expressed lncRNAs in disease- and post-surgical treatment state . For this purpose , we mapped all of the sequenced reads detected in leukocytes from PD patients pre- and post-DBS and from matched healthy control ( HC ) volunteers against the largest currently available lncRNA database , the GENCODE database version 7 [56] . This database consists of reconstructed transcript models using Exonerate [57] and Scripture [58] , and is based on the high-depth transcriptomic data from 16 human tissues made publicly available from Illumina [56] . The current version of the GENECODE database covered 14 , 880 partially identified lncRNAs based on their chromatin signatures or position relative to protein coding genes . Alignment of the RNA-Seq leukocyte data to GENCODE revealed an average of 6 , 209 leukocyte expressed lncRNAs ( across all the leukocyte sequenced libraries ) , of them as many as 5 , 862 were present in all the libraries . Generally , PD patient leukocytes at all clinical stages ( either pre- or post- DBS treatment , and comparing on- and off- electrical stimulation ) contained less lncRNAs than in HC ( Figure 2A ) . We applied differential expression analysis of all the lncRNAs that were expressed in all the libraries using EdgeR [59] , [60] . The analysis used an over-dispersed Poisson model ( to account for both biological and technical variability ) and Empirical Bayes methods ( to moderate the degree of over-dispersion across transcripts and improve the reliability of the results ) . The normalization approach employed an empirical strategy that equates the overall expression levels of genes between samples under the assumption that the majority of them are not differentially expressed [59] , [60] . Overall , the analysis revealed 596 PD leukocytes modified lncRNAs ( p<0 . 05 ) . Briefly , the fold change tagwise dispersion plot of the lncRNAs detected in PD patient leukocytes was slightly skewed towards positive log fold change , indicating a general up-regulation trend in PD leukocytes ( Smear plot is illustrated under Figure 3A ) . The biological coefficient of variation ( BCV ) and multidimensional scaling clustering ( MDS ) plots are given under Figure S1 . Among all of the PD leukocyte altered lncRNAs compared to HC ( with uncorrected p<0 . 05 ) , 13 passed FDR correction ( FDR<0 . 05 , Table 1 ) . Overall , 11 of them were annotated by GENCODE 7 . 0 as novel entities . Those were well supported by locus-specific transcript evidence or evidence from a paralogous or orthologous locus while not being currently represented in the two databases of HUGO Gene Nomenclature Committee ( HGNC ) database [61] and RefSeq [62] . Two of the altered lncRNAs were annotated as known two additional databases , and overall −8 of the altered lncRNAs were on the sense and 5 on the antisense strand . The majority of the lncRNAs found to be differentially expressed in PD leukocytes belong to the novel multi-exon RNA processing ( RP- ) lncRNAs family , with four of them showing locus conservation with zebra fish and one with mouse [63] . One of the disease leukocyte-altered lncRNAs ( RP11-124N14 . 3 , transcript name RP11-124N14 . 3-001 ) showed high abundance ( with an average count level of 2 , 460 in all the sequenced libraries ) , whereas the rest of the lncRNAs found to be differentially expressed in PD showed middle to low abundance levels ( hundreds-to numerous counts; Table 1 ) . The PD leukocyte-altered lncRNAs further had different transcript size , some shorter ( for example , RP11-533O20 . 2 of 2 exons and non-conventional length for lncRNA of 161 nucleotides ( nt ) ) , some with middle length ( e . g , RP11-462G22 . 1 – of 879 nt ) and some longer ( e . g U1 , with length of 1 , 548 nt and RP4-705O1 . 1 – 1 , 518 nt ) . Notably , one of the DBS up-regulated lncRNAs , RP11-120K24 . 2 , was reported to be up-regulated in the brain of Autism disorder patients [64] . The disease-mediated increase observed by RNA-Seq analysis in U1 and RP11-462G22 . 1 was faithfully validated by real-time RT-PCR in the PD leukocytes ( Figure 2D , one-tailed t-test p = 0 . 049 for RP11-462G22 . 1 and RPL19 as reference gene , non-significant for U1; details under Methods ) , which further raised the question of the relevance of the observed leukocyte alterations to the PD brain degenerative process . Next , quantitative RT-PCR ( qRT-PCR ) in two PD-related brain tissues , the Amygdala and the SN , validated the leukocyte PD observed increase of potential ceRNA RP11-462G22 . 1 ( Figures 2E and 2F , two-way ANOVA p = 0 . 049 , TUBB3 served as a reference gene ) . Validation in both brain regions by qRT-PCR of the disease RNA-Seq detected differential expression of the lncRNA RP11-79P5 . 3 ( LncBTF3-4 , which is conserved in the zebra fish ) have confirmed its observed PD leukocyte up-regulation in PD brains as well ( Figure 2E and 2F; two-tailed t-test p = 0 . 03 , TUBB3 served as a reference gene ) . The increase of U1 , the second leukocyte altered lncRNA which was detected as altered in PD leukocytes through leukocyte RNA-Seq analysis and was predicted to target numerous miRNAs , was confirmed by qRT-PCR in PD leukocytes ( Figure 2D ) . In the brain , it was detected by PCR only in the Amygdala ( and not in the SN ) , where it also exhibited elevation under PD as in PD blood leukocytes ( Figures 2D and 2F ) . The DBS treatment induced differential expression changes in lncRNAs . ( Figure 3B ) . Overall 663 lncRNAs ( uncorrected p<0 . 05 ) were modified post-DBS as compared with the pre-DBS ( disease ) state of the same patients exhibiting mainly post- treatment down regulation ( Smear plot is illustrated under Figure 3B ) . While 428 lncRNAs decreased post- compared to pre-DBS , only 235 increased post-treatment . Thus , overall , the DBS expressed a wider effect on lncRNAs as compared with the disease . The treatment induced opposite global direction of change compared to the disease , which mainly induced leukocyte lncRNA increases . The full list of differentially expressed lncRNAs with RNA-Seq count values is given under Table S2 ( MDS and BCV plots are illustrated under Figure S2 ) . Of the total number of DBS modified lncRNAs , 18 passed FDR correction ( Figure 2C ) . Of the highly significant treatment-altered lncRNAs ( FDR<0 . 05 , Table S2 ) , 9 were on the sense and 9 on the antisense strands . Of the DBS-modified lncRNAs , 14 showed a decrease and only 4 were increased post- as compared with pre-DBS ( Figure 2C ) . Four of the lncRNAs that were modified post- as compared to pre-DBS in the blood leukocytes were among the top PD leukocyte modified lncRNAs ( as shown in Figures 2B and 2C , color highlighted ) : RP4-705O1 . 1 , RP11-533O10 . 2 , RP11-425I13 . 3 and RP11-79P5 . 3 ( of which disease increase was validated by qRT-PCR in patients pre-DBS compared to HC ) . All the disease- and treatment- shared lncRNAs showed inverse direction of expression changes post- as compared with pre-DBS . The short one hour electrical stimulation cessation ( OFF-stimulus ) induced very mild alteration of lncRNA expression in the patients' leukocytes , as it did not induce any lncRNA differential expression that passed FDR correction ( Fold change plot under Figure 3C , MDS and BCV plot under Figure S3 ) . Nonetheless , 110 LncRNAs showed uncorrected significance level of p<0 . 05 ( Table S2C ) , completely balances in terms of down- and up-regulation – 55 lncRNAs exhibited leukocyte decrease upon stimulation cessation , and 55 increase . These included two molecules that were annotated by GENCODE as putative lncRNAs - transcripts that contain 3 or fewer exons and are supported by 1 or 2 Expressed Sequence Tags ( ESTs ) , but not 3 . To further challenge the significance of PD leukocyte modified lncRNAs , we analyzed an additional , independent recently released RNA-Seq dataset from post mortem brain samples of PD patients and healthy donors ( Array expression accession number E-GEOD-40710 ) . This dataset was composed of the 3′ UTR of polyadenylated mRNA sequencing data ( PA-Seq ) of transcripts from cortical tissue samples of PD patients and unaffected controls [65] . We selected 6 PD and 6 age- and gender- leukocyte-matched ( males ) non-affected samples from this dataset for analysis . MDS plot ( illustrated under Figure S4 ) revealed that two of the unaffected control samples as potential outliers , and therefore we excluded these samples from further analysis , and analyzed for disease differential expression only the remaining 6 PD and 4 unaffected individual samples from the external PD brain RNA-Seq dataset ( count values are given under Table S3A ) . Overall , a larger number of lncRNAs were expressed in the brain as compared with blood leukocytes ( n = 7 , 189; Table S3A ) . Of these , 3 , 495 lncRNAs were also among the lncRNAs that were detected as expressed in PD blood leukocytes either pre- or post-DBS as well as in control ( HC ) leukocytes samples ( Table S3B ) . Differential expression analysis of the PD brain-expressed lncRNAs ( Fold change smear plot based on the tagwise dispersion analysis is given under Figure 3D ) identified 242 lncRNAs as significantly altered in PD brains ( FDR<0 . 05 , Table S3C ) . Of these , 181 were up-regulated and only 61 were down- regulated in the PD brain samples ( Figure S4 and Figure 3D ) . Of the overall 963 lncRNAs that were significantly altered in the PD brain RNA-Seq dataset ( uncorrected p<0 . 05 , Table S3D ) , 569 ( 59% ) were also detected as expressed in the PD leukocyte RNA-Seq dataset . Of these , 135 have passed FDR threshold brain dataset ( Table S3D ) . These included 2 of the lncRNAs that were significantly changed in PD patients leukocytes compared with HC leukocytes ( FDR<0 . 05 ) : RP13-507P19 . 2 and RP11-79P5 . 3 , which were validated by qRT-PCR in both PD leukocytes and independent PD and unaffected control brain samples that were obtained from the Netherlands Brain Bank ( NBB ) ( Figure 2D–2F ) . On the other hand , of the 13 lncRNAs that changed in PD leukocytes pre-DBS compared to HC , 7 were also detected overall as expressed in the independent PD brain RNA-Seq data set , 2 of which were also altered in the PD brain RNA-Seq dataset ( ( p<0 . 05 , Table S3D and Table 1 ) . We have validated one of these ( RP11-79P5 . 3 ) by qRT-PCR in both the leukocytes and the independent PD brain sample set in both the amygdala and SN . Of all of the PD leukocyte-modified lncRNAs ( having uncorrected significance level of p<0 . 05 ) , 59 were significantly altered in the brain dataset as well ( brain RNA-Seq data uncorrected p<0 . 05 , Table S3D ) . Taken together , these findings highlight the relevance of the leukocyte-differentially expressed lncRNAs to the PD degenerative process overall . Since lncRNA functions are believed to be closely related to their secondary structures [66] [67] , we applied a stochastic sampling method of structure prediction which maximizes the expected accuracy of the prediction [68] . The differentially expressed lncRNAs emerged as able to form complex stem-loop secondary structures . The secondary structure of U1 ( also termed lnc-SPATA21-1 ) entailed a large number of stem-loop structures ( Figure 4A ) ( predicted structure computed with [69] , see methods for details ) . Two of the disease-altered lncRNAs , U1 and RP11-462G22 . 1 were predicted to target miRNAs and thus are potential competitive endogenous RNAs ( ceRNAs ) . Using a support vector machine-learning algorithm ( details under methods ) , U1 was predicted to bind 8 different miRNAs ( listed under Table 1 ) . These included hsa-miR-188-3p which controls dendritic plasticity and synaptic transmission [70] , as well as hsa-miR-125b , which promotes neuronal differentiation and inflammation in human cells by repressing multiple targets [71] ( Figure 4A , enlarged ) . RP11-462G22 . 1 ( also termed lnc-FRG1-3 ) exhibited a yet more complex secondary stem-loop structure with a larger number of loops compared to U1 and was predicted to target 21 different miRNAs , identifying this lncRNA as well as a potential ceRNA ( Table 1 ) . These included the mitochondrial calcium import regulator hsa-miR-25-5p ( complementary sequence enlarged under Figure 4B ) . Similarly to other lncrRNAs , which are generally poorly conserved between species , both U1 and RP11-462G22 . 1 were not found neither in the mouse nor in the zebra fish genomes [63] . Identifying the U1 spliceosomal lncRNA as disease- and treatment-altered in PD patients' leukocytes called for exploring possible concurrent splicing and transcript structure modifications in the same patients RNA samples . While we have previously reported alternative splicing alterations in PD leukocytes pre- and post-DBS through exon and junction microarray analyses [47] , [48] , RNA-Seq data analysis offers several important advantages in expression studies . Being independent of predefined probes ( as compared with microarray platforms ) , RNA-Seq allows unbiased identification of novel junctions , exons and transcript isoforms as well as altered polyadenylation choices at high resolution . The in-depth coverage allows a more accurate measurement of exon inclusion or exclusion and identification of lowly abundant junctions and exons . Moreover , this analysis further enables detection of reciprocal pairs of junctions , one that includes and the other that excludes an exon . To fully exploit these virtues , we developed an updated version of AltAnalyze ( version 2 . 0 ) for RNA-Sequencing analysis ( http://www . altanalyze . org/ ) . Extending the program that was initially developed and introduced for junction microarray analyses [37] , [72] , we introduce a significantly improved workflow ( Figure 1 ) . This is reflected in its user-friendly pipeline for the analysis of both known and novel splicing events directly from SOLiD BioScope processed RNA-Seq results ( but also from other platforms , such as TopHat [73] , HMMSplicer [74] and SpliceMap [75] ) . The obtained analysis results can be directly integrated with alternative exon or junction array datasets . Unlike existing RNA-Seq analytical pipelines , this new version of AltAnalyze can directly evaluate differential gene expression , identify known and novel alternative exons and junctions and alternative Poly-A sites , perform combinatorial exon and junction analyses and evaluate these effects at the level of protein domains , miRNA targeting and enriched biological pathways , as a fully automated and user-friendly pipeline . We have developed a full analysis scheme to analyze using a wide range of analysis approaches gene , exon and junction levels of count data . This tool is able to analyze count information that is obtained from various platforms and analysis methods of data produced by RNA-Seq experiments ( Methods ) . We have applied the full scheme of the new version of AltAnalyze ( version 2 . 0 ) on leukocyte RNA-Seq data from PD patients pre- and post-DBS and matched HC ( Figure 1 ) . Splicing-Index ( SI ) algorithm ( described in detail under [76] and [77] ) was applied on the PD leukocyte RNA-Seq data based on the above-described novel targeted RNA-Seq module of AltAnalyze . The SI analysis enabled us to detect 1 , 652 alternatively spliced exons in 1 , 221 distinct genes ( Table S4A and S4B ) in PD as compared with HC , including such that belong to the splicing factor HNRNPF . The SI exon level analysis based on RNA-Seq read counts yielded many more splicing products in PD leukocytes' RNA than was previously detected by both our PD leukocyte microarrays ( exon and junction ) analyses [78] . Notably , 34 modified genes were detected in both the RNA-Seq and microarray technologies , including the elongation factor EIF2AK3 and the interleukin receptor IL1RL1 . Also , the majority of disease-induced changes , 1313 were exon inclusion ones and only 339 – exclusion events , similar to the enrichment of inclusion events seen in cortical samples from Alzheimer's disease ( AD ) patients [79] . To further identify high confidence splicing events , we implemented linear regression analysis approach for analyzing all the reciprocal junction pairs detected in the RNA-Seq libraries ( i . e reciprocal junction pairs: all the pairs of junctions one of which includes an exon in the transcript and one excludes it ) . Following sample-specific database construction of all the known and novel junctions available in the samples , the test is performed on all the junction pairs detected in the RNA-Seq libraries ( both novel and known ) . Using this novel RNA-Seq implemented analysis approach on the sequenced PD leukocyte RNA , we could identify all the detected known and novel splice junctions . The novel RNA-Seq linear regression tool of the AltAnalyze RNA-Seq module subsequently identified 315 reciprocal junction pairs ( Table S4C ) as significantly changed in the disease , the majority of them ( 228 junction pairs ) identified as novel by de-novo predictions that followed dataset-specific database construction ( enabled in the new RNA-Seq adapted version 2 . 0 of AltAnalyze ) . Only 87 of the disease-altered junctions existed in current genomic databases , and thus the majority of these junctions could not be previously detected by microarray analyses . Of the total changed junctions , 157 and 158 spliced junction pairs induced exon inclusion and exclusion events in 144 different genes ( Table S4D ) , suggesting that unique splicing events occur under PD . Some events were , however annotated as affecting more than one type of transcript structure variation , yielding an overall of 179 such PD-associated annotations . These primarily involved alternative cassette exons ( N = 142 , Figure S6 ) . Other types of detected events included alternative 3′-ends ( n = 11 ) , alternative 5′-ends ( n = 7 ) , alternate promoters ( n = 5 ) , mutually exclusive exons ( n = 2 ) and 2 trans-splicing inducing splice junction pairs events . To enable a more rigorous detection of alternative exons , we implemented and applied for RNA-Seq the Analysis of Splicing by Isoform Reciprocity ( ASPIRE ) approach [80] through the novel RNA-Seq version of AltAnalyze ( version 2 . 0 ) . This identified pairs of alternative exons and reciprocally expressed exclusion junctions through combined junctions and exons analysis based on RNA-Seq read counts for both types of gene structures . Subsequent analysis of the corresponding read counts from leukocyte RNA of PD patients compared to control samples identified 105 inclusion and 88 exclusion events in 192 exon-junction pairs ( Table S4E ) . Exon-level SI analysis of RNA-Seq libraries from RNA samples of patients pre-DBS compared to matched healthy control volunteers detected 1353 alternative events , in 1 , 069 distinct genes ( Table S4F ) . Of these , 748 were inclusion and 320- exclusion events; extending our previous reports of widespread influence of DBS on splicing direction . Linear regression analysis showed DBS-induced massive reduction in spliced junction pairs , particularly the fraction of novel ( compared to known ) junctions in which splicing events were detected ( Figure 5A ) . Hierarchical classification ( HCL ) of the PD and HC samples based on the expression of spliced junction pairs in the leukocyte RNA-Seq read counts , correctly classified patients from control samples ( Figure 5B ) . The resolution of RNA-Seq is higher compared to microarrays , and the technology enables detection of novel gene structural elements and events . Nevertheless , changes in 20 of the RNA-Seq disease detected leukocyte genes were also detected as altered under PD in our previous exon array analysis of a larger cohort of PD patients pre- and post-DBS [47] , [48] . These included the motor movement disorder dystonia related LRRC16A gene [81] . Linear regression junction level analysis of all the reciprocal junction pairs expressed in the RNA-seq data from the same patients , post-DBS treatment as compared with the RNA-Seq leukocyte data from the pre-DBS ( disease ) state revealed 137 pairs of AS changes ( Table S4G ) . Of these , 85 were un-annotated in current genomic databases thus consisting of novel junction pairs ( Figures 5A ) . The DBS alternatively spliced junctions were structurally a part of 73 different distinct genes ( Table S4H ) . The detected events consisted of 53 exon-inclusion and 84 exon-exclusion events . The RNA-Seq read count of the DBS- altered reciprocal spliced junction pairs correctly classified patients post- from pre- surgical state ( Figure 5C ) . We implemented the robust ASPIRE [82] analysis approach for RNA-Seq data analysis and applied it on the RNA-seq data of patients' leukocytes post- compared to pre-DBS on stimulation while combining both exons and junctions count data for the analysis . This revealed 108 AS changes , 49 of those representing inclusion and 59 - exclusion events ( Table S3I ) . These evented occurred in 17 genes that were also detected in our previous FIRMA analysis of exon microarrays data of a larger cohort of patients , including the inflammatory mediator IRAK1 [83] . Functional analysis using the AltAnalyze Gene Ontology ( GO ) [84] Elite module ( GO-Elite ) adopted for RNA-Seq detections highlighted enrichment in response to oxidative stress , ribonucleoprotein binding , transcriptional repression and histone methylation ( Table S4O ) . Exon level analysis using SI measure of the RNA-Seq read counts revealed drastically reduced splicing changes following one hour of electrical stimulation cessation ( OFF-Stim ) compared to PD leukocytes from pre-DBS patients in 865 exons ( Table S4J ) of 778 genes ( Table S4K ) . These included an inclusion event in the mitochondrial matrix gene Sirt3 , which was recently found to induce aging-associated degeneration [85] . 496 ( 63% ) of the off-stimulus detected alternative splicing events were exclusion events . Exon-level SI analysis detected changes in 11 genes that were previously detected by us in PD patient leukocytes through exon microarray analyses of a larger PD cohort ( including the sequenced samples ) [47] , [48] , including the ubiquitin-specific protease regulator USP13 . Linear regression analysis of all the reciprocal splice junction pairs found as expressed in the OFF- compared to ON-stimulus leukocyte mRNA-Seq data revealed 81 junction pairs as changed ( Table S4l ) in 36 genes ( Table S4m ) ; Of these , 68 were novel junctions , and 14 were known ( Figure 5A , right bar graph ) . The altered junction pairs correctly classified leukocyte RNA from OFF-stim and ON-stim samples from one hour earlier ( Figure 5D ) . A combined analysis of both exons and junctions quantified by the ASPIRE analysis yielded 70 OFF-stim induced junction level splicing changes , which included 28 inclusion and 42 exclusion events ( Table S4N ) . Functional analysis through the GO-Elite module of AltAnalyze detected enrichment in immune effector process , natural killer cell proliferation , cytokine production , protein transport and regulation of response to stress ( Table S4P ) . Splicing modifications , and especially those affecting the 3′-untranslated region ( 3′-UTR ) could potentially modify miRNA-binding sites . Implementing a module detecting miRNA enrichment in the novel version of RNA-Seq adopted program of AltAnalyze ( details under http://www . altanalyze . org/ ) , we identified potential miRNA binding sites in the regions detected through both junction- and exon-level splicing analyses ( either through a single feature – exon/junction- or a dual feature , i . e pairs of junctions/exons ) . Enrichment analysis for potential miRNA binding sites in the genes detected as alternatively spliced by junction-level linear regression analysis of PD leukocyte RNA-Seq read counts compared with HC revealed 20 potential miRNA target sites in these genes ( Table S5A ) . These spanned the two forms of hsa-miR-133 , previously linked to PD ( hsa-miR-133a , predicted to bind the GTPase GSN and the cancer-linked FRG1B [86] , and hsa-miR-133b , also predicted to bind FRG1B . Enrichment analysis of miRNA binding sites in transcripts detected as alternatively spliced by exon level SI analysis of leukocytes from PD patients pre-DBS compared to controls detected 364 miRNA-target binding predictions ( Table S5B ) , including the synaptic plasticity human hsa-miR-188 and the inflammation controlling hsa-mir-125 . Enrichment analysis for miRNA-binding sites in the DBS-spliced genes detected by SI analysis predicted 481 such sites ( Table S5C ) . These included predicted binding of spliced genes to three forms of hsa-miR-376 ( a–c forms ) , hsa-mir-544 targeted at the same spliced gene ( ITGAL ) and the inflammation-related hsa-mir-150 [87] predicted to target the disease spliced gene FGD4 . In comparison , the novel RNA-Seq linear regression AltAnalyze module identified only 5 miRNA-target pairs in DBS-modified transcripts ( Table S5D ) . Also , the differentially spliced junction pairs detected by linear regression analysis showed no enrichment in miRNA binding sites when comparing post-DBS to one-hour stimulation cessation . Nevertheless , the splicing-index genes detected based on exon-level SI analysis of leukocyte read counts from patients post-DBS off stimulation compared to the on state exhibited 262 such predictions ( Table S5E ) . Splicing modifications may further modify human protein binding domains . To assess the potential impact of the detected splicing changes on protein interactions , we tested for over representation of protein domains using z-score calculations ( details under http://www . altanalyze . org/ ) for all the identified PD-related splicing events in the detected regions . The SI exon level analysis of PD compared to controls yielded 311 statistically significant ( adjusted ) changes in domain-target pairs ( Table S6A ) , including metal binding domains and ubiquitin motifs . Similar analysis at the linear regression junction level of patients pre-DBS samples compared to HC yielded 16 statistically significant ( adjusted ) changes in domain-target pairs ( Table S6B ) ( of a total of 12 , 529 domains analyzed ) , including metal-binding domains . Examples include the immune complement component ITGAX ( also called CD11C ) and ITGAM ( also called CD11B ) . The DBS treatment induced changes in 351 domains in the transcripts detected by SI analysis of the RNA-Seq samples of patients post-DBS on stimulation compared to the pre-DBS state ( Table S6C ) and 11 domains in the transcripts detected by the linear regression junction level analysis ( having adjusted p<0 . 05 ) ( Table S6SD ) out of a total of 12 , 539 protein domains detected in the RNA-Seq libraries . In contrast , the short one hour electrical stimulation cessation induced 281 predicted changes in functional binding domains ( after p-value adjustment ) in the linear regression detected junction pairs ( Table S6E ) and 44 enriched binding domains were detected by SI analysis on RNA-Seq detected exons that were modified upon electrical stimulation cessation post-DBS ( Table S6F ) . Mutations in prion-like domains of the splicing regulator heteronuclear ribonucleoprotein HNRNP A/B were recently linked with Amyotrophic lateral sclerosis ( ALS ) and rare proteinopathies [88] . Intriguingly , we detected PrLD domains that have RNA recognition motif ( RRM ) in 8 genes that were identified as undergoing alternative splicing modifications ( in either exons , junctions or both ) in PD patients leukocytes pre-DBS as compared with healthy control volunteers . Mutations in the Prion-like domains of three of these ( FUS , EWSR1 and TAF15 ) ( Supplementary Figure S5 and Table S7A ) were recently reported as causally involved in other neuropathologies and human degenerative proteinopathies [50] . The Prion-like domains in PD-spliced transcripts included HNRNP A beta ( HNRNP A/B ) ( Supplementary Figure S5 ) . Correspondingly , analysis through the SI module for splice junction arrays by AltAnalyze of RNA samples produced from primary neurons of mice with ablated ( KO ) FUS or HNRNPA1 [89] detected 147 exclusion events ( Table S7B ) . The identified genes included IMPA2 , which is associated with schizophrenia and bipolar disorder [90] , the enolase NO1 which is differentially expressed under stress [91] , Mclf2 , a rho guanine nucleotide exchange factor that interacts with the mental retardation and autism related gene interleukin-1 accessory protein-like 1 ( Il1RAPL1 ) , TMPR555 , a trans-membrane serine protease whose presynaptic distribution on motor neurons in the spinal cord suggests an important role in neural development [92] and the JNK signaling pathway activator TCEA3 [93] . Additionally , splicing alterations of HNRNPA1 were previously associated with selective loss of HNRNP A/B and with massive exon inclusions in AD entorhinal cortex , and lentiviral-mediated suppression of HNRNP A/B impaired electrocorticography in the mouse brain [79] . ASPIRE analysis detected splicing changes spanning 151 transcripts in a splice junction microarray dataset of HNRNP A/B silencing of human embryonic kidney cells [94] . The affected genes included the nucleosome stability histone H2A , BCL2 which is involved in striatal neurons and considered to be a compensatory mechanism in PD [95] , MLLT10 involved in lymphoblastic lymphoma [96] , Chorod1 involved in brain development [97] , the potential neuro-protector PON2 [98] and the iron homeostasis involved gene FBXL5 [99] . Of the genes detected as undergoing AS changes upon electrical stimulation cessation , 15 belonged to the proteins family that contains the RNA Binding motif RBM33 and included Prion-like domains or RNA Recognition Motifs ( RRMs ) . These spanned RBM5 , RBM19 , RBM25 and RBM39 , among others ( Table S7A ) . In the RBM5 , RBM19 and RBM25 genes , the prion-like domains were found as present both in altered exons as well as in exon-junction boundaries . Overall , six disease-spliced junctions were included in prion-like domain regions , including in the FUS and RBM33 genes , and 3 of the off-stimulus AS genes showed enrichment in prion-like protein domains ( Prion-IPR000817 ) . Functional enrichment analysis of the 114 disease-detected transcripts identified by the ASPIRE analysis on both exon and junction quantifications served to explore disease-related pathways . The analyzed transcripts were highly enriched in immune system pathways , including regulation of leukocyte-mediated immunity ( Figure 6 ) as well as disease-related pathways such as nuclear transport , regulation of GTPase activity and synaptic transmission . Other affected pathways include protein import into the nucleus , known to be impaired in degenerative proteinopathies due to mutations in HNRNP A/B [50] as well as regulation of the neuro-immune CDC42 Rho GTPase; in the brain , CDC42 binds to collybistin and participates in bringing GABAergic receptors to anxiolytic synapses [100] , whereas in lymphocytes it regulates cell division [101] , perhaps explaining part of the immune mal-functioning that is a characteristic PD phenotype . The disease effect at the transcript level was estimated based on structural elements for the global population of human genes . For this purpose , all the EnsEMBL ( 66 ) and UCSC ( 65 ) mRNA transcripts were compared to each aligned read identified in the RNA-seq samples . De-novo predictions for transcript level structure of all the human genome annotated genes using the constructed AltAnalyze database detected 633 , 054 known exons ( of a total of 705 , 345 detected ones in the RNA-Seq samples ) . The detected exons were located in 3′ un-translated regions ( 3′-UTR ) , 5′-UTR , C- and N-termini , as well as in rare structural elements ( such as nonconventional AT/AC ending introns ) . Overall , of 633 , 054 previously known and overall 705 , 345 exons detected in all the sequenced libraries , 343 , 400 exons ( 48% , Figure 7A , pie illustration on the left side ) remained un-annotated at the level of transcript structure . The exons detected in all the RNA-Seq samples were primarily composed of cassette exons ( 33% , Figure 7A , left pie ) . The rest of the annotated exons ( overall 18% of all the detected exons ) were functionally annotated to 13 different transcript-level splicing structures ( Figure 7A , right pie ) . These included alternate 3′ intron ends ( 3% ) , alternate 5′ intron ends ( 2% ) , alternate N terminals ( 2% ) , alternate C termini ( 1% ) , alternate promoters ( 2% ) , exon region exclusion ( 1% ) and 580 last exons in the transcripts . 654 of the samples expressed exons detected in the RNA-Seq libraries were linked to strange intron ends ( not GT/AG , GC/AG or AT/AC ) and 44 to AT/AC intron ends . 1% of all the detected exons were bleeding exons ( initiating or terminal exons that overlap with an intron of another transcript ) and 1% were mutually exclusive expressed exons . Overall , 2% of the exons were linked to intron retention events and 3% - were located in alternative Polyadenylation ( Poly-A ) sites ( Figure 7A ) . Overall , the exons in samples of PD leukocytes showed significantly different frequencies of transcript functional relevance as compared with healthy samples ( Figure 7B , light blue bars ) . A goodness of fit ( Chi-square ) test yielded a statistically significant difference between the distributions of specific events in the disease at the transcript level annotation ( compared to the global population of exons detected in all the sequenced samples ) for all the event types ( except for alternate 3′ intron ends ) ( Figure 7B , black and dashed lines ) . The post-DBS patient leukocyte samples showed lower proportions of last transcript exons , strange intron ends , bleeding exons and cassette exons as compared with the pre-DBS ( disease ) state ( Figure 7B , orange bars ) . In contrast , the proportion of exons annotated with other types of transcript level events ( including alternate promoters and intron retention ones ) increased following DBS ( Figure 7B , orange bars ) . The short period of one hour of electrical stimulation cessation reduced the proportions of all the event types , as compared to the stimulated state ( and in some cases , also compared to the pre-DBS disease state ) . However , the proportion of alternative 3′ intron ends was increased ( Figure 7D and Table S8 ) . The events with reduced proportions upon stimulation cessation included last transcript exons , strange intron ends , bleeding exons , alternate C and N termini , alternative 5′ intron ends and intron retention sites . The proportion of cassette exons remained similar in the off- compared to the on- stimulus state ( but lower compared to the disease state in both ) . Alternative Polyadenylation ( Poly-A ) predictions were incorporated into the novel version of AltAnalyze , which was adopted for RNA-Seq analysis and enabled reporting transcript event annotations . Exon regions overlapped with Poly-A binding sites that underwent alternative splicing modifications . A targeted analysis of the Poly-A sites in the sequenced libraries revealed increased alternative Poly-A site choices in PD leukocytes as compared with normal controls ( Figure 7C , middle plot and Table S4 ) . This increase was attenuated by the DBS treatment yet was largely regained ( to even a higher proportion than in the disease ) following one hour OFF stimulation ( Figure 7C ) . It was previously shown that complex alternative RNA processing generates unexpected diversity of poly-A polymerase isoforms [102] , which might be the case observed in the PD leukocytes RNA-Seq data .
We present here a comprehensive approach to analyze whole-transcriptome RNA-Seq data obtained via various platforms , using measurements of both splice junctions and exons , independently and in combination through various analysis methods , which enable identification and analysis of both know transcript variant as well as novel ones . The workflow enables additionally identification of transcript structures modifications , and integration with protein binding sites and microRNA annotations . We have re-implemented a diverse set of splicing-directed analysis methods ( ASPIRE , linear regression , FIRMA and splicing-index ) that were originally developed to analyze splice-sensitive microarray data [51] [103] [104] , for the analysis of complex RNA-Seq data of protein-coding transcripts . The new RNA-Seq analysis workflow enables de-novo identification of genome-specific transcript structures through sample specific database construction based on the experimental specific read counts . We also incorporated for the first time prediction of Poly-A sites in the novel AltAnalyze version described here ( version 2 . 0 ) . This workflow enabled us to detect novel exons and junctions in protein coding RNA molecules , as well as a large range of splicing events under PD pre- and post- brain stimulation both on- and one hour off- electrical stimulation ( which re-induces the disease motor symptoms ) , as compared with healthy control volunteers . We also present here a workflow for detection and differential expression analysis of lncRNAs in whole transcriptome RNA-Seq data . Together , the novel analysis workflow and unique RNA-Seq dataset enabled us a widespread analysis of differential splicing as well as to detect lncRNAs and characterize their differential expression in both the disease and treatment states . At the transcript level , the DBS-induced increase in alternative ends , as well as in intron retention and alternative promoter usage , was accompanied by a 50% decrease in the number of ‘bleeding exons’ ( that ‘leak’ into other transcripts ) . The number of cassette exons ( present in certain transcripts but not in others ) was predictably highest among all the possible types of splicing events in all the sequenced samples . Specifically , we observed increase in the frequency of cassette exons and intron retention events both in the disease and following DBS , as compared with the global population of expressed exons . Notably , non-conventional AT/AC and GT/AG ending introns were predictably very rare , in all the tested clinical conditions , as compared with the other types of transcript structural variations and disease-modified Poly-A choices . Our deep survey characterized leukocyte-expressed lncRNAs in both patients and control volunteers and identified 5 lncRNAs that are over-expressed in the disease and inversely decrease following DBS . These include the spliceosome component U1 , supporting the notion of disease-involved splicing modulations . Also , increased levels of the muscular dystrophy-associated RP11-462G22 . 1 ( lnc-FRG1-3 ) may be relevant to the muscle rigidity in PD , one of the six disease hallmark motor symptoms . Another disease-modified lncRNAs that decreased post-DBS ( RP11-79P5 . 3 ) was also found as differentially expressed by analysis of an additional external , independent PD brain RNA-Seq data-set [65] and its disease up-regulation was successfully validated by qRT-PCR in the leukocytes as well as in two brain regions from an additional set of PD and unaffected control brain samples , in both the Amygdala and SN . So far , only a few large-scale studies have revealed fundamental characteristics of lncRNAs including their low levels of expression , temporal and spatial patterns of expression , sequence conservation and association with histone modifications [105] . Functional assays have also revealed diverse mechanisms through which lncRNAs act to regulate protein-coding genes at both the transcriptional and translational levels . However , to date there is insufficient data on the relationship between sequence , expression and pattern of newly identified lncRNAs [106] . The relatively low sequence and transcriptional conservation between species further complicate these studies . Yet , the identification of alternative , still unidentified features may produce a framework with which to accurately predict the functions of un-annotated lncRNAs [105] . An independent brain dataset analyzed in the current study exhibited a large number of lncRNAs commonly expressed in leukocytes from PD patients , thus we provide here an exceptionally rich resource for lncRNA expression in PD human leukocytes and brain regions . We recently profiled differentially expressed miRNAs of PD patients' leukocytes pre- and post-DBS by small RNA deep sequencing [107] , concurrently with alternative splicing changes of their predicted target genes . That study involved analysis by a junction array-adopted version of AltAnalyze . Here , we use the AltAnalyze target prediction module to detect potential miRNA binding sites within regions detected as undergoing splicing modifications by the RNA-Seq analyses , as well as putative protein binding domains . To detect disease and treatment-affected pathways , the splicing-sensitive results were re-analyzed using the functional AltAnalyze analysis module GO-Elite [108] for over-representation analysis ( ORA ) of pathways , ontologies and other gene sets . We believe that our current approach and results will provide a useful resource for biomedical researchers of movement and neurodegenerative disorders , and that our suggested analysis workflow may maximize the observations obtained by analyzing RNA-Seq data through simultaneous detection of novel junctions , exons and splice isoforms in a data-specific manner through comprehensive yet sensitive detection of alternative splicing events . Our lncRNA analysis workflow and results will also provide an important resource to the biomedical community . Currently , 31% of the human genome bases in sequenced transcripts are annotated as intergenic ( located between coding genes ) . Of these , lncRNAs are rapidly emerging as important and fascinating regulatory factors across a diverse catalogue of molecular , genetic and cellular processes , but phenotypic consequences of their differential expression , as well as sequence and structure derived functionality are still an Enigma . Here , in addition to comprehensive detection of both junction and exon level splicing changes in protein coding transcripts , we also fully characterized the disease- and treatment-expressed lncRNAs , and found large disease-induced expression changes in 13 lncRNAs ( of the over 6000 lncRNAs detected in the leukocytes overall ) , including such that are involved in RNA processing . We have validated the RNA-Seq observed disease alternations through real-time RT-PCR for three lncRNAs , including two potential ceRNAs predicted to bind numerous miRNAs . Although only recently detected , lncRNAs raise a great interest to the scientific community due to their tremendous influence on our perception of genes . It is clear now that they can function at the molecular level [109] , but their potential role in human neurodegenerative diseases was not reported yet . Certain lncRNAs function as transcriptional regulators of neighboring protein-coding genes by cis- or trans-modulation [110] , enhance or repress nearby protein-coding genes [109] , operate as epigenetic gene regulators through histone or DNA modification [111] ( for example , in muscular dystrophy ) [112] , and act as precursors or decoys for small RNAs [113] . Thus , the expression map of lncRNAs in human leukocytes and specifically , in PD patients' pre- and post- DBS treatment may become an important resource . Specifically , both miRNA-binding lncRNAs and splicing modulations have been demonstrated to impact miRNA binding site integrity , which has been proposed to be an important mechanism in regulating miRNA-RNA sensitivity [51] . Emerging evidence further demonstrates a role for lncRNAs in regulating both miRNA targeting [114] , possibly competing with the protein coding targets of the sponged miRNAs , and splicing factors [115] . For example , the lncRNA MALAT1 modulates SR splicing factor phosphorylation [116] , whereas miR-188-5p which is complementary to the PD-induced lncRNAs targets the alternative splicing regulatory factor SFRS1 ( SF2/ASF ) [117] ( which we previously reported as modified in PD patients through exon microarray analysis ) . Two of the PD differentially induced lncRNAs predictably bind many complementary miRNAs , and were further increased following DBS treatment . RP5-875O13 . 1 ( lnc-SPATA21-1 ) showed complementarity to 8 miRNAs and RP11-462G22 . 1 to 21 miRNAs , supporting the notion that lncRNAs may function in PD as protective decoys preventing the functioning of their complementary miRNAs . That miRNAs may present lncRNA-trapped , possibly non-functional versions , further suggests that quantifying miRNA levels in biological sources may be insufficient to predict their functioning potential . The DBS treatment potentially exacerbated this reaction , upon induction of changes in large number of lncRNAs . These two lncRNAs may hence belong to the newly discovered competitive endogenous RNAs ( ceRNAs ) lncRNA class , originally described as transcribed retropseudogenes that retain the miRNA-binding function of their parent mRNAs , which currently include lncRNAs [114] . CeRNAs have been proposed to function as miRNA ‘decoys’ or ‘sponges’ , thereby de-repressing levels of protein coding transcripts that share with the ceRNAs the same miRNA response elements [114] . Although ceRNA-mediated regulation represents an elegant mechanism by which lncRNAs may control protein function through miRNA mediators , the proportion of lncRNAs that act as ceRNAs remains unknown [55] . Of note , secondary sequence structures were so far not studied for lncRNAs , and our current observation for secondary structure enabling possible miRNA sponging for two of the disease differentially expressed lncRNAs calls for future studies involving lncRNA secondary structures predictions . Future comparative study of various species will provide further insights into structure-based functionality of lncRNAs [118] . So far , Knockout models for specific lncRNAs did not produce any phenotypes . However , evidence for their importance stems from lncRNAs involvement in cancer and other human diseases , and evolutionary analyses suggest that lncRNAs represent a new class of non-coding genes whose importance should become clearer upon further experimental investigation [119] . We anticipate some of these associations will be made clearer by longitudinal studies that will include larger cohorts of PD patients as well as targeted lncRNA knockout models that will experimentally validate a link between splicing events with lncRNA differential expression . The discovery of at least one lncRNA regulated in our PD patients that affects splicing , highlights additional potential candidate lncRNA spliced targets consistently identified via RNA-Seq , junction and exon microarray analyses . Importantly , we found highly complex and previously unknown splicing and alternative poly-A patterns in healthy controls' leukocytes and a conspicuous decline of this rich variability in PD leukocytes . Together , these findings support the notion of a massive impact of both lncRNAs and the existence AS changes that cause a wide range of transcript-level structure modifications in PD . Blood cells provide an accessible source for biomarker identification , and although accurate identification of disease biomarkers in the blood has proven difficult in the past , blood biomarkers were recently found for both neurological diseases as well as psychiatric disorders [34] , [35] , [36] . Future studies in larger cohorts of Parkinson's patients will enable verification of disease markers in the blood . Here , we employed a non-biased full leukocyte RNA-sequencing followed by detection of known and novel splicing events and transcript functional level annotations concurrently with detection of poly-A sites . This allowed us to profile both known and novel structural transcript changes in PD pre- and post-treatment ON- and OFF-stimulus at an unprecedented depth . At the structural level , mutations in the prion-like domains of splicing factors such as heteronuclear ribonucleoprotein AB ( hnRNP A/B ) and FUS were recently shown to lead to pathological protein fibrils [50] . While their involvement in sporadic neurodegenerative processes is still incompletely understood , findings of hnRNP A/B decline in Alzheimer's disease [79] suggests that impaired splicing regulation might be involved in the emergence of sporadic neurodegenerative processes , including PD . Splicing alternations were also reported to occur early on in Alzheimer's disease ( AD ) , and failed nuclear transport and fibril formation by splicing factors harboring prion-like domains , such as hnRNP A/B and FUS was recently implicated in Amyotrophic Lateral Sclerosis ( ALS ) [50] . It is hence noteworthy that we found AS changes in the prion-like domains of the non-mutated variants of these transcripts and identified protein transport to nuclei as a primarily impaired signaling network in PD leukocytes . Additional predictions involve neuro-immune signaling , with a specific focus on the CDC42 Rho GTPase which functions both in controlling anxiety and in defense against viral infection and general immune cell activities , both phenomena known in PD patients and which emerged in our network analysis as changed in the disease . Post-mortem brain studies of sporadic PD , highlighted mitochondrial dysfunction as being central to the disease [120] , and it was further pointed out as contributing to the pathogenesis of other neurodegenerative diseases such as Ataxia [121] . Pink1 and Park2 may act in a quality control pathway preventing the accumulation of dysfunctional mitochondria , and regulators that control Park2 translocation into the damaged mitochondria were recently elucidated [120] , revealing that this pathway is much more complicated than previously appreciated , and suggesting that other , yet unknown , regulators also contribute to the process . Here , we have charted the first whole transcriptome genome-wide splicing map of Parkinson's leukocytes through characterization of both known and novel junctions and exons via multileveled analysis of high throughput long RNA sequencing . Wide annotations of alternate promoters , splicing and alternative poly-A sites allowed us to identify and quantify both disease- and treatment-induced splicing shifts , miRNA binding site modifications , putatively changed protein-protein interactions and other transcript structural changes in the three tested states of the participant patients ( disease , post-treatment ON- and OFF- stimulus ) . We also noted shifts in splice patterns in PD leukocytes as compared with the global splicing map of the human genome , which was partially sustained post-DBS presenting specific attenuation of disease-derived increase in the frequency of Poly-A choices . To identify inclusion exons expressed in a reciprocal nature relative to a corresponding exclusion junction we have implemented and applied on the RNA-Seq data a more stringent algorithm based on the ASPIRE analysis approach . Importantly , we provide here a full resource of leukocyte-expressed lncRNAs in both disease and healthy states and specifically in PD . In summary , we developed a novel computational approach and a user friendly tool for analyzing whole-transcriptome RNA-Seq data through sample specific database construction . Our workflow includes identification of novel splice junctions , exons and splicing events , including such that involve novel variants , in protein-coding genes . We combined both exon- and junction- level analyses by applying this newly developed version of AltAnalyze for RNASeq analysis to gain deep insight into gene expression and splicing aberrations in PD and search for electrical stimulation -induced changes , concurrently with global detection and differential expression of the leukocyte expressed lncRNAs . RNA-seq comprehensive analyses thus enable new insight to leukocyte transcriptome data , which becomes an important resource for researchers of neurodegenerative diseases overall , and our results will provide insights into DBS-treatable diseases overall ( including mental disorders [122] ) . In particular , lncRNAs may be future novel biomarkers for PD and other neurodegenerative and neurological conditions and an important tool in future personalized neurology .
PD patients and matched controls were recruited to the study according to the declaration of Helsinki ( Hadassah University Hospital , Ein-Kerem , approval number 6-07 . 09 . 07 ) and have signed informed consent prior to inclusion in the study . Blood leukocytes were collected from 3 PD patients pre- and post- bilateral sub-thalamic ( STN ) -DBS neurosurgery while being on stimulation and following a short 1-hour of stimulation cessation and from 3 healthy age-matched control healthy volunteers ( HC ) . The age , disease duration and Body-Mass Index ( BMI ) of the study patient participants are given under Table ST1 . All the patient study volunteers that passed our stringent set of exclusion criteria signed informed consent forms prior to inclusion in the study ( clinical parameters of the recruited volunteers are given under ST1 ) . To control for variability in the leukocytes expression profiles that stem from other factors ( such as infections , or other diseases ) , volunteers were assessed for their clinical background and state and fulfilled detailed medical history questionnaires . Exclusion criteria for participant patients included depression and past and current DSM Axis I and II psychological disorders ( SM ) , chronic inflammatory disease , coagulation irregularities , previous malignancies or cardiac events , or any surgical procedure up to one year pre-DBS . Potential volunteers that did not fulfill these inclusion criteria were not recruited to the study . All patients went through bilateral STN-DBS electrode implantation ( Medtronics , USA ) and were under dopamine replacement therapy ( DRT ) both pre- and post-DBS ( on significantly reduced dosage post-DBS with t-test p<0 . 01 ) , the last medication administered at least five hours pre-sampling . The clinical severity of the disease was assessed by a neurologist by the Unified PD Rating Scale ( UPDRS ) [123] . Controls were recruited among Hadassah hospital staff and researchers at the Edmond J . Safra Campus ( Jerusalem ) . All study volunteers underwent stringent filtering prior to inclusion in the study . The exclusion criteria for the healthy control volunteers included smoking , chronic inflammatory diseases , drug/alcohol usage , major depression , previous cardiac events , fever within up to three months prior to inclusion in the study and past year hospitalizations . Blood collection was conducted in a fixed range of hours ( 10AM–14PM ) . In order to reduce expression profile variability that depends on the time of sampling . To ensure accurate inspection of in-vivo leukocyte expressed RNA , the collected venous blood ( 9 ml blood using 4 . 5 ml EDTA ( anti-coagulant ) tubes ) was immediately filtered using the LeukoLock fractionation and stabilization kit ( Ambion , Applied Biosystems , Inc . , Foster City , CA ) . To ensure high RNA quality , the leukocyte-enriched samples were immediately incubated in RNALater ( Ambion ) ( http://www . affymetrix . com/support/technical/technotes/blood_technote . pdf ) . Stabilized filters and serum samples were stored at −80°C . RNA extraction followed the manufacturers' alternative protocol instructions for RNA extraction from LeukoLock filters . Briefly , cells were flushed ( TRI-Reagent Ambion ) into 1-bromo-3-chloropropane-containing 15 ml tubes and centrifuged . 0 . 5 and 1 . 25 volume water and ethanol were added to the aqueous phase . Samples were filtered through spin cartridges , stored in pre-heated 150 µl EDTA; RNA was quantified in Bioanalyzer 2100 . Determination of RNA quality and quantity were conducted using the Eukaryote Total RNANano 6000 kit ( Agilent ) . RNA was frozen and stored in −80°C immediately after production . RNA quality was assessed by running the samples on Agilent RNA 6000 Nano-gel ( #5067-1511 ) . For each library Ribosomal RNA of 5 ug total RNA was removed using Invitrogen RiboMinus kit ( #A10837-08 ) and then sample was concentrated using the RiboMinus Concentration Module ( Invitrogen ) . Ribosomal RNA removal was verified by RNA 6000 Nano gel analysis . Library construction was conducted according to SOLiD Whole Transcriptome Analysis Kit ( PN4425680 ) protocol , fragmentation ( by RNase-III ) was verified on Agilent RNA 6000 Pico Kit ( #5067-1513 ) and 150 ng fragmented RNA ware used for further protocols . cDNA samples were run on 4% Agarose gel , 150–250 base pairs ( bp ) sized fragments were cut and extracted using Qiagen Min-Elute Gel-Extraction Kit ( #28604 ) , gel was dissolved by intensive vortex and not by heating . Libraries were amplified for 12 cycles using bar-coded primers supplied in SOLiD Transcriptome Multiplexing Kit ( Ambion , #4427046 ) . Libraries were quantified using the Kapa ABI SOLiD Library Quantification Kit ( KK4833 ) and diluted for final analysis on Agilent High Sensitivity DNA Kit ( #5067-4626 ) . 500 uM libraries were used for emulsion PCR according to Applied Biosystems SOLiD-3 System Template Bead Preparation Guide ( 4407421 ) to prepare for sequencing on the SOLiD-3 platform . RNA-Seq reads ( . csfasta files ) and quality scores ( . qual files ) were obtained using the SOLiD instrument software: SOLiD-3 . SOLiD-3 . System software analysis was used for all the primary data analysis including image analysis , bead finding , quality metrics and color calls . The software applications used to set and control data analysis included SOLiD software suite under license agreement . The suit included: Instrument Control Software ( ICS ) , SOLiD Experimental Tracking System ( SETS ) , and SOLiD Analysis Tools ( SAT ) V3 . 0 . Job management by the Job Manager used the Corona-Lite v4 . 0 platform . Sequencing was run on the Applied Biosystems SOLiD 3 System . Images of each cycle were analyzed , data was clustered and normalized . For each tag , a sequential ( sequence-ordered ) set of color space calls was produced . Quality metrics were produced through normalization . Two probe sets were used to maximize the fraction of “mappable” amplified beads , read length and sequencing throughput for sequencing of the 50-bp reads . Five rounds of primers ( A , B , C , D and E ) were used to sequence template by ligation of di-base labeled probes . As the libraries were size fragmented , the set of primers used was specific to the P1 Adaptor . For each library three types of raw data files were created: . csfasta ( the sequenced reads in color space ) , . qual and . stats . The quality values given in the . qual files ( estimate of confidence given for each color call ) , q for a particular call , is mathematically related to its probability of error ( p ) , and is calculated as follows: q = −10log10p . The SOLiD q values are similar for those generated by Phred and the KB basecaller for capillary electrophoresis ( described in detail under [124] ) . The algorithm relies on training ( calibration ) to a large set of control data and color calls for which the correct call is known . In SOLiD-3 system , the correct call is determined by mapping the read to a known reference sequence . The secondary data analysis included matching of the reads to reference genome and generation of base space sequences . Each library was mapped using SOLiD BioScope ( v1 . 3 ) software ( life technologies , applied biosystems , Carlsbad , California ) via cloud computing . The count reads were mapped to the UCSC human genome version 19 ( February 2009 GRCh37/hg19 assembly , homo_sapiens . GRCh37 . 56 . dna . toplevel . fa database ) twice: once to receive exon quantification ( using the counttag tool ) and once more to receive junction level quantification , using the BioScope splice junction extractor tool . The * . gff and * . sam files were created during this analysis step . The BioScope alignment software Mapreads was used . Count merging that employs discontinuous word pattern search algorithms was performed in both pipelines . The bed coordinates of the Gencode v . 7 human long non-coding RNAs database were downloaded from the GENCODE lncRNA data page of the CRG Bioinformatics and Genomics Group [http://big . crg . cat/bioinformatics_and_genomics/lncrna_data] and complemented with other non-coding transcript information available from the EnsEMBL BioMart version 0 . 7 query interface to the EnsEMBL Genes 72 – GRCh37 . p11 database ( www . ensembl . org ) Genome coordinates in bed format corresponding to the mapped reads for all samples used the Lifescope Lifetech 2 . 5 . 1 software and UCSC hg19 masked reference database as obtained by the original . sam files with SAMtools , SAMtools view and bedtools bamToBed . These read bed files were intersected with the genome coordinates of the above-mentioned lncRNAs using the bedtools intersectBed program , requiring a 90% overlap of each sequence read with a target lncRNAs . Lists of sequence tags corresponding to lncRNAs were obtained by intersection of the bed tools . The count information of all of the detected leukocyte lncRNAs was first filtered . LncRNAs that did not present read count in three or more libraries , and ones that did not exist in EnsEMBL were filtered out . The remaining lncRNAs ( overall , 6430 ) were analyzed using the Bioconductor edge-R [59] software version 3 . 0 . 1 to detect differential expression in PD patients pre-DBS compared to healthy control volunteers , post-DBS on stimulation as compared with pre-DBS state and post-DBS off electrical stimulation as compared with post-DBS on electrical stimulation . This analysis module is particularly suitable to use on small number of rate replicate samples . The results were annotated using the BioMart integrated annotation database query interface [130] , using the human genome reference consortium assembly build version 37 ( GRch37 , hg19 ) and GENCODE version 7 [131] . A method that maximizes the pseudo-expected accuracy of the model served for prediction of RNA secondary structure [68] . The binding affinity of the lncRNAs to potential targets as sponge was computed using MirTarget2 [132] , which implements a support-vector-machine ( SVM ) -based miRNA target prediction algorithm that scans all of the seed-matching sites in the potential targets and predicts both conserved and non-conserved miRNA targets in mammals . Dissected brain tissues ( amygdala and SN ) from PD patients ( n = 5 , 4 males and 1 female ) and unaffected ( non-demented ) controls ( n = 5 , 3 females and 2 males ) were provided by the Netherlands Brain Bank ( NBB ) ( ST9 ) . Ethical approval and written informed consent from the donors or the next of kin was obtained in all cases . These tissues were kept at −70 degrees until use and served for RNA extraction and qRT-PCR validation tests . RNA was extracted from brain tissues using the QIAGEN ( Venlo , Netherlands ) easy kit , which ensures full representation of all RNA length groups . Briefly , brain tissue was homogenized with 700 µL QIAzol lysis buffer and subsequently lysed for 5 minutes and mixed with 140 µL of chloroform to allow full neutralization . Centrifugation for 15 minutes ( in 12 , 000× g and 4°C ) followed suspension of 3 minutes . The aqueous phase was mixed with 1 . 5 volume of ethanol , loaded on RNA-binding spin column and centrifuged for 30 seconds in 8400× g . The column was washed with 700 µL RWT buffer ( 85% ethanol ) and twice with 500 µL RPE buffer ( QIAGEN , 70% ) , for the removal of DNA and protein remnants , respectively . Washing was performed by centrifugation for 30 seconds in 8400× g and discarding the flow-through . Finally , spin column was centrifuged for 1 minute in 21 , 067× g to further dry the column . 50 µL of nuclease-free water were used to elute the RNA , which was immediately put on ice to prevent degradation . The RNA concentration was determined by Nanodrop-1000 , and its integrity was assessed with 1% agarose gel and identification of two distinct bands ( 28S and 18S rRNA ) . The DNA remnants were degraded using Sigma Aldrich ( 3 , 050 Spruce St . , St . Louis , Missouri 63 , 103 , United States ) DNAse1 Amplification Grade ( Sigma Aldrich ) . 800 ng of RNA were diluted in 8 µL of nuclease-free water , and then mixed with 1 µL of DNAse and 1 µL of 10× reaction buffer ( Sigma Aldrich ( 3 , 050 Spruce St . , St . Louis , Missouri 63 , 103 , United States ) ) . Following 15 minutes of incubation in 25°c , 1 µL of stop solution ( Sigma Aldrich ( 3 , 050 Spruce St . , St . Louis , Missouri 63 , 103 , United States ) ) was added , and the mixture was incubated for 10 minutes in 70°c on a MJ Research PTC 200 Thermal Cycler ( GMI Inc . , 6511 Bunker Lake Blvd , Ramsey , MN 55303 , United States ) . The entire volume of the mixture ( 800 ng of RNA in 11 µL volume ) was used for cDNA preparation using the Quanta qScript cDNA Synthesis Kit ( Quanta biosciences Inc . , 202 Perry Parkway , Suite 1 . Gaithersburg , MD 20877 , USA ) . For each reaction 11 µL RNA were mixed with 4 µL ×5 Reaction Buffer , 4 µL Nuclease-Free Water and 1 µL Reverse Transcriptase ( except for the no-RT controls , where reverse transcriptase was not added ) , for a final 20 µL reaction volume . Mixture was put in a 200 µL PCR tube , and placed in a MJ Research PTC 200 Thermal Cycler ( GMI Inc . , 6511 Bunker Lake Blvd , Ramsey , MN 55303 , United States ) programmed for 5 minutes in 22°c , 30 minutes in 42°c and 5 minutes in 85°c . cDNA was then diluted 1∶10 , by adding 180 µL DDW . iTaq Universal SYBR green supermix 2× ( Biorad Inc . , Hercules , California , US ) was used for both the target and reference genes used for normalization . 10 µL of SYBR supermix ( Biorad Inc . , Hercules , California , US ) , 1 µL of 10 µM of each left and right primers , and 8 µL of cDNA were used for each reaction . Reaction was performed on a Biorad ( Hercules , California , US ) CFX96 Touch Real-Time PCR Detection System . The protocol used for product amplification was 95°c for 3 minutes , 95°c for 15 seconds and 51 repeats of 60°c for 30 seconds , then melting curve was performed by increasing the temperature from 67 . 0 to 94 . 6°c in 0 . 3°c increments every 5 seconds . The data was obtained using Bio-Rad CFX Manager 3 . 0 software . For each primer pair and tested samples , triplicate PCR reactions were tested . Triplicates that were not tightly grouped ( more than 1 . 5% cycles apart ) were removed from further calculations as outliers . If the triplicates were not tightly grouped , and no outlier could be identified , the triplicate was re-run ( and the whole sample was omitted from further analysis if the outcome of the re-run still showed un-tight grouping ) . The primers ( Sigma Aldrich ( 3 , 050 Spruce St . , St . Louis , Missuri 63 , 103 , United States ) ) used for the lncRNA targets and reference genes are as follows: RP11-462G22 . 1 – forward primer GAGCTGCCTTTCATCTGGTC , reverse primer GGTAGTGCTTTGCCTCATCC; U1 – forward primer GAACCCCGAGTCCACTGTAA , reverse primer TGAACCCCGTTATGTCAGGT; RP11-79P5 . 8 – forward primer CTCGGCTTCGACTTTAGCTG , reverse primer CTTCTTTTTCACCGCTCCTG; TUBB3 – forward primer GCAACTACGTGGGCGACT , reverse primer GGCCTGAAGAGATGTCCAAA; RPL-19 – forward primer GCTCGATGCCGGAAAAACAC , reverse primer GCTGTACCCTTCCGCTTACC . The sequences ( in fastq or Color Space format ) were mapped against the GENCODE V7 . 0 reference database using the Bowtie aligner software [133] version 1 . 0 . 0 against the FastA files from GENCODE lncRNA catalogue at CRG ( http://big . crg . cat/computational_biology_of_rna_processing/lncrna ) . This target dataset consisted of 14 , 880 sequences in FastA format and was used also for the leukocyte RNA-Seq data alignment . The run parameters were set as in the alignment of the leukocyte RNA-Seq count reads ( -k1 and –best ) . The aligned sequences were parsed with in-house generated scripts ( Genomnia srl , Milan , Italy ) and transformed in transcript count tables with EnsEMBL GENCODE transcript ID identifiers . Several transcripts that were not included in EnsEMBL were eliminated from the dataset prior to the analysis ( since considered as ‘retired’ transcripts ) . Cross datasets comparisons were performed with the VLOOKUP Excel function based on the ID column . Differential expression analysis of PD samples compared to normal controls was performed with the Bioconductor EdgeR software [59] version 3 . 4 . 2 on R ( version 3 . 0 . 2 , 64 bit ) , using Trimmed Mean of Ms ( TMM ) normalization [60] and exact test . The tagwise dispersion was calculated for each lncRNA ( Supplementary Figure S3 , PD leukocyte compared to controls as an example ) , and was moderated by EdgeR toward a common value inferred by all the examined genes . The dispersion parameter determined how to model the variance for each gene in each comparison and dataset . The common variation is inferred from all the datasets . The variance under a negative binomial model computed as where EM is the estimated mean and D – the dispersion , for each gene in each comparison and dataset . The fold change calculation uses the square root of the dispersion as the biological coefficient of variation , inferred by Poisson distribution ) , and the tagwise dispersion parameter determines how to model the variance for each gene for the differential expression analysis . All the raw and processed RNA-Seq whole transcriptome profiling files ( . csfasta , . qual , . stats , . gff , . bam , . tab , . contig . range and . bed files ) were deposited under the Gene Expression Omnibus depository ( GEO ) [52] and are available under series accession number GSE42608 . The independent PD brain RNA-Seq dataset was obtained from the Array Express repository [134] ( accession number E-GEOD-40710 ) . | Long non-coding RNAs ( lncRNAs ) comprise a novel , fascinating class of RNAs with largely unknown biological functions . Parkinson's-disease ( PD ) is the most frequent motor disorder , and Deep-brain-stimulation ( DBS ) treatment alleviates the symptoms , but early disease biomarkers are still unknown and new future genetic interference targets are urgently needed . Using RNA-sequencing technology and a novel computational workflow for in-depth exploration of whole-transcriptome RNA-seq datasets , we detected and analyzed lncRNAs in sequenced libraries from PD patients' leukocytes pre and post-treatment and the brain , adding this full profile resource of over 7 , 000 lncRNAs to the few human tissues-derived lncRNA datasets that are currently available . Our study includes sample-specific database construction , detecting disease-derived changes in known and novel lncRNAs , exons and junctions and predicting corresponding changes in Polyadenylation choices , protein domains and miRNA binding sites . We report widespread transcript structure variations at the splice junction and exons levels , including novel exons and junctions and alteration of lncRNAs followed by experimental validation in PD leukocytes and two PD brain regions compared with controls . Our results suggest lncRNAs involvement in neurodegenerative diseases , and specifically PD . This comprehensive workflow will be of use to the increasing number of laboratories producing RNA-Seq data in a wide range of biomedical studies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome",
"sequencing",
"genome",
"expression",
"analysis",
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"genome",
"analysis",
"tools",
"functional",
"genomics",
"gene",
"expression",
"biology",
"genomics",
"molecular",
"cell",
"biology",
"neuroscience",
"transcriptomes"
] | 2014 | Long Non-Coding RNA and Alternative Splicing Modulations in Parkinson's Leukocytes Identified by RNA Sequencing |
In vertebrate embryos , the earliest definitive marker for the neural plate , which will give rise to the entire central nervous system , is the transcription factor Sox2 . Although some of the extracellular signals that regulate neural plate fate have been identified , we know very little about the mechanisms controlling Sox2 expression and thus neural plate identity . Here , we use electroporation for gain- and loss-of-function in the chick embryo , in combination with bimolecular fluorescence complementation , two-hybrid screens , chromatin immunoprecipitation , and reporter assays to study protein interactions that regulate expression of N2 , the earliest enhancer of Sox2 to be activated and which directs expression to the largest part of the neural plate . We show that interactions between three coiled-coil domain proteins ( ERNI , Geminin , and BERT ) , the heterochromatin proteins HP1α and HP1γ acting as repressors , and the chromatin-remodeling enzyme Brm acting as activator control the N2 enhancer . We propose that this mechanism regulates the timing of Sox2 expression as part of the process of establishing neural plate identity .
Sox2 is a transcription factor that plays multiple critical roles during embryonic development in vertebrates . In embryonic stem ( ES ) cells , as well as in adult central nervous system ( CNS ) stem cells , Sox2 expression is required for the maintenance of multipotency and for the ability of cells to self-renew [1] . Sox2 is also expressed in cells that retain their ability to proliferate and/or acquire glial fates , whereas it is down-regulated in cells that become postmitotic and differentiate into neurons [2–4] . In addition , it is also transiently expressed outside the CNS in cranial sensory organs derived from the placodes and in subsets of peripheral nervous system ( PNS ) cells [5 , 6] . In all vertebrates studied to date , Sox2 is also a general marker for the very early developing neural plate . In the chick , for example , Sox2 expression starts at the late primitive streak stage ( stages 4–4+ [7] ) in the future neural territory [8 , 9] . A morphologically recognizable neural plate only becomes visible after the beginning of Sox2 expression [8] . Importantly , Sox2 function is required for development of the neural plate [10] . Time-course experiments have shown that induction of Sox2 requires the same period of exposure to organizer-derived signals ( the tissue responsible for inducing the neural plate in the normal embryo [11–13] ) as is required to induce a mature neural plate [14–17] . For these reasons , Sox2 is considered to be the earliest definitive marker for the neural plate [18 , 19] . The complex expression profile of Sox2 is controlled by multiple regulatory elements , each responsible for directing expression to a specific subset of expression sites . A very compelling analysis of the noncoding regions of Sox2 in the chick embryo [20] revealed as many as 25 distinct conserved enhancers , of which two account for the expression of this gene in the early neural plate at stages 4+–5 . One of these enhancers , named N2 , is responsible for the initial expression ( stage 4–4+ ) and is activated in a large domain corresponding to the entire forebrain/midbrain and most of the hindbrain . The other , N1 , drives expression in the future caudal hindbrain and spinal cord and is activated a little later ( around stage 5 ) [20 , 21] . To understand the processes that define the neural plate , it is essential to understand how the activity of these two elements , and especially N2 , is regulated in the embryo . Analysis of the N2 enhancer reveals multiple putative binding sites for known transcription factors [20 , 21] . However , the spatial and temporal expression patterns of these factors do not provide an obvious explanation for the time of onset of Sox2 expression in normal development ( unpublished data ) . Furthermore , to date , no single secreted factor or any combination thereof has been found to induce either Sox2 expression or a neural plate in competent cells not normally fated to form part of the neural plate [13 , 19] . We therefore directed our attention to nuclear factors that might regulate this enhancer . Here , we provide evidence that a group of coiled-coil proteins interact with each other and with chromatin-remodeling factors and heterochromatin proteins to regulate the activity of the N2 enhancer . We propose that this is part of a mechanism that regulates the time of onset of expression of Sox2 in the nascent neural plate .
A recent study [22] using the P19 cell line demonstrated that the chromatin-remodeling enzyme Brahma ( Brm ) can activate Sox2 by binding directly to the N2 enhancer . Is Brahma also involved in regulating Sox2 expression in the normal embryo ? To test this , we introduced a mutated version of Brahma ( BrmK755R , which does not bind ATP and is therefore unable to remodel chromatin [23] ) by electroporation into the prospective neural plate of embryos at stage 3–3+ . This resulted in strong inhibition of Sox2 expression in the electroporated domain ( Figure 1A and 1B; 5/6 ) , unlike controls electroporated with green fluorescent protein ( GFP ) ( Figure 1C and 1D; 0/5 expressing ) . However , Brm is expressed ubiquitously in the embryo [24]; what mechanisms prevent premature expression of Sox2 ? A good candidate is the transcriptional repressor HP1α , which binds directly to Brahma-related proteins at a highly conserved site [25] and which is also ubiquitously expressed in early embryos ( Figure 2 ) . Consistent with this , overexpression of HP1α in the neural plate represses Sox2 ( Figure 1E and 1F; 3/3 ) . Could HP1α be an endogenous inhibitor of Sox2 expression ? To address this , we took advantage of the fact that both the chromoshadow domain and the chromodomain are necessary for the function of HP1 proteins [26 , 27]: targeting to chromatin requires interaction of the chromoshadow domain with a chromatin-tethered partner , as well as binding of the chromodomain to a methylated Lys9 of histone H3 [28] . We therefore made a dominant-negative form of HP1α ( ΔHP1α ) consisting of its isolated chromoshadow domain ( which can bind to Brahma-related proteins but lacks repressor activity [25] ) . When ΔHP1α is misexpressed as a line extending from the neural plate into the peripheral , nonneural ectoderm ( see Materials and Methods , “Design of assays” ) , Sox2 is induced ( Figure 1G and 1H; 6/7 ) , whereas similar electroporation of GFP has no effect ( Figure 1I and 1J; 0/8 ) . This suggests that HP1α activity is required to prevent expression of Sox2 in the nonneural ectoderm . In embryos in which ΔHP1α was expressed as a line , we observed that Sox2 was up-regulated , not only in the embryonic nonneural ectoderm ( prospective epidermis ) , but also in the more peripheral area opaca epiblast ( extraembryonic ectoderm ) ( Figure 1G ) . We were surprised by this observation because until now , various factors ( such as bone morphogenetic protein [BMP] antagonists [16 , 17 , 19] ) have been described that can expand the neural plate domain , but never as far as the extraembryonic epiblast , and none can induce a separate domain of Sox2 expression in this region . The only treatment described to date that can induce neural markers in the area opaca is a graft of the organizer , Hensen's node , which is able to generate a complete , patterned nervous system in this region [13 , 29–32] . These observations define the area opaca as a particularly rigorous location in which to test for the neural inducing ability of factors ( see Materials and Methods , “Design of assays” ) . Electroporation of ΔHP1α in this region dramatically induces Sox2 , showing that HP1α normally represses Sox2 expression ( Figure 1K and 1L; 8/8 ) . In contrast , electroporation of GFP in the same region has no effect ( Figure 1M and 1N; 0/10 ) . As an additional control , since HP1α may have more general activity as a transcriptional repressor , we also tested whether ΔHP1α can also induce other early embryonic genes using Brachyury , a marker for mesoderm expressed at this stage of development . It did not ( Figure 1O and 1P; 0/3 ) . This result also confirms that the induction of Sox2 is direct , rather than a consequence of prior induction of mesoderm by ΔHP1α . Likewise , electroporation of BrmK755R or Brm had no effect on neural or mesoderm markers ( 0/5 for each; unpublished data ) . To test whether the inducing activity of ΔHP1α requires Brahma , we introduced ΔHP1α together with BrmK755R . This combination fails to induce Sox2 ( Figure 1Q and 1R; 0/12 ) , suggesting that HP1α normally inhibits Sox2 expression through a Brm-dependent mechanism ( Figure 1S ) . In Xenopus , the gene encoding the coiled-coil protein Geminin is expressed in the early prospective neural plate , and its misexpression induces neural markers [33] . More recently , it has been shown that Geminin interacts genetically with Drosophila Brahma , that it binds directly to its vertebrate homologs Brg1 and Brm ( at the same site as does HP1α ) , and that Geminin knock-down abolishes Sox2 expression [25 , 34] . Could Geminin be responsible for releasing the repression of Brm activity by HP1α ? To test this , we cloned the chick homolog of Geminin . Before and during early gastrulation , Geminin is expressed in a large domain , which then ( from stages 4–4+ ) becomes restricted to the neural plate ( Figure 3 ) . The early expression of chick Geminin resembles that of “pre-neural” genes ( such as Sox3 , ERNI , and Churchill ) , which precede the initiation of Sox2 expression and which are induced by fibroblast growth factor 8 ( FGF8 ) [8 , 9 , 35 , 36] . We therefore tested whether FGF can also induce Geminin . Indeed , Geminin can be induced by FGF8-soaked beads ( 9/10; Figure 3H , arrow ) , but not by control beads ( 0/10; Figure 3H , arrowhead ) . When misexpressed as a line extending laterally from the neural plate , Geminin strongly induces ectopic Sox2 ( Figure 4A and 4B; 8/8 ) . Geminin can also strongly induce Sox2 expression when introduced into the extraembryonic epiblast ( Figure 4C and 4D; 20/20 ) . To test whether this induction requires the chromatin-remodeling activity of Brm , we cointroduced Geminin and BrmK755R: the mutated chromatin remodeler abolishes the induction of Sox2 by Geminin ( Figure 4E and 4F; 0/15 ) , suggesting that Brm activity is required for Sox2 induction by Geminin . Together , these results suggest that early in development Sox2 expression is constitutively repressed by the presence of HP1α bound to Brm at the N2 enhancer of Sox2 , and that later in development , FGF may release this inhibition through induction of Geminin , which competes HP1α away from the protein complex ( Figure 4G ) . Geminin is already expressed at the beginning of gastrulation ( Figure 3 ) , long before Sox2 ( which appears at stage 4 [8 , 9] ) , suggesting that an additional mechanism must exist to prevent premature Sox2 activation . A good candidate for this repression is ERNI , which is broadly expressed in the epiblast at early stages but is rapidly down-regulated from the prospective neural plate at stage 4+ [36] , around the time when Sox2 starts to be expressed . To test whether ERNI can inhibit the induction of Sox2 by Geminin , we cointroduced them into the area opaca: ERNI does indeed inhibit the induction of Sox2 by Geminin ( 6/14 with very weak induction , 8/14 with no induction; Figure 5A–5C ) . The above findings are consistent with the idea that ERNI normally functions to repress Sox2 expression at very early stages of development . However , it is unlikely that down-regulation of ERNI transcription is sufficient to relieve this inhibition because Sox2 expression begins at stages 4–4+ [7] , when some ERNI transcripts can still be detected within the prospective neural plate [36] . Therefore , an endogenous inhibitor is likely to exist whose expression should begin at around this time ( stage 4–4+ ) . To identify such an inhibitor , a two-hybrid screen was performed using ERNI as bait and a library of cDNAs from stage 3–6 chick embryos ( Figure 9A ) . Only one partner was found , encoding a small coiled-coil domain protein which we named BERT ( Figure 9B ) . An equivalent human protein ( SCOCO , corresponding to the fragment shown in bold in Figure 9B ) was previously isolated as a partner of human ARL1 , a component of the Golgi apparatus [44] , and a nematode homolog ( unc69 ) was found to be essential for neural development [45] . BERT is expressed ubiquitously at low levels at all stages , but is up-regulated specifically in the prospective neural plate from stage 4–4+ ( Figure 10 ) , just prior to when Sox2 expression appears [8 , 9] . When misexpressed as a line across the nonneural epiblast , BERT acts like the dominant-negative ERNI constructs: it induces strong expression of Sox2 ( Figure 9C and 9D; 15/15 ) , which also acquires a neural plate-like morphology ( Figure 9D′: note that the thickened ectoderm characteristic of the neural plate has greatly expanded on the electroporated side; see arrowhead on right ) . Mesodermal markers ( Brachyury , Chordin , and BMP4; 0/5 , 0/4 , 0/4 , respectively; unpublished data ) are not induced , showing that this expansion of the neural plate is a direct effect . These findings indicate that BERT has an activity compatible with it being an endogenous antagonist of ERNI , regulating not only Sox2 expression , but also the onset of neural plate development . To confirm this , we examined the effects of coexpressing BERT with Geminin+ERNI ( which does not induce Sox2; see above and Figure 5A and 5B ) in the area opaca . Indeed , when all three constructs are cointroduced , induction of Sox2 is seen ( 12/12; Figure 9E and 9F ) . Is BERT required to control the onset of Sox2 expression in the neural plate ? To address this , we designed a fluorescein-labeled Morpholino oligonucleotide ( MO ) to the 5′ end of the coding sequence ( see Materials and Methods ) and introduced this ( together with GFP ) by electroporation into the prospective neural plate at stage 3–3+ and examined Sox2 expression at stages 4+–5 . BERT-MO caused down-regulation of Sox2 expression in this domain ( Figure 9G and 9H; 5/6 ) , unlike control MO ( Figure 9I and 9J; 0/5 ) . Staining of BERT-MO–electroporated embryos with an antibody against BERT/SCOCO ( see Materials and Methods ) confirmed that the MO does indeed inhibit translation of BERT protein ( Figure 9K ) in the electroporated domain ( Figure 9L ) . Together , these findings implicate BERT as an endogenous antagonist of ERNI , required to regulate the onset of Sox2 expression in the neural plate . From the results presented above , the evidence that BERT binds to ERNI directly is based entirely on the two-hybrid screen used to isolate BERT . To confirm that the two proteins can interact physically , we used BiFCo assays , which further revealed that BERT , Geminin , and ERNI all bind to each other through their coiled-coil domains ( Figure 11 and Table 1 ) . This finding raises the possibility that BERT disrupts Geminin-ERNI dimers by binding to both proteins , thus removing ERNI-HP1γ from the complex to activate Sox2 ( Figure 9M ) . To test this further , we used BiFCo competition assays ( Figure 12 ) . When BERT is added to Geminin-Venus ( N ) +ERNI-Venus ( C ) , fluorescence is lost ( Figure 12A and 12B ) . When Dlx5 is used as a control instead of BERT in this assay , there is no effect ( Figure 12C ) . Conversely , when BERT-Venus ( C ) is added to Geminin-Venus ( N ) +ERNI , fluorescence is generated ( Figure 12D and 12E ) , and this is not mimicked by addition of Dlx5-Venus ( C ) ( Figure 12F ) . Likewise , when BERT-Venus ( C ) is added to Geminin+ERNI-Venus ( N ) , fluorescence is produced ( Figure 12G and 12H ) , which is not mimicked by the use of Dlx5-Venus ( C ) instead of BERT-Venus ( C ) ( Figure 12I ) . Together , these findings support the idea that BERT can disrupt Geminin-ERNI heterodimers by binding to both proteins . The experiments described above tested the protein–protein interactions and their effects on Sox2 expression , but their physical association with the N2 enhancer was extrapolated from published results in a cultured cell line , unrelated to the early neural plate [22] . To test whether these interactions can regulate Sox2 expression directly at the N2 enhancer , we coelectroporated a reporter construct consisting of the N2 enhancer and a minimal TK promoter [20] driving expression of LacZ together with either Geminin alone , with Geminin+ERNI , or with Geminin+ERNI+BERT , into the extraembryonic epiblast . No expression of the reporter was seen when it was coelectroporated with the control construct pCAβ-GFP ( Figure 13A and 13B; 0/16 ) or with Geminin+ERNI ( unpublished data; 0/6 ) . However , expression was induced by both Geminin ( unpublished data; 6/6 ) and by Geminin+ERNI+BERT ( Figure 13C and 13D; 5/5 ) . This shows that ERNI can block the activity of the N2 enhancer of Sox2 and that BERT inhibits this . Finally , to confirm that these proteins do indeed interact physically with the N2 enhancer , we conducted chromatin immunoprecipitation ( ChIP ) assays using chromatin extracted from embryonic day ( E ) 7 . 5 mouse embryos and an antibody against mouse Geminin . The antibody specifically precipitates the N2 enhancer of Sox2 ( Figure 13E , lane 3 ) , unlike control experiments performed either without chromatin ( Figure 13E , lane 1 ) or without anti-Geminin antibody ( Figure 13E , lane 2 ) . These findings demonstrate that Geminin does indeed associate physically with the N2 enhancer of Sox2 in vivo at an appropriate stage in development .
The most parsimonious model to explain our findings in terms of how Sox2 expression is regulated in the early neural plate comprises the four steps shown in Figures 1S , 4G , 7E , and 9M . Since Brm and HP1α are expressed ubiquitously ( [24] and results from the present study ) , we propose that there is a basal state in which Brm is bound to the N2 enhancer of Sox2 [22] , but the latter is not expressed because the repressor HP1α occupies the chromoshadow-binding domain of Brm ( Figure 1S ) . Early in development , FGF activity induces both ERNI [18 , 36] and Geminin ( this study ) in the epiblast . Geminin binds to the chromoshadow-binding domain of Brm , displacing HP1α ( Figure 4G ) . However , the interaction of ERNI with Geminin recruits the transcriptional repressor HP1γ , thus continuing to prevent premature expression of Sox2 in the epiblast ( Figure 7E ) . Later in development ( stage 4–4+ ) , BERT is up-regulated within the neural plate , where it binds to both ERNI and Geminin and displaces ERNI-HP1γ complexes away from Brm , freeing the latter to activate N2 and thus Sox2 expression ( Figure 9M ) . At around the same time , ERNI transcription starts to be down-regulated in the neural plate . This model accommodates all of our results and those in the literature; its significance is explored further in the following sections . The N2 enhancer of Sox2 is about 550 bp long and is predicted to contain multiple binding sites for transcription factors [20] , many of which are expressed in the epiblast prior to the stage at which Sox2 expression is initiated . In principle , binding of the appropriate activators to the N2 enhancer should turn on Sox2 . However , the spatial and temporal patterns of expression of these factors do not account for the timing or spatial distribution of Sox2 transcription at this stage in development , as many of them are expressed ubiquitously ( unpublished data ) . We therefore propose that , irrespective of the binding of putative activators to the N2 enhancer , the conformation of chromatin , maintained in a closed configuration by HP1 proteins , prevents activation at early stages . It is only when HP1 proteins are removed and the chromatin-remodeling activity of Brm is released that N2 is activated . Chromatin-remodeling complexes may turn out to have a widespread role in the transcriptional activation of specific genes , as exemplified by Smad-activated genes whose transcriptional regulation also requires the activity of such complexes [46] . Likewise during skeletal muscle differentiation , chromatin-binding proteins “mark the spot” for activation of genes by other transcription factors together with chromatin remodeling by SWI/SNF proteins: MyoD binding to chromatin is regulated by the homeodomain protein Pbx1 in cooperation with the Brahma-related enzyme Brg1 [47–50] . To our knowledge , however , this is the first report suggesting that a SWI/SNF chromatin-remodeling complex can recruit HP1 proteins to a specific enhancer to repress transcription of a target gene . Our model proposes mutually inhibitory interactions between several proteins . Why does Sox2 need to be regulated by such a complex mechanism , rather than by merely recruiting a single or a few activators to a simple enhancer ? We suggest that this is one in a series of steps that act to separate different functions for signals that are common to different developmental processes . Previously , we showed that 3–5 h of exposure to signals from the organizer ( Hensen's node ) is sufficient to induce transient expression of the pre-neural marker Sox3 , but not sufficient to induce later neural plate markers ( such as Sox2 ) , and that the BMP antagonist Chordin can stabilize the expression of Sox3 induced by such a graft ( but again not induce Sox2 ) [16] . Based on these findings , we conducted a screen to identify genes induced within 5 h of exposure to the organizer [36] . We identified several genes induced within this time , among them ERNI , which is induced very rapidly , within 1–2 h . FGF8 is sufficient to mimic this effect , and during normal development , ERNI is expressed even before gastrulation , in a domain identical to that covered by the underlying hypoblast ( which expresses FGF8 ) . FGF is required for both mesodermal [51–54] and neural induction [36 , 55 , 56] . How do cells that have received FGF signals decide between these two incompatible fates ? A likely scenario is that cooperation with other factors , present at different times and in different locations , contributes to refine this choice . To allow this to happen , it may be necessary for cells to retain a “memory” that they have received FGF signals yet be prevented from being allocated prematurely to inappropriate fates . ERNI appears to fulfill such a role: while it is expressed , cells are multipotent , as its early domain of expression encompasses the prospective neural and mesendodermal domains as well as some nonneural ectoderm . At the end of gastrulation , ERNI transcription starts to be down-regulated from the future neural plate , remaining only at the border between neural and epidermal domains [36 , 57] . At the same time , BERT is up-regulated in the domain that is losing ERNI expression while Sox2 starts to be expressed in the same domain ( stage 4–4+ ) . This sequence of events could help to explain why it takes such a long time ( about 9 h ) following a graft of a node for Sox2 expression to begin and for a neural plate to be induced [14–17] . Consistent with the proposal that ERNI is part of a mechanism to prevent premature expression of Sox2 , we have observed that transfection of BERT into the prospective neural plate region of stage 2–3 embryos can induce premature expression of Sox2 ( unpublished data ) . The present and previous studies [36] reveal that FGF signaling activates ERNI as well as Sox3 and Geminin expression in the epiblast . However , FGF does not induce BERT , whose expression is also not regulated by BMP antagonists or any combination of known factors implicated in neural induction to date ( unpublished data ) . In future , it will be interesting to determine whether BERT is induced by some other combination of factors or whether its expression is regulated simply by a cell-autonomous timer in cells that are still in the epiblast at the end of gastrulation , but does not require input from other cells . In all likelihood , the mechanisms responsible for regulating Sox2 expression and the acquisition of neural fate will turn out to be considerably more complex , and our model does not rule out additional mechanisms . It will be interesting in future to investigate whether other developmentally expressed genes are regulated by similar processes . Our findings provide a mechanism for how Sox2 expression is initiated as part of the events that define the early neural plate . We propose that ERNI functions as an inhibitor of premature Sox2 expression during early gastrulation: cells expressing ERNI are multipotent and can generate any cell type . Cells that remain in the epiblast at the end of gastrulation and acquire expression of BERT to activate Sox2 , which , most likely together with other genes involved in neural specification , assigns a neural plate fate .
Fertile hens' eggs ( Brown Bovan Gold; Henry Stewart & Co . ) were incubated at 38 °C to the desired stages . Electroporations were performed as described [35] . The coding region of full-length ERNI , ERNI coiled-coil domain ( aa 1–164 ) , chick BERT , chick Geminin , human BrmK755R ( kind gift from Dr A Imbalzano ) , mouse HP1α , mouse HP1α chromoshadow domain ( aa 106–180 ) , and mouse HP1γ chromoshadow domain ( aa 118–176 ) were cloned into pCAβ and electroporated at 0 . 2 μg/μl ( except ERNI and BrmK755R and HP1α , which were used at 0 . 4 μg/μl ) together with 1 μg/μl of pCAβ-GFP , which was used to mark the electroporated cells . The N2-TK-LacZ reporter plasmid was constructed from N2-TK-GFP , kindly provided by Dr H . Kondoh , and was electroporated at 1 μg/μl . FGF8b ( Sigma ) was delivered bound to heparin beads ( prepared as described [19] ) at 50 μg/ml . In situ hybridization and immunostaining for GFP were performed as described [35] . To establish the role of different components in regulating the expression of Sox2 , three different types of assays were used for gain- and loss-of-function experiments . First , to assess the effects on endogenous expression of Sox2 in the normal neural plate , constructs were introduced into the prospective neural plate at mid-primitive streak stage ( stage 3–3+ ) and the embryos incubated about 6–9 h so that the embryo had reached stages 4+–7 , just beyond the stages at which Sox2 expression begins ( 4+ ) and also because at these stages the neural plate is still open , allowing easier visualization of expanded expression . Please note that stages 4+–7 are particularly short , this entire period lasting only about 3 ± 1 . 5 h at 38 °C . To determine whether a construct can induce ectopic expression of Sox2 , two different locations were chosen . In one set of assays , the construct is introduced as a continuous line between the prospective neural plate of the embryo and the inner aspect of the extraembryonic epiblast , covering most of the prospective epidermis . In the other assay , the construct is introduced as a discrete domain within the inner third of the extraembryonic ( area opaca ) epiblast and the embryos incubated 12–15 h ( by which time they have reached stages 6–9 ) . The reasons for choosing both of the latter two assays for induction is that extensive embryological studies have revealed differences in their reactivity to neural inducing stimuli . For example , inhibition of BMP signaling is sufficient to expand the endogenous neural plate laterally ( and BMP misexpression to narrow it ) , but only when the territory is continuous with the embryo's own neural plate [13 , 16 , 17] , suggesting that induction of neural markers by certain stimuli in this region requires cellular continuity with the neural plate and/or its border . On the other hand , a graft of the organizer ( Hensen's node ) is able to induce a complete , patterned ectopic nervous system from the extraembryonic epiblast of the inner area opaca [13 , 29–32] . A period of 9–13-h contact is required to induce Sox2 after a graft of the organizer , which is why 12–15 h was chosen in this assay . To date , no single factor or any combination thereof has been found to mimic this activity of the organizer . It is therefore particularly important , to assess the full inducing properties of a treatment , to test its ability to induce Sox2 in the area opaca . We therefore used all three assays to compile a more comprehensive understanding of the inducing or inhibiting activities of each of the constructs in this study . A translation-blocking MO against BERT with the sequence CAGCGTCCATGTCAGCGTTCATCAT , targeting the 5′ end of the ORF of the gene or a standard control MO ( Gene Tools LLC ) , both labeled with fluorescein , were electroporated by injecting a small volume ( about 0 . 1 μl ) of a stock of the MO at 1 mM exactly as described for electroporation of constructs ( see above ) . Antibody against human SCOCO was kindly provided by Dr . Richard Kahn . This was used in whole mounts by indirect immunoperoxidase with anti-rabbit-HRP using the same method as described for GFP ( see above ) . For two-hybrid screens with embryonic cDNA , poly-A RNA was isolated from 600 chick embryos ( stage 3–6 ) using the Ambion Poly ( A ) Pure Kit . The mRNA was used to synthesize a cDNA library which was cloned into the pMyr vector using the CytoTrap XR Library Construction Kit ( Stratagene ) . The library was transformed into XL10-Gold Ultracompetent Cells ( Stratagene ) . Full-length ERNI was cloned into the pSOS vector and used as bait in the CytoTrap two-hybrid screen , which was performed according to the manufacturer's instructions ( Stratagene ) . For two-hybrid screens on chick ES cells , poly-A RNA was isolated from ES cells [58] . cDNA was synthesized using Stratagene's cDNA synthesis kit and introduced into pGAD424 vector ( Clontech ) , and this was transformed into XL1-blue MRF' bacteria by electroporation . All plasmids , yeast strains , and media used were purchased from Clontech . The bait ENS-1/ERNI coding sequence was cloned in NdeI/SalI sites of pGBKT7 , introduced into AH109 yeast , and checked for lack of self-activation of the reporter . Screening was performed according to the Yeast Protocols Handbook ( Clontech ) . pGAD424 recombinant plasmids from 18 candidates were purified , of which seven encoded the CHCB2 protein [41] and all included the chromoshadow domain . The smallest one , encoding the 87 carboxy-terminal amino acids , was used in further experiments . The full ENS-1/ERNI coding sequence was cloned into pGADT7 and various truncated forms ( Figure 5 ) subcloned into pGBKT7 . Point mutations were introduced into pGBKT7:ENS-1 using the QuikChange site-directed mutagenesis kit from Stratagene and checked by sequencing . Yeast two-hybrid assays were performed by rapid cotransformation of strain AH109 . The Xenopus Geminin amino acid sequence was used to BLAST the GenBank EST database . The full-length chick homolog sequence was recovered and cloned by PCR from the CytoTrap cDNA library described above . The N- and C-terminal halves of Venus ( aa 1–154 and 155–229 ) were PCR-amplified from pCS2 vectors and cloned into pcDNA3 . 1A . Geminin , ERNI , BERT , and human E2F3 were cloned in frame into the 5′ end of each of the two Venus halves , giving rise to six plasmids expressing each of the three genes fused to either of the two Venus halves . Dlx5 control vectors were a kind gift of Andrew Bailey . COS cells and cES cells were transfected as described [16] , and the cells were observed the next day by epifluorescence in a compound microscope . The method used closely followed one previously described [59] . Briefly , 20 E7 . 5 mouse embryos were fixed in 4% formaldehyde , homogenized in lysis buffer , and sonicated . Cell extracts were harvested by centrifugation , incubated overnight with an antibody against mouse Geminin ( Santa Cruz Biotechnology FL-209 , 5 μg ) , and then immunoprecipitated with Protein-A-Sepharose . Precipitates were heated to reverse the formaldehyde cross-linking . The DNA fragments in the precipitates were purified by phenol/chloroform extraction and EtOH precipitation and used as a template for a PCR , using the following mouse N2-specific primers: forward: AACTCTCATAGCCCTAACTGTC , reverse: CCCTCCTCTCCTAATCTCCTTATGG . After 20 cycles of amplification , one-tenth of the reaction product was used as a template for a second round of a further 20 cycles . The final PCR products were run on a 1% agarose gel .
The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank ) accession number for the chick homolog of Geminin is EU118174 , and for BERT , it is EU118175 . | During early development , when the embryo has three layers of cells ( ectoderm , mesoderm , and endoderm ) , a region of the ectoderm called the neural plate becomes specified to generate the entire nervous system . One of the earliest molecular markers for the neural plate is the transcription factor Sox2 , which is critical for cells to acquire their neural fates and also defines neural progenitor character . We know very little about the intracellular mechanisms by which the neural plate cells acquire these fates . Here , we show that recruitment of transcriptional repressors to chromatin-remodeling complexes regulate the onset of Sox2 expression . Competitive interactions between three proteins , ERNI , BERT , and Geminin , modulate the choice of repressors and regulate Sox2 expression . During gastrulation , when the three embryonic cell layers form , ERNI recruits the repressor HP1γ to prevent Geminin from activating Sox2 prematurely . By the end of gastrulation , this repression is counteracted by competitive binding of BERT to ERNI and Geminin , causing activation of Sox2 . We propose that this mechanism regulates the timing of Sox2 activation in the very early neural plate and thus helps to define the domain that will give rise to the nervous system . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"developmental",
"biology",
"cell",
"biology",
"neuroscience",
"molecular",
"biology"
] | 2008 | A Mechanism Regulating the Onset of Sox2 Expression in the Embryonic Neural Plate |
Various characteristics of complex gene regulatory networks ( GRNs ) have been discovered during the last decade , e . g . , redundancy , exponential indegree distributions , scale-free outdegree distributions , mutational robustness , and evolvability . Although progress has been made in this field , it is not well understood whether these characteristics are the direct products of selection or those of other evolutionary forces such as mutational biases and biophysical constraints . To elucidate the causal factors that promoted the evolution of complex GRNs , we examined the effect of fluctuating environmental selection and some intrinsic constraining factors on GRN evolution by using an individual-based model . We found that the evolution of complex GRNs is remarkably promoted by fixation of beneficial gene duplications under unpredictably fluctuating environmental conditions and that some internal factors inherent in organisms , such as mutational bias , gene expression costs , and constraints on expression dynamics , are also important for the evolution of GRNs . The results indicate that various biological properties observed in GRNs could evolve as a result of not only adaptation to unpredictable environmental changes but also non-adaptive processes owing to the properties of the organisms themselves . Our study emphasizes that evolutionary models considering such intrinsic constraining factors should be used as null models to analyze the effect of selection on GRN evolution .
The genetic basis of organismal evolution is one of the fundamental problems in biology [1]–[7] . The modes of selection for phenotypes would influence the fixation probabilities of the mutations that affect the phenotypes [8] , and the profile of the mutations fixed during the course of evolution would determine the architecture of the genomes and the genetic systems underlying the phenotypes [9] . However , because genetic systems would modify the phenotypic effects of the mutations , the properties of the genetic system would influence the rates and directions of phenotypic evolution as well as the mutational robustness and evolvability [10]–[15] . Therefore , both phenotypes and genetic systems have evolved by mutually influencing each other . Gene regulatory networks ( GRNs ) constitute important parts of such genetic systems and are involved in various biological processes such as environmental responses in unicellular organisms and cell differentiation in multicellular organisms [4] , [16] , [17] . Recent theoretical and experimental studies have revealed that complex GRNs have evolved by successive gene duplication , changes in regulatory interactions , and particularly in prokaryotes , horizontal gene transfer [18]–[20] . In addition , recent studies have addressed the structural features of complex GRNs such as redundancy , scale-free outdegree distributions and exponential indegree distributions [4] , [21]–[24] and the contribution of these features to genetic characteristics such as mutational robustness and evolvability [25]–[29] . One important question with regard to the evolution of complex GRNs is the evolutionary origin of these structural and mutational properties . Various evolutionary processes simultaneously influence GRN evolution and these properties are interrelated . It is thus difficult to identify the factors that have promoted the evolution of these properties , which could evolve as a result of being directly influenced by selection and also incidentally as a result of other factors [30]–[33] . Thus , to identify the factors responsible for the evolution of the properties of complex GRNs , it is necessary to consider not only selection but also various mutational processes and constraining processes . Selection for phenotype is one of the most important driving forces of organismal evolution . However , the impact of phenotypic selection on the evolution of GRNs is unclear . The mode of selection strongly influences the fate of mutations and the profile of mutations fixed during the course of evolution ultimately determines the architecture of GRNs . Thus , it is important to examine how different modes of phenotypic selection would affect the evolution of GRNs . However , there are significant limitations to our general understanding of the processes of adaptation in evolutionary biology . Many previous studies on the evolution of mutational robustness with respect to GRNs have focused on the fixation of phenotypically neutral mutations under stabilizing selection with a constant optimal environment [25] , [34] . On the other hand , the fixation of beneficial mutations for phenotypic adaptation under changing environments is limited [29] . Many studies have suggested that some examples of GRN architectures are related to mutational robustness and evolvability [11] , [26] , [35] , [36] . Theoretical studies have proposed that these genetic properties appear to be evolvable traits [29] , [37]–[40] and that these genetic properties could play a significant role in organismal evolution [41] . However , it is unclear how mutational robustness and evolvability influences the process of GRN evolution . Certain properties of GRN might have evolved through non-adaptive processes such as mutations and biophysical constraints on gene regulation [30] , [31] , [42]–[45] . Mutations in particular is the ultimate source of genetic variation . Thus , the biased properties of mutations can potentially influence the tendency of an organism to evolve . For example , the probability of a transcription factor binding site formation as a result of mutations could vary by several orders of magnitude mainly owing to the extensive variation in the size of potential cis-regulatory regions among organisms [31] , [46] , and the rate of gene deletion could be several times higher than the rate of gene duplication in certain organisms [47] , [48] . Moreover , it has been suggested that the horizontal transfer of regulatory genes is observed to a lesser extent than that of phenotypic genes [20] . Several studies have suggested that certain characteristic features of complex GRNs , such as redundancy and scale-free degree distributions could evolve as an inevitable outcome of mutations [30] , [31] . However , these previous studies have not considered certain essential evolutionary processes such as selection and gene duplication . It is therefore unclear whether such characteristic features of complex GRNs evolved as a result of selection or as a result of the inherent properties of the mutations . The purpose of this study was to identify the evolutionary causes of various structural and mutational properties of complex GRNs , such as redundancy , indegree and outdegree distributions , mutational robustness , and evolvability . For this purpose , we constructed an individual-based model of GRN that dynamically controls gene expression levels and allows populations to evolve under various fluctuating conditions of selection with various kinds of mutations such as gene duplication and deletion , cis- , trans-regulatory mutation and horizontal gene transfer . In this study , to explore selective conditions that promote the evolution of complex GRNs , we first examine the evolution of GRNs under various conditions of fluctuating selection . Second , for showing the adaptive mechanisms for the evolution of complex GRNs , we examine the fitness effect of all the mutations that arose during the evolution . Third , to explore whether internal factors of organisms promote or inhibit the evolution of GRNs , we examined the impact of gene expression cost , constraints on expression dynamics , and several types of mutational biases such as the relative rates of gene duplication and deletion , the possibility of formation of new transcription factor binding sites and horizontal gene transfers . Finally , on the basis of the results of the above analyses , we discuss the major evolutionary causes of various properties of complex GRNs , i . e . , redundancy , scale-free out-degree distributions , exponential in-degree distributions , mutational robustness , and evolvability .
Before presenting the results , we provide a brief description of our model ( see Methods for details ) . The model represents a single regulatory module that controls gene expression in response to specific external stimuli ( Fig . 1A ) . We assume that the populations comprise haploid asexually reproducing individuals . Individuals have their own genomes , and a genome of an individual determines the individual's GRN structure . Individuals of a population at generation = 0 are clonal and have 10 regulatory genes ( R1 , … , R10 ) and 2 phenotypic genes ( P1 , P2 ) and the GRN of an individual has a random structure . The expression levels of each gene are restricted to a range of [0 . 0 , 10 . 0] . The phenotype of an individual is defined as the combination of steady-state expression levels of phenotypic genes . Thus , an individual phenotype is represented as a vector , . Individuals reproduce according to their fitness value . A fitness value of an individual depends on phenotypic selection and the cost of gene expression . The phenotypic selection is defined as a Gaussian function , where an individual phenotype that is closer to an arbitrarily defined optimum has higher fitness ( Fig . 1B ) . When an offspring is produced , various types of mutations such as gene duplication , gene deletion , cis- and trans-regulatory mutation , basal transcription level mutation , and horizontal gene transfer are expected to occur with certain probabilities . Under given simulation conditions , a population is allowed to evolve for 50 , 000 generations , and 60–100 replicated populations are examined under a simulation conditions . Throughout the simulation studies , a set of parameter values is used as a standard set of conditions ( Table 1 ) . Then , in order to examine the influence of a certain factor on GRN evolution , only 1 parameter value is changed while the other parameters are kept at standard values . The standard values are determined by approximating those of yeast because of the availability of appropriate yeast data [49] , [50] . To elucidate the selective conditions for the evolution of complex GRNs , we first examined the evolution of GRN under various conditions of fluctuating phenotypic selection . For that purpose , we compared the structures of GRNs after simulation runs for 50 , 000 generations with standard parameter values under various fluctuating conditions of phenotypic selection . The fluctuation of phenotypic selection was modelled by shifting the position of an optimal phenotype by generation . The initial position of the optimum was set as the phenotype of founding individuals at generation = 0 . We assumed 2 types of optimum shift , a random-walk and a cyclic optimum shift for exploring the impact of the difference in the direction of the optimum shift ( Fig . 1B ) . For both types of optimum shift , we analyzed the optimum shifts with various amplitudes ( d ) and frequencies ( f ) . In the random-walk optimum shift , the optimum shifts away from the previous position by a constant distance ( d ) in a random direction for each 1/f generation . In the cyclic optimum shift , there are 2 alternative optima that are spaced at a constant distance ( d ) , and the optimum is switched from one to another for each 1/f generation . Figure 2 shows the structures of GRNs after 50 , 000 generations of evolution under various fluctuations of phenotypic selection . As a proxy for the GRN structure , we first examined the number of regulatory genes that were responsible for the expression of phenotypic genes ( denoted as core genes in Fig . 1A ) . In our model , not all regulatory genes were responsible for the expression of phenotypic genes because some regulatory genes were not transcribed . An example of such an untranscribed gene is the R5 gene in Fig . 1A ( silent ) . In other cases , regulatory genes did not regulate phenotypic genes either directly or indirectly . An example of this is the R4 gene in Fig . 1A ( pseudo-expression ) . The results show that under the random-walk optimum shift , the number of core and pseudo-expression genes in evolved GRNs increases with the increase in amplitudes ( d ) and frequencies ( f ) of the optimum shifts . However , under the cyclic optimum shift , GRNs with a slightly large number of regulatory genes evolve only when the optimum shift has high amplitude and low frequency . While the random-walk optimum shift with higher amplitude and frequency tends to promote the evolution of complex GRNs , the number of both core genes in the evolved GRN is relatively small when both the amplitude and the frequency are extremely high . To clarify the relationship between the intensity of optimum fluctuation and the evolution of complex GRNs , we analyzed the time-averaged fitness from the 0 generation to the 50 , 000th generation in each population ( Fig . 3A ) . Because the time-averaged fitness of a population becomes smaller as the intensity of optimum fluctuation becomes stronger , the time-averaged fitness of a population may be used as a good indicator of the intensity of optimum fluctuation ( Fig . 3B ) . We examined the relationship between the time-averaged fitness during evolution and the structure of GRNs after evolution ( Fig . 3C ) . The results show that the maximum number of core genes is observed when the time-averaged fitness is at a middle level . The results indicate that the evolution of complex GRNs is most efficiently promoted when the intensity of the optimum fluctuation is moderate . However , the evolution of complex GRNs is disturbed when the intensity of optimum fluctuation becomes too strong . Generally , populations with low fitness would be exposed to a high risk of extinction in nature . Thus , realistically , complex GRNs would evolve under a moderately strong optimum shift , e . g . , small and frequent ( d = 10−1 , f = 10−1 ) optimum shift in a random direction or a large and infrequent ( d = 100 , f = 10−3 ) optimum shift in both random and cyclic directions . On the other hand , simple GRNs would evolve under a small and infrequent optimum shift ( d = 10−3 , f = 10−3 ) , and this selective condition corresponds to a pure stabilizing selection with a fixed optimum . Thereafter , to examine the relationships between GRN structures and mutational properties such as the mutational robustness and evolvability , we examined the phenotypic effect of various types of mutations after simulation runs for 50 , 000 generations . For that purpose , a single mutation was introduced into an individual and then the phenotypes of mutant individuals were compared with those of the original individuals . One thousand randomly chosen individuals in a population were examined for each type of mutation . In addition , to clarify the multilateral aspects of mutational robustness and evolvability , we classified the mutations into 3 types according to their phenotypic effect . The Non-effect mutations cause no phenotypic changes . The Loss-of-phenotype mutations cause loss of the expression level at least one phenotypic gene ( Pi<10−2 ) or prevent the expression from reaching a steady state . Significant mutations cause phenotypic changes but do not also produce the effect of a Loss-of-phenotype mutation . In addition , we measured the size of phenotypic changes caused by Significant mutations . Only the results of mutations against core genes are presented here since mutations against non-core genes generally have no phenotypic effect ( data not shown ) . Figure 4 shows the relationships between GRN structures and the phenotypic effect of mutations in evolved GRNs under various conditions of phenotypic selection . Several tendencies were derived from the results . First , trans-regulatory mutations , gene deletion , and gene duplication have similar effects , and these mutations are unlikely to represent Loss-of-phenotype mutations in complex GRNs . Second , most of the cis-regulatory mutations were Non-effect mutations , while most of the other types of mutations were rarely Non-effect mutations . Third , the extent of phenotypic changes caused by Significant mutations was generally small in complex GRNs . These results suggest that the complex GRNs confer both mutational robustness , i . e . , a low proportion of Loss-of-phenotype mutations ( PL ) and small phenotypic changes in Significant mutations ( DS ) ) as well as evolvability , i . e . , high proportions of Significant mutations ( PS ) and a high mutational target size . On the contrary , simple GRNs have low evolvability . i . e . , low PS and small mutational target size and fragility , i . e . , high PL and large DS . However , because the mutational target size is small in simple GRNs , spontaneous mutations are less likely to arise . Thus , although simple GRNs are fragile when a mutation is artificially introduced such as a gene knockout in the laboratory , they are robust to spontaneous mutations under natural conditions . Then , to confirm whether mutational target sizes and PS are associated with actual evolvability , we examined the rates of phenotypic adaptation in response to a benchmark selective condition by using populations obtained after 50 , 000 generations of evolution ( see Methods ) . The results show that both the mutational target sizes and PS are positively correlated with the rates of phenotypic adaptation ( Fig . S1 ) . Thus , the mutational target size and PS examined in laboratory experiments could be a good indicator of actual evolvability . From the present results , we suggest that the evolvability of a target phenotype in a population could be defined as follows: ( 1 ) where a unit of mutation is a gene in this study . Because mutations occurring in core genes were considered in this analysis , a mutational target size in this analysis is the number of core genes , and the PS value is the probability of Significant mutations occurring with respect to core gene mutations . In this section , we analyze adaptive mechanisms explaining the present results where a certain mode of fluctuating selection remarkably promotes the evolution of complex GRNs . In the present study , complex GRNs have high evolvability , which is positively correlated with mutational target sizes and PS . Because high evolvability is considered to be favorable under conditions of fluctuating selection [29] , [39] , it is possible that the evolution of complex GRNs is promoted by its high evolvability . Thus , we first analyzed the evolution of GRNs without the influence of evolvability . To remove the influence of evolvability , we controlled the mutational target size and PS . The mutations in the present model were assumed to occur at a per-gene mutation rate , so that individuals with large numbers of genes ( i . e . , large mutational target size ) had high mutation rates per individual . Thus , we kept per-individual mutation rates constant regardless of the number of regulatory genes ( see Method for details ) . The results showed that the effect of constant mutation rates per individual is almost the same as the assumption of constant mutation rates per gene ( Fig . S2 ) . Then , to remove the influence of PS , we set PS at 1 regardless of the difference in GRN structures ( PS = 1 , PL = PN = 0; see Methods for detail ) . The results were almost the same as those obtained without controlling the PS value ( Fig . S3 ) . These results indicate that evolution of complex GRNs in our model could be promoted without the influence of evolvability and that the influence of evolvability on GRN evolution might be small . The other possible mechanism for complex GRNs is fixations of beneficial mutations through phenotypic adaptation . To elucidate how the mutations contribute to the phenotypic adaptation , we analyzed fitness effects ( i . e . , a difference in fitness between a mutant and its original individual ) for all the mutations that arose during the 50 , 000 generations of evolution in each population . In this analysis , we removed the cost of gene expression ( c = 0 ) from the original model since we wanted to obtain the fitness effects caused only by the differences in phenotypes . Figure 5 shows the relationships between the intensity of optimum fluctuation and the fitness effects of various kinds of mutations . The results showed that mutations are likely to be beneficial when the fluctuation is at a moderate level ( Fig . 5 red points ) . On the contrary , mutations are likely to be neutral when the fluctuation is strong ( Fig . 5 blue points ) and are likely to be deleterious when the fluctuation is weak ( Fig . 5 black points ) . The results indicate that the evolution of complex GRNs is caused by the fixation of beneficial mutations through phenotypic adaptation . Then , we examined the difference in the number of gene duplications and deletions that showed beneficial fitness effects , because fixation rate of gene duplications need to be greater than those of gene deletion for the evolution of complex GRNs . The results show that gene duplications are more likely to be beneficial than gene deletions particularly when the optimum fluctuation is moderate ( Fig . 6 ) . We then examined the relationship between the number of core genes in the evolved GRN and the number of beneficial gene duplications and gene deletions that occurred during the evolution . The results show that the number of core genes in GRN becomes larger as the number of beneficial gene duplications become more than those of gene deletions ( Fig . 7 ) . These results indicate that the evolution of complex GRNs are promoted mainly by phenotypic adaptations acquired through the more frequent fixation of beneficial gene duplication than through gene deletion . The above analysis showed that the fixation of beneficial gene duplication by phenotypic selection is an important adaptive factor for promoting evolution of complex GRNs . However , the genes included in the complex GRNs of the above analysis are generally too abundant to be regarded as a single regulatory module . Thus , it is reasonable to expect the existence of certain constraining factors to restrict the evolution of complex GRNs in real organisms . We examined the impact of certain examples of internal constraining factors that are inherent in organisms , such as ( i ) the functional constraints on gene expression dynamics , ( ii ) cost of gene expression , and ( iii ) the biased properties of mutations , in the subsequent analysis .
Our study showed that the mode of fluctuation in phenotypic selection has a remarkable impact on the evolution of GRNs . In particular , it was found that fluctuating phenotypic selection with random-walk optimum shift strongly promotes the evolution of complex GRNs with high mutational robustness and evolvability . On the other hand , phenotypic selection with cyclic optimum shift contributes only slightly under limited conditions . By examining a fitness effect of all the mutations during evolution , our study has determined that phenotypic adaptation by beneficial gene duplication represents a major factor that promotes the evolution of complex GRNs . Our study has shown that evolution of complex GRNs is promoted when the intensity of optimum fluctuation is moderate . This phenomenon was thought to occur because the fitness effects of Significant mutations depends upon the intensity of optimum fluctuation ( see Fig . S10 for illustration ) . When the fluctuation is weak , most phenotypic changes produced by mutations are likely to be deleterious because the phenotype of the current population is very close to optimal and mutated phenotypes are more likely to be far from optimal . Thus , phenotypic selection tends to inhibit the fixation of the mutations under these conditions . When the intensity of the optimal fluctuation is moderate , certain phenotypic changes produced by mutations would be beneficial since the mutated phenotypes have a greater chance of being located closer to the optimum than the current population . Thus , phenotypic selection tends to promote the fixation of the mutations under these conditions . When the fluctuation is strong , the position of the optimum would be too far from both the current population and the mutated phenotypes . Thus , most phenotypic changes induced by the mutations would not change the fitness . Such phenotypic changes are selectively neutral and would be fixed only by genetic drift . Thus , phenotypic selection does not play any role in the fixation of mutations under these conditions . We observed the relationship between phenotypic effects ( Significant , Non-effect , and Loss-of-phenotype ) and the fitness effects ( beneficial , neutral , and deleterious ) of mutations . Non-effect mutations are always neutral by definition . Loss-of-phenotype mutations are usually deleterious because the movement of the optimum was assumed to avoid the vicinity of 0 . 0 in our model . On the contrary , Significant mutations could show all three of the fitness effects . Particularly , Significant mutations that cause only small phenotypic changes are not exactly neutral , but mutations having very small fitness effects are known to behave like neutral mutations . Such mutations are referred to as “nearly neutral” . Thus , our study regards the mutations as neutral when their fitness effects are smaller than 10−2 . The exact judgment on near neutrality theoretically depends upon the population size and the selection differential . In our analyses , we did not adopt a precise judgment with respect to neutrality since it was difficult to calculate selection differential precisely . We defined values ranging from 10−4 to 10−1 as indicating neutrality , and the results were qualitatively unaffected by changing these values ( data not shown ) . We analyzed the fitness effect of a mutation at the time of incidence of the mutation in a population . However , because the position of the optimum fluctuates over generations , the fitness effect estimated in our analysis was not a complete indicator for judging the fate of a mutation , particularly when the fluctuation was very frequent . Although such a factor might make it more difficult to detect the fitness effects of the mutation , the present analysis showed high statistical significance . Thus , we believe that the analysis is valid and that the mechanism explained above would operate for most populations of our simulation . Although the detailed mechanisms of how gene duplication is more likely to become beneficial than gene deletions under conditions of fluctuating selection are unclear , we can conclude that there should be differences in the phenotypic effects between gene duplication and deletion . To address this problem , we need to perform detailed analyses with regard to the sizes and directions of phenotypic changes caused by mutations . Fixation of a mutation occurs not only through selection but also through genetic drift . A role of genetic drift in the fixation of a mutation becomes stronger when the efficiency of selection becomes weak . The fixation probabilities of mutations by genetic drift depend on the mutation rates , i . e . , mutations that occur more frequently will be fixed more frequently than the other kinds of mutations . Thus , for the evolution of complex GRNs through genetic drift , the following two conditions must be satisfied: ( i ) gene duplications occur more frequently than gene deletions; ( ii ) selection is ineffective for fixation ( i . e . , very small population size , weak strength of selection , and strong optimum fluctuation ) . This study showed that when the intensity of optimum fluctuation was weak , evolution of complex GRNs was effectively restricted even when gene duplications occur more frequently than gene deletions ( Fig . 13 ) . This indicates that the effects of selection were much larger than those of genetic drift , and thus , the conditions where the evolution of complex GRNs is promoted only by genetic drift might be limited . The complex GRNs not only had a larger number of core genes but also had pseudo-expression genes . In our study , pseudo-expression genes are produced by loss-of-function because of trans-regulatory mutations . These results indicate that phenotypic changes occurring through loss-of-function by trans-regulatory mutations are likely to contribute to phenotypic adaptation . Loss-of-function mutations are generally considered deleterious in molecular evolution . However , our study showed that a loss-of-function mutation could become somewhat beneficial when one of the duplicated genes loses its function under conditions of fluctuating selection . Our study showed that the evolution of GRNs under various selective and constraining conditions would produce not only core genes but also non-core genes . While silent pseudogenes have been commonly observed in various species , only a small number of silent regulatory genes were observed in our model . This might be because the loss of gene expression by basal transcription level mutations and cis-regulatory mutations rarely occurred in our model . Silencing of gene expressions through loss-of-function mutations at the transcription factor binding sites and the promoter regions were commonly observed in real organisms . As far as we know , while the actual rates of these mutations were unknown , the mutations might occur more frequently than those in our simulations . Moreover , most non-core genes in our model were pseudo-expression genes . This might be because these pseudo-expression genes were produced by loss-of-function through trans-regulatory mutations , and the mutations were likely to become beneficial under fluctuating selection in our model . Although such pseudo-expression genes might be wasteful , recent studies have revealed that significant fractions of non-coding RNA are composed of transcribed pseudo-genes [54] , [55] . Thus , the presence of pseudo-expression genes in GRNs in this study is not necessary unrealistic , and the transcribed pseudo-genes present in real organisms might be products of adaptation of gene expression under fluctuating selection . While pseudo-expression genes and silent genes do not involved in functional parts of GRNs , these competent constitute significant part of genomic contents . Thus , our study indicates that the mode of phenotypic selection could influence not only GRN structures but also genomic architectures . While cis-regulatory mutations have been receiving considerable attention in the studies on gene expression evolution , various other mutations , including gene duplications , gene deletions , and trans-regulatory mutations could also influence gene expression through dosage effects . Although dosage effects of gene duplication and deletion have been well recognized , the selective conditions that promote the fixation of these mutations are unknown . Our study demonstrates that these mutations were fixed by selection when the direction of selection was randomly fluctuated . Functional protein dosage increases with gene duplications , but decreases with gene deletions and loss-of-function by trans-regulatory mutations; thus , these mutations appeared to become beneficial more often under fluctuating selection in this model . While our study emphasizes the beneficial aspects of dosage effect , several studies in some multicellular organisms suggested that dosage effects negatively influence fitness . For example , small-scale duplications in some kinds of genes , such as developmental genes and transcription factors , might be limited because a quantitative balance between different proteins through molecular interactions is important for the functioning of these types of genes [56] , [57] . In our analysis , the fixations of mutations that have dosage effects were strongly inhibited even if the emergence of mutations was positively biased when the intensity of optimum fluctuation was small ( Fig . 13 ) . Although our study focused on the selection by external environments , functional constraints through molecular interactions between proteins in a cell seem to be important in real organisms . Thus , it is necessary to consider such biophysical processes in future studies on GRN evolution . While , many studies about the evolution of development and GRN have focused on relatively discrete spatial and temporal changes of gene expression ( i . e . , heterotopy and heterochrony ) where importance of cis-regulatory mutation is proposed to play major role [1] , [3] , [29] , [58] , [59] . Our study focused on a single regulatory module producing the continuous changes of gene expression level in a specific cellular type ( heterometry [60] ) . Such continuous differences in the levels of gene expressions in a specific cellular type are also often correlated with the variations in fitness and quantitative traits in multicellular organisms [61]–[63] . Likewise , a steady-state gene expression level in response to specific environmental stimuli is also correlated with fitness in unicellular organisms [51] , [64]–[66] . In real organisms only a part of GRN is used in a specific condition [67] , and such a condition-specific sub-network appears to be a regulatory module that controls specific cellular function [17] , [68] , [69] . Thus , a significant part of phenotypic evolution could be represented as the quantitative changes in gene expression by single regulatory module . Changes in the direction of selection for phenotypes owing to spatio-temporal environmental fluctuation is one of the major driving forces of organismal evolution [64]–[67]; however , most studies in the field of evolutionary biology are focused on evolution under stabilizing selection with fixed optimum or directional selection in a fixed direction [37] , [65] , [70] . Only a few studies have dealt with adaptation toward a moving optimum [8] or a cyclically fluctuating optimum [29] , [71] . Interestingly , while it is difficult to elucidate historical patterns of fluctuation in the direction of selection , a study on the long-term evolutionary patterns of various quantitative traits from fossil records showed that the most of the traits fit to the evolutionary models of random-walk or stasis rather than prolonged directional selection [72] . Thus , the modes of fluctuation in phenotypic selection assumed in this paper might be plausible . Our results demonstrated that steady-state constraints on GRN expression dynamics could significantly restrict the evolution of complex GRNs . This is because the constraints would decrease the proportion of mutations that could contribute to phenotypic adaptation . A previous study demonstrated that signaling pathways that evolved under constraints for different response dynamics would show the different levels of complexity [73] . This indicates that the strength of constraints might depend on the type of expression dynamics , and the result of that study might be compatible with our results . Our results demonstrated that the fitness load of gene expression costs significantly restricted the evolution of complex GRNs . This is because the cost of gene expression would increase both the deleterious effects of gene duplications and the beneficial effects of gene deletion . Previous studies have suggested that even small costs of gene expression have significant impacts on the evolution of gene expression in microorganisms [51]–[53] . However , the fitness loads of a single gene duplication/deletion might be generally very small; thus , the impacts of expression costs on GRN evolution have not been sufficiently studied . By using individual-based simulations that could deal with very large population sizes , we could demonstrate that even small costs of gene expression could have significant impacts on GRN evolution . Our study showed that some mutational bias had a considerable impact on GRN evolution . The probability of transcription factor binding site formation by regulatory mutations ( Cmut ) mainly affected the number of core genes in GRNs and the shape of indegree distributions in complex GRNs , but not the outdegree distributions ( Figs . 10–12 ) . In particular , some GRNs that evolved with lower Cmut showed exponential distributions as observed in microorganisms , while those with higher Cmut showed single-peaked Poisson distributions ( Fig . 11 ) . These results indicated that the exponential indegree distributions observed in real microorganisms might be due to their small Cmut rather than being a direct product of selection; in addition , indegree distributions of global GRNs in multicellular organisms might be single peaked , although only a part of the GRNs were identified in multicellular organisms . In contrast , the scale-free outdegree distribution depended on GRN complexity ( i . e . , phenotypic selection ) but not on Cmut ( Fig . 12 ) . Thus , scale-free outdegree distributions can be considered as a by-product of complex GRNs that evolved through phenotypic selection . Most studies have focused only on the scale-free feature of biological networks and have ignored the differences between outdegree and indegree distributions . Lynch ( 2007 ) [31] argued that scale-free degree distributions could evolve as a result of mutational bias where the gain rates of regulatory interactions were much smaller than loss rates ( i . e . , low Cmut ) . However , the study presented only an indegree distribution rather than outdegree distributions , and the shape of the indegree distribution appeared to be exponential . Hence , our results might correspond to those of Lynch ( 2007 ) . However , our results might contain some biases since we obtained the degree distributions by assembling separated regulatory modules . If the removal of regulatory interactions between regulatory modules in real GRNs disrupts the scale-free and exponential properties of degree distributions , GRN models that contain multiple regulatory modules would be necessary . On the other hand , if the removal of regulatory interactions between regulatory modules does not disrupt the degree distributions in real GRNs , real GRNs might be regarded as the assembly of complex regulatory modules , even if there are some connections between modules . Although the precise mechanisms for the evolution of these degree distributions were unclear from our study , gene duplications and changes in regulatory interactions by trans-regulatory mutations might be necessary factors for the evolution of scale-free properties . We conducted an additional analysis by changing the rates of cis- and trans-regulatory mutations . The results showed that the decreased rates of cis-regulatory mutation did not affect the number of core genes , indegree distributions , and outdegree distributions ( Fig . S4 , S5 , S6 ) . On the other hand , decreased rates of trans-regulatory mutations decreased the number of core genes ( Fig . S4 ) and disrupted the shape of outdegree distributions ( Fig . S6 ) . These results imply that the gains of regulatory interactions through trans-regulatory mutations might contribute to the increase of core genes and the establishment of scale-free outdegree distributions . In our results , indegree distributions mostly fitted to the Poisson distributions rather than the exponential distribution , and the distribution peaked as Cmut increased . In contrast , outdegree distributions mostly fitted to scale-free distributions and the shape did not depend on Cmut . We hypothesized the mechanisms of these result as follows . The mechanism of indegree distributions: Theoretically , a random network where a link ( input and output regulatory interactions ) between two randomly selected nodes ( genes ) exists at constant probability ( Cmut ) is supposed to have the Poisson distribution for both its indegree and outdegree distributions where the average ( generally denoted as λ ) is equal to the average degree of nodes ( i . e . , λ = N×Cmut , where N is the number of nodes in the network ) . Because an increase of Cmut would increase the average degree of nodes ( N×Cmut ) , the change in the shape of the indegree distributions owing to the changes in Cmut seemed natural from this equation . The Poisson distribution with very low average values ( λ<1 ) is almost indistinguishable from the exponential distribution , and the estimated value of Cmut for organisms in which exponential indegree distribution was reported are generally small . Thus , the reported indegree distributions in these organisms might fit the Poisson distribution rather than the exponential distribution . The mechanism of outdegree distributions: Bhan et al ( 2002 ) showed that the joint effects of node duplication and link rewiring would change the degree distributions from Poisson to scale-free [19] . While the study did not distinguish indegree and outdegree distributions ( i . e . the degree was sum of the indegree and outdegree ) , we presume that both would be scale-free if the distributions were analyzed separately . In addition , while we assumed that the establishment of regulatory interactions depended on the cis-regulatory and coding regions , Bhan's study did not have such assumption . Thus , the differences we observed between indegree and outdegree distributions might be attributed to our assumption . This assumption might disrupt the changes of indegree distributions from Poisson to scale-free even if gene duplication and regulatory mutations occurred . On the other hand , we supposed that the change in outdegree distributions from Poisson to scale-free is due to the joint effects of gene duplication and trans-regulatory mutations in our model , because the change in outdegree distributions from Poisson to scale-free was also disrupted when the rates of trans-regulatory mutations became low ( Fig . S6 ) . To detect the actual mechanisms for degree distribution , more detailed examinations on how various mutations change the regulatory interactions are necessary . Recent studies in yeasts have revealed higher rates of gene duplication and deletion than previously thought [49] and the abundance of copy number variations in some model organisms [74] . In addition , the contributions of copy number variations to gene expression variations have also been elucidated [75] . Surprisingly , the estimation of the expected number of gene duplication and deletion per genome is even higher than those of base substitutions [49] . Furthermore , while relative rates of gene duplication and deletion were said to be biased toward a high deletion rate , the estimated duplication rate is several times higher than the deletion rate [49] . Thus , genetic drift might have a larger effect to promote the complex GRNs in real organisms . In contrast to the relative rate of gene deletion and gene duplication , horizontal transfer of single regulatory genes did not contribute to the evolution of complex GRNs under any conditions of phenotypic selection ( Fig . 14 ) . The results indicate that duplication of regulatory genes is indispensable for the evolution of complex GRNs and also that the minority of horizontally transferred regulatory genes against phenotypic gene in bacterial species were not due to natural selection but due to an inherent property of the mutation . However , we only considered the horizontal transfer of a single regulatory gene in this model . Some studies have reported that functionally related genes are often clustered and that transcription factors and their target genes tend to exist close to each other in a genome [42] , [45] , [76] . Therefore , simultaneous horizontal transfer of transcription factors and their target genes might be necessary for successful horizontal transfer of regulatory genes . Our study demonstrates that various constraining factors inherent in organisms could show significant impacts on GRN evolution . While redundant duplicated genes are common in various species , some microbial organisms , such as Escherichia coli , were known to have only a small number of duplicated genes and very few pseudo genes in their genome . This indicates that these organisms might show mutational biases toward high deletion rates or be living under the strong influence of selection for expression costs . Many studies have suggested that various biophysical factors involved in transcriptional regulations ( e . g . , molecular properties of DNA and proteins , physical structures of the nucleus and chromosomes , spatial arrangements of gene order in genomes , and stochastic noises of chemical reactions ) would be important for the evolution of GRNs and genomic architectures [42]–[45] , [77]–[79] . Thus , these biophysical constraints should be considered for the extension of the model . We would like to emphasize the striking importance of considering constraining factors in evolutionary models to analyze the effect of selection on GRN and genomic evolution . Generally , many evolutionary biologists are often interested in questions such as how differences in genomes or GRNs among species are caused by selection or whether some characteristic properties of GRNs such as network motifs are evolved by selection or not . In the field of molecular evolution , evolutionary models have generally considered various mutational biases ( e . g . , base substitution model ) . In contrast , in the studies on biological network evolution , mathematical random network models have been usually used as null models to detect the effects of selection . Thus , previous studies on network evolution might be ineffective in detecting the effects of selection . To solve this problem , evolutionary models considering these constraining processes ( e . g . , mutation biases and biophysical factors ) must be used as null models to detect the effects of selection . Our study demonstrated that complex GRNs confer high mutational robustness ( i . e . , mutations against core genes are unlikely to cause Loss-of-phenotype and have only a small phenotypic effect ) and evolvability ( i . e . , a larger mutational target size and a mutation are likely to change the phenotype ) ( Fig . 4 ) . In contrast , simple GRNs confer only mutational robustness because of their small mutational target size . Increased core genes in complex GRNs are mostly functionally redundant duplicated genes in our model; thus , the proportion of mutations that cause Loss-of-phenotype seems to be small in complex GRNs . At the same time , an increase of redundant genes might reduce the contribution of each redundant gene to phenotypic expression; thus , the size of phenotypic change by Significant mutations might be small in complex GRNs . On the other hand , mutations against core genes generally unlikely to be Non-effect; thus , a decrease of Loss-of-phenotype mutations in complex GRNs leads to the increase of Significant mutations . Many studies in evolutionary biology have studied the relationship between the mode of fluctuating selection and the evolution of mutational properties , such as genetic canalization and evolvability . Previous studies showed that genetic canalization ( a kind of genetic robustness , which is defined as phenotypic insensitivity to mutation or a lower genetic variance of phenotype ) would evolve under stabilizing selection and cyclically fluctuating selection with particularly small and frequent optimum shift [25] , [71] , [80] . Also , decanalization or higher evolvability would evolve under randomly fluctuating selection and cyclically fluctuating selection with large and infrequent optimum shifts [29] , [39] , [71] . In this study , simple GRNs evolved under conditions where genetic canalization is expected to evolve , while complex GRNs evolved under condition where decanalization is expected to evolve . Because a population continuously needs to follow in the movement of the optimum shift under selective conditions where decanalization was favored , the evolution of complex GRNs was promoted by phenotypic adaptation by gene duplication in these selective conditions . The relationship between robustness and evolvability is a key to understanding how organisms can withstand mutations . However , multiple definitions of mutational robustness and evolvability made it difficult to understand the relationship and evolutionary origins of these features . For example , mutational robustness is defined as a property that reduces the phenotypic and lethal effects of mutations [25] , while evolvability is defined as the ability to promote high evolution rates of an existing trait [11] , [37] and the emergence of a novel trait [27] . These definitions of mutational robustness and evolvability indicate that they are interrelated with each other and include several distinct properties concerning the effect of mutations on phenotype and viability . Because robustness and evolvability are such complex traits , various mutational effects such as Loss-of-phenotype , Non-effect , and Significant should be considered to understand these mutational properties . For example , most quantitative traits in wild populations have substantial genetic variations ( evolvability ) ; however , systematic analysis of gene knockout experiments has revealed that mutations are unlikely to cause lethal outcomes and are likely to show only small phenotypic effects ( mutational robustness ) [81] , [82] . However , the type of biological systems that could consistently achieve both mutational robustness and evolvability and the modes of environmental conditions by which such genetic systems could evolve remain unknown [83] , [84] . Our study revealed that both mutational robustness and evolvability could be consistently achieved by complex GRNs that evolved under randomly fluctuating environments . Genetic canalization has long been regarded as a proxy of mutational robustness in biology [80] . Thus , our results might be confusing since complex GRNs that evolved under the condition of decanalization have higher mutational robustness than simple GRNs that evolved under the condition of canalization in several aspects ( mutations against core genes are unlikely to cause Loss-of-phenotype and are likely to cause only small phenotypic changes ) . In the studies on the evolution of genetic canalization with GRN [25] , [85] , genetic canalization was defined as a smaller average phenotypic effect of mutations . However , because these studies did not distinguish Non-effect and Significant mutations , it is not clear whether the smaller average phenotypic effect of mutations is due to the larger proportion of Non-effect mutation ( i . e . high PN ) or the smaller phenotypic effect of Significant mutations ( i . e . small DS ) . Generally , some genes do not contribute to the expression of other genes when GRNs evolve under stabilizing selection , and these genes are called frozen components [86] , [87] . If frozen components correspond to non-core genes in our model , mutations in frozen components mostly would not affect gene expression patterns of GRNs ( i . e . , Non-effect ) . Thus , the evolution of genetic canalization in previous GRN models showed robustness mainly due to larger proportion of Non-effect mutations rather than to smaller phenotypic effect of Significant mutations . In addition , these studies showed that evolution of genetic canalization in GRNs was associated with the evolution of shorter developmental time to establish a steady-state gene expression pattern [25] , [85] . The results also indicated the evolution of simple GRNs ( smaller number of core genes ) in these models [87] . Additionally , Huerta-Sanchez and Durrett ( 2005 ) revealed that the evolution of genetic canalization in the model was due to the selection for increased viability against mutations rather than phenotypic selection [88] . In other words , selection for increased viability under pure stabilizing selection promotes smaller mutation rates of the phenotype ( i . e . smaller mutational target size ) rather than smaller phenotypic effects of mutations . The importance of mutational robustness in evolution has been pointed out [41] . Robustness mechanisms generally would lessen the number of mutations that show deleterious effects and would increase the number of mutations that potentially contribute to phenotypic adaptation . Thus , robustness mechanisms are considered to have some effects that promote the evolvability of organisms in general . For example , in our analysis , the robustness conferred by redundant duplicated genes and other robustness mechanisms that reduce functional constraints that act on expression dynamics would lessen the proportion of Loss-of-phenotype mutations and increase the proportion of Significant mutations . Consequently , the rate of evolution in the existing trait ( a kind of evolvability ) is increased . Some studies argued that robustness mechanisms would lessen the number of mutations that show some phenotypic effects and would increase the number of neutral mutation . Moreover , these studies argued that the accumulation of such neutral mutations would aid the evolution of a novel phenotype ( another definition of evolvability ) when the environmental or genetic background was changed [83] , [84] . Thus , the decrease of deleterious effects of mutations through some robustness mechanisms , including redundancy , Hsp proteins , posttranscriptional processes , and protein-protein interactions , might have some effects that can promote organismal evolvability in general . Our results showed that the evolution of GRNs could occur when the effects of evolvability are absent ( Fig . S2 and S3 ) . The level of evolvability appeared to saturate at relatively small numbers of core genes ( <10 or so ) ( Fig . S1 ) . Because phenotypic selection strongly promoted the evolution of complex GRNs that had very large number of genes in our model , the effects of evolvability on GRN evolution might not be detected in our study . The level of complexity of a single regulatory module in real organisms is not so high; thus , the selection for evolvability might actually be effective for promoting complex GRNs . Our study did not analyze the effects of evolvability in GRNs with such small number of genes . Estimating the effects of evolvability alone on GRN evolution without the influence of phenotypic selection would be possible if the evolvability of each genotype is examined , and the genotype would be artificially selected according to their evolvability . Then , if evolution of complex GRNs is observed through the analysis , we can demonstrate that selection for evolvability alone could promote complex GRNs . However , such selective conditions might be unrealistic in nature , and selection for evolvability is inevitably coupled with phenotypic selection . Thus , it may be generally difficult to distinguish the effects of these two factors . The role of evolvability in organismal evolution is an interesting subject in understanding the origin of biological diversity . Contrary to the ordinary phenotypes , evolvability is not a property of an individual . Instead , it is a property of a “genotype . ” Because the evolvability of an original genotype itself would change by mutations , it should be applied only for short-term evolution . However , depending on the properties of target systems , e . g . , very low gene duplication rates , the evolvability of an original genotype might be invariant to the mutations and might be applied even for long-term evolution . The mechanism by which properties of evolvability depend on the target systems will be an interesting subject in the future . To analyze mutational robustness and evolvability , we used the phenotypic effects of mutations rather than fitness effects . This is because the phenotypic effects of mutations can be observed in laboratory experiments for real organisms , but the fitness effects of mutations differ depending on external environments and are very difficult to be measured . One of our aims in the present study was to clarify the multilateral aspects of mutational robustness and evolvability by using data available in laboratory experiments . While experimental noise would bring some difficulty in estimating the phenotypic effects , it would be possible to distinguish the effects of mutations such as Loss-of-phenotype , Non-effect , Significant , and also the mutations that change the expression dynamics through examining temporal changes of gene expression or variance of the expression . A distinction between Significant mutations and Loss-of-phenotype mutations in our analysis was actually helpful because the increased number of core genes mainly contributed to the increased number of Significant mutations and decreased number of Loss-of-phenotype , but minorly contribute to Non-effect mutations . Thus , it would be difficult to reveal the relationship between the structure and genetic properties of GRNs without distinguishing between Significant and Loss-of-phenotype mutations in our analysis . We believe that such a distinction between mutations is useful in understanding mutational robustness and evolvability . Moreover , these diverse mutational effects might have fundamental importance in biological evolution . For example , some studies proposed that even mutations that were neutral at the time it arose ( called cryptic genetic variation ) might contribute to phenotypic adaptation because such cryptic genetic variations could contribute the phenotypic variation following changes of genetic and environmental background [89] , [90] . In our simulations , we considered the unsteady dynamics of phenotypic gene expression as lethal . In addition , while the loss of phenotypic gene expression were not assumed to be lethal , the mutation was generally deleterious in our analysis ( data not shown ) because in our simulations , the movement of optimum was assumed to avoid around Pi = 0 . However , the mutations did not necessarily become lethal/deleterious in real organisms . For example , gene essentiality would be reduced under conditions where selective pressure might be weak , such as laboratory conditions or intrabody environments . Moreover , some studies have revealed that even loss of gene expression could be beneficial for phenotypic adaptation under a certain environment in nature [91] . In addition , unsteady expression dynamics might be favorable under fluctuating environments through its increased temporal variance of gene expressions . Recent technological advances have allowed us to perform not only whole genome expression analysis but also analysis of expression dynamics at the single-cell level [92] . By shifting the viewpoint regarding the effect of mutations from the changes in steady-state expression levels to the changes in expression dynamics , we could deal with broader aspects of GRN evolution and could understand organismal evolution in general . Some predictions from our study might provide useful hypotheses that can be tested by experimental data in real organisms . For example , Roth ( 1989 ) conducted an experiment on microbial evolution and showed that a duplication-containing strain was fixed under conditions of growth limitation because of the availability of a carbon and energy source , but the strain was displaced by other strains afterward [93] . Experimental evolution under unexpectedly fluctuating environments might promote increasing number of gene duplication . The present models did not consider several important factors such as duplication of receptor and phenotypic genes , stochastic noise , pleiotropy , and complexity of gene regulation . Duplication and divergence of receptors and target genes are commonly observed in microbial GRN evolution [94] . Extending our models would make it possible to study GRN evolution in broader contexts , such as evolution of new functions , adaptation to novel environment , and evolution of complex phenotypes . Many studies have revealed that even simple genetic networks could show robustness against the noise of gene expressions by means of some local network architectures called network motifs [95]–[98] . Examining how gene expression noise affects the evolution of GRNs under various fluctuating selection conditions would be interesting . In multicellular organisms , many transcription factors work at several developmental stages or in multiple cell types . The pleiotropic property of genes would be necessary for GRNs of multicellular organisms . Molecular interactions between the DNA , transcription factors , and transcription machinery are extremely complex . While simulating all of these interactions is impossible , considering some interactions is necessary to explore their importance . A sequence-based GRN model that considers molecular interactions might aid in detailed quantitative analyses of the evolution of gene expression and GRNs [99] .
The GRN of each individual had M phenotypic genes and N regulatory genes . Each gene was composed of a cis-regulatory region and a coding region . A cis-regulatory region was composed of L cis-sites that are potentially recognized by specific transcription factors ( boxes in Fig . 1A ) , and each cis-site had two parameters called the cis-number and the interaction coefficient . On the other hand , a coding region ( diamonds in Fig . 1A ) had a parameter called the trans-number . The cis- and trans-number values determined which regulatory gene product ( i . e . , transcription factor ) would bind to a cis-site . For example , the product of a regulatory gene with a trans-number of 5 would bind to cis-sites with a cis-number of 5 ( see Fig . S9 for an illustration ) . The value of the interaction coefficient determines the strength of transcriptional activation/repression when a regulatory gene product binds to the cis-site . A cis- and trans-number was assigned an integral number in the range [1 , n] . An interaction coefficient had a real value in the range [−5 , 5] . A cis-number represented a specific DNA sequence of m base pairs . The possible number of motifs ( n; the possible number of colors in Fig . 1A ) produced by m base pairs of DNA sequence was calculated as: ( 2 ) where 1/2 indicates the direction of motifs against the promoter . Multiple binding sites for the same transcription factor were allowed to exist in a cis-regulatory region . However , not all the cis-sites were bounded by transcription factors because the possible number of motifs ( n ) was much greater than the number of regulatory genes that actually existed in a genome ( N ) ; n≫N . Generally , the length of DNA sequences that were recognized by a transcription factor ( m ) was 5–10 bp; thus , we assumed m≈7 . 14 for all the regulatory genes ( this corresponded to n = 9950 ) . A GRN was represented by a dynamic system whose state was represented by the expression levels of the network genes , which were denoted as: ( 3 ) where Ri ( t ) and Pi ( t ) are the expression levels of the regulatory gene i and phenotypic gene i at developmental time t , respectively . The gene expression state at t = 0 is the initial gene expression state . The initial gene expression state for all genes was set at the 0 . 0 expression level . Thus , the initial gene expression state was represented as: ( 4 ) Certain genes were assumed to have a positive basal transcription level ( described below ) , and these genes began to express without transcriptional activation by regulatory genes soon after the beginning of development . The expression level of each gene would change by the following equation: ( 5 ) where Gi ( t ) is the expression level of gene i ( Ri or Pi ) and xi ( t ) is the regulatory input to gene i at developmental time t . The Φ value was defined by: ( 6 ) where K and Emax was constant that determines threshold against regulatory input and the maximum gene expression level and was set at 15 and 10 for all the genes , respectively . This value restricted the expression level to a range [0 . 0 , 10 . 0] for all genes . The regulatory input to gene i at developmental time t was calculated as: ( 7 ) where bi is the basal transcription level of gene i ( bi≥0 ) , L is the number of cis-sites , Bji is the interaction coefficient of the cis-site j of gene i , and Eji ( t ) is the sum of the expression levels of all the regulatory genes that bind to the cis-site j of gene i at developmental time t . We assumed that half of the genes in a GRN at generation 0 had b = 1 , while the other half of genes had b = 0 . We considered the equilibrium steady-state expression levels of phenotypic genes as individual phenotype , which was described as: ( 8 ) The steady state was achieved when the following variance-like criterion was met for all the phenotypic genes ( 9 ) where is the mean expression level of the phenotypic gene i over the developmental time from ( t−50 ) to t , and V determined the degree of steady-state levels that were required for the viable phenotypic expression ( V = 10−4 for standard parameter values ) . In addition , an individual that did not reach the steady state within the developmental time of 500 was considered to be lethal . For the modeling of external signals , we assumed that the R1 gene was a receptor transcription factor . R1 can exist either in the active state ( R1+ ) , which can control transcription , or in the inactive state ( R1− ) , which cannot control transcription . If external signals were present , all products of the R1 gene stayed active throughout the developmental process; however , if the external signals were absent , all products of the R1 gene stayed inactive throughout . Thus , an individual had two phenotypic states: ( in the presence of an external signal ) and ( in the absence of an external signal ) . The fitness value of an individual ( F ) was calculated by ( 10 ) where S is the suitability of the individual's phenotype to the environmental conditions , and Q is the cost of expressing the phenotype . The suitability of phenotype ( S ) was determined by the following Gaussian function: ( 11 ) where D is the Euclidean distance between the phenotype of an individual ( ) and the optimal phenotype ( ) , σ represents the strength of phenotypic selection ( σ = 1 ) , and or is the optimal phenotype in the presence or absence of an external signal , respectively . Because expressing phenotypic genes in the absence of external signals would be wasteful , we assumed for all simulation conditions . On the other hand , the state was assumed to change temporally , as described in the main text , according to fluctuations in external conditions . The cost of expressing the phenotype ( Q ) was described as: ( 12 ) where c is the fitness load per unit of gene expression and Ri ( t ) and Pi ( t ) are the expression levels of the R and P genes i at developmental time t , respectively ( c = 10−5 for standard parameter values ) . Then , the probability of reproduction ( Wi ) that a copy ( i . e . , offspring ) of individual i was created for the next generation was described as: ( 13 ) where Z is the effective population size ( Z = 105 ) . Thus , in creating the next generation , one individual was selected according to the probability , and this procedure was repeated until we got Z viable offspring . When a copy of an individual ( offspring ) was created , mutation would occur at a certain probability . Six types of mutations ( gene duplication , gene deletion , cis-regulatory mutation , trans-regulatory mutation , basal transcription level mutation , and horizontal gene transfer ) were assumed in the model , and the per-gene mutation rates for each type of mutations were denoted as μBTL , μCIS , μTRA , μDUP , μDEL , and μHOR , respectively . When a gene duplication ( or gene deletion ) was assumed to occur , one regulatory gene was randomly copied ( or erased ) along with its cis-regulatory and coding regions . When a cis-regulatory mutation ( or trans-regulatory mutation ) was assumed to occur , a cis-site ( or the coding region ) was randomly chosen and the value of its cis-number ( or trans-number ) was replaced by the value drawn from the uniform distribution of the integer [1 , n] . When a basal transcription level mutation was assumed to occur , a regulatory or phenotypic gene was randomly chosen , and the value of the basal transcription level of the gene ( b ) was increased ( +1 ) or decreased ( −1 ) . When a horizontal gene transfer was assumed to occur , a regulatory gene was randomly created by assigning values drawn from uniformly distributed integer [1 , n] to each cis- and trans-number , and by assigning values drawn from uniformly distributed real number [−5 , +5] . R1 is the receptor for upstream signals; thus , duplication and deletion were not assumed to occur in this gene . Although the per-gene mutation rate of the cis-regulatory region was not estimated , based on the mutation rate per nucleotide per generation ( 10−10 ) and the extent of the cis-regulatory region of genes ( 102–105 bp ) [50] , [100] , the mutation rate including the cis-regulatory region would be in the order of 10−8–10−5 per gene per generation . Background rates of gene duplication and deletion are approximated in the order of 10−6 per gene per cell division ( generation ) in yeast; on the other hand , per-generation mutation rates in multicellular organisms could be 1 to nearly 3 orders of magnitude greater than that in yeast because germ-line cell divisions occur 9 times in nematodes , 36 times in flies , and 200 times in humans [49] . Therefore , the mutation rates used in this study are probably realistic . Because offspring are assumed to be subject to a mutation at per-individual mutation rate regardless of the number of genes , only a single mutation was always introduced to the offspring . The per-individual rate of each mutation was set at values 10 times larger than the per-gene rates because a founder individual has 10 regulatory genes . For setting PS = 1 and PN = PL = 0 , we reintroduced a mutation to the original offspring until the mutation showed Significant phenotypic changes when the offspring were subjected to a mutation . Notably , this procedure did not cause multiple mutations in the offspring . We assumed PS = 1 only for gene duplication , gene deletion and trans-regulatory mutation , and the other type of mutations are assumed as same as the original model . Because , the value of PS in gene duplications , gene deletions and trans-regulatory mutations were well correlated to the structure of GRNs . On the other hand , cis-regulatory mutations and basal transcription level mutations originally had very low value of PS , and the procedure of setting PS = 1 in these mutations would disrupt the evolution of GRNs in unwilling manner . To examine the rate of phenotypic adaptation of an evolved population , a new optimum was placed at a distance away ( d = 1 ) from the mean phenotype of the population . Then , under this benchmark selective condition , the population was allowed to evolve for 1000 generations . During the benchmark evolution , changes in the Euclidean distance between the optimum and the mean phenotype of populations were examined . The probability that a binding site of a particular transcription factor is present in a cis-regulatory region depended on the size ( base pair ) of the binding site ( m ) and the cis-regulatory region ( L ) [70] . The number of DNA motifs ( n ) produced by m base pairs of DNA sequences was calculated as n = ( 1/2 ) ×4m as described by Equation 2 . Thus , the probability of the presence ( Cmut ) and absence ( 1−Cmut ) of a binding site of a particular transcription factor in a cis-regulatory region was represented by the following equation ( see Fig . S9 for an illustration ) . ( 14 ) where L is the number of cis-sites ( i . e . , potential binding sites ) in a cis-regulatory region . The value of L was changed to control the value of Cmut in the simulation . For the standard parameter values , we used L = 100 , n = 9550 ( m = 7 . 14 ) , and , thus , Cmut≈0 . 01 . This probability was applied to all combinations of any transcription factor and any cis-regulatory region in a GRN; therefore , the value of the GRN connectivity density ( C; the proportion of the number of existing regulatory interactions against the number of possible regulatory interactions in a GRN ) tended to approach the value of Cmut if sufficient numbers of regulatory mutation were accumulated . To represent the relative rate of gain and loss of regulatory interactions , Lynch ( 2007 ) used a different parameter , α = μl/μg , where μl and μg are the rate of loss and gain of a transcription factor binding sites , respectively [31] . We approximate that α = μl/μg≈ ( 1−Cmut ) /Cmut . Lynch ( 2007 ) inferred α = 103−102 for prokaryotes , 102−101 for unicellular eukaryotes , and 101−100 for multicellular eukaryotes; this corresponds to Cmut = 10−3–10−2 for prokaryotes , 10−2–10−1 for unicellular eukaryotes , and 10−1–100 for multicellular eukaryotes . To prepare the initial population , we created a founder individual for each population . A founder individual had Mini phenotypic genes and Nini regulatory genes; its GRN structure was randomly generated with a certain connectivity ( Cinit ) . We used Mini = 10 , Nini = 2 , and Cinit = 0 . 5 for the standard simulation condition because the low value of Cinit made it difficult to obtain viable founder individuals . Qualitatively similar results were obtained with the various values of Mini , Nini ( Fig . S7 and S8 ) , and Cinit . We assumed b = 1 for half of the genes and b = 0 for the other half . To create a founder individual with a certain value of Cinit , we determined the range of values for cis- and trans-numbers ( ninit ) according to Equation 14 ( L = 100 , standard parameter value; thus , ninit = 145 for Cinit = 0 . 5 ) . We then initialized the genome of the founder individual by setting the random integral number between 1 and ninit for each cis- and trans-number , and the random real number between −5 and +5 for each interaction coefficient . This procedure assured that the GRN connectivity ( C ) of the founder individual equaled to Cinit . Then , the cis-numbers that were not used for regulatory interactions were rerandomized by setting a random integral number ranging [1 , n]; however , the numbers that were already assigned as trans-numbers were excluded from the rerandomization . This procedure assured the sufficient complexity of the composition of the cis-regulatory regions of the founding individual . We assumed that the viable founder individual should have appropriate phenotypic values , in which the expression levels of all phenotypic genes are <0 . 01 in and >2 . 0 in . | Various organismal traits , including the morphology of multicellular species and metabolism in unicellular species , are determined by the amount and combinations of proteins in the cell . The complex regulatory network plays an important role in controlling the protein profiles in a cell . Recent studies have revealed that gene regulatory networks have many interesting structural and mutational features such as their scale-free structure , mutational robustness , and evolvability . However , why and how these features have emerged from evolution is unknown . In this paper , we constructed an evolutionary model of gene regulatory networks and simulated its evolution under various environmental conditions . The results show that most features of known gene regulatory networks evolve as a result of adaptation to unpredictable environmental fluctuations . In addition , some internal organismal factors , such as mutational bias , gene expression costs , and constraints on expression dynamics , are also important for GRN evolution observed in real organisms . Thus , these GRN features appear to evolve as a result of not only adaptation to unpredictable environmental changes but also non-adaptive processes owing to the properties of the organisms themselves . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"evolutionary",
"biology/microbial",
"evolution",
"and",
"genomics",
"computational",
"biology/transcriptional",
"regulation",
"genetics",
"and",
"genomics/gene",
"expression",
"evolutionary",
"biology/genomics",
"computational",
"biology/evolutionary",
"modeling",
"developmental",
"biology/cell",
"differentiation",
"computational",
"biology/systems",
"biology",
"evolutionary",
"biology/developmental",
"evolution"
] | 2010 | Evolution of Gene Regulatory Networks by Fluctuating Selection and Intrinsic Constraints |
Bacteria use a variety of secreted virulence factors to manipulate host cells , thereby causing significant morbidity and mortality . We report a mechanism for the long-distance delivery of multiple bacterial virulence factors , simultaneously and directly into the host cell cytoplasm , thus obviating the need for direct interaction of the pathogen with the host cell to cause cytotoxicity . We show that outer membrane–derived vesicles ( OMV ) secreted by the opportunistic human pathogen Pseudomonas aeruginosa deliver multiple virulence factors , including β-lactamase , alkaline phosphatase , hemolytic phospholipase C , and Cif , directly into the host cytoplasm via fusion of OMV with lipid rafts in the host plasma membrane . These virulence factors enter the cytoplasm of the host cell via N-WASP–mediated actin trafficking , where they rapidly distribute to specific subcellular locations to affect host cell biology . We propose that secreted virulence factors are not released individually as naked proteins into the surrounding milieu where they may randomly contact the surface of the host cell , but instead bacterial derived OMV deliver multiple virulence factors simultaneously and directly into the host cell cytoplasm in a coordinated manner .
Nosocomial infections contribute $4 . 5 billion to annual healthcare costs in this country alone , with an estimated 2 million nosocomial infections occurring in the US annually , resulting in 99 , 000 deaths [1] . Many of these nosocomial infections are caused by Gram-negative pathogens , and interaction of these pathogens with the host is often mediated by secreted virulence factors . Bacteria have evolved mechanisms for the secretion of virulence factors into the host cell to alter host cell biology and enable bacterial colonization , and these mechanisms typically require that bacteria be in intimate contact with the host . For example , the Type III secretion system ( T3SS ) and Type IV secretion system ( T4SS ) deliver proteins directly into the host cytoplasm from an extracellular bacterial pathogen's cytoplasm [2] utilizing transport machines that act as macromolecular syringes [3] . Delivery of extracellular bacteria or bacterial products can also occur via endocytosis initially into the lumen of the host endocytic compartment , then movement to the host cytoplasm via lysis of the endocytic compartment or delivery of the proteins across the endocytic membrane via the Type III Secretion System ( T3SS ) [3] . For several decades , work by Beveridge's group has characterized bacterial-derived outer membrane vesicles ( OMV ) to be a novel secretion mechanism employed by bacteria to deliver various bacterial proteins and lipids into host cells , eliminating the need for bacterial contact with the host cell [4]–[7] . OMV are 50–200 nm proteoliposomes constitutively released from pathogenic and non-pathogenic species of Gram-negative bacteria [8] , [9] . Biochemical and proteomic analyses have revealed that OMV are comprised of lipopolysaccharide , phospholipids , outer membrane proteins , and soluble periplasmic proteins [8] , [9] . Many virulence factors that are periplasmic proteins are enriched in OMV , for example , Escherichia coli cytolysin A ( ClyA ) , enterotoxigenic E . coli heat labile enterotoxin ( LT ) , and Actinobacillus actinomycetemcomitans leukotoxin [10]–[12] . Beveridge's group and others have reported that some secreted virulence factors from P . aeruginosa , including β-lactamase , hemolytic phospholipase C , alkaline phosphatase , pro-elastase , hemolysin , and quorum sensing molecules , like N- ( 3-oxo-dodecanoyl ) homoserine lactone and 2-heptyl-3-hydroxy-4-quinolone ( PQS ) [6] , [7] , [13] , [14] , are also associated with P . aeruginosa OMV [8] , [9] . Whether these secreted virulence factors packaged in OMV are eventually delivered to the host and the mechanism by which this occurs is currently unknown . A recent study suggested that E . coli OMV fuse with lipid rafts in the host colonic epithelial cell , but the delivery and intracellular trafficking of the OMV cargo was not characterized [15] . Thus , we investigated the possibility that OMV deliver multiple secreted virulence factors into the host cell through a lipid raft-mediated pathway , eliminating the need for intimate contact of the pathogen with the host .
Based on reports that multiple virulence factors are packaged in OMV , we hypothesized that these virulence factors could be simultaneously delivered in a coordinated manner in OMV to the host cell by the microbe . We tracked four P . aeruginosa secreted factors , including alkaline phosphatase , β-lactamase , hemolytic phospholipase C , and Cif , previously reported to be packaged in OMV [6] , [7] , [13] , [14] , [16] . We chose these secreted virulence factors because they play important roles in host colonization , for example alkaline phosphatase promotes biofilm formation [17] , [18] , β-lactamase degrades host antimicrobial peptides , hemolytic phospholipase C is cytotoxic and promotes P . aeruginosa virulence [19] , and Cif is a recently characterized toxin that inhibits CFTR-mediated chloride secretion in the airways [16] and thereby likely reduces mucociliary clearance . To purify OMV from bacterial products not packaged in OMV , like pilus , that may also elicit a host response , we modified a published protocol [14] utilizing high-speed differential centrifugation and density gradient fractionation to isolate OMV from an overnight P . aeruginosa culture supernatant . The Cif protein , as well as a protein present in the membrane of OMV , Omp85 [20] , were identified in purified bacterial-derived OMV ( Figure S1 ) . When airway cells were treated with isolated and purified OMV for ten minutes all four OMV proteins examined were detected in host airway epithelial cell lysate ( Figure 1A ) . By contrast , these virulence factors were not detected in lysates of control cells treated with vehicle ( Figure 1A ) . Therefore , OMV deliver multiple virulence factors to host airway epithelial cells in the absence of bacteria , thus providing a mechanism for bacteria to alter host cell physiology without the need for intimate contact with the host . To explore the significance of OMV in the delivery of virulence factors into the host cytoplasm , we examined the cytotoxic effect of P . aeruginosa OMV on host airway cells using the CellTiter 96 AQueous One cytotoxicity assay . OMV were cytotoxic after a delay of 8 hours ( Figure 1B ) , although virulence factors could be detected in the cytoplasm of host cells after 10 minutes ( Figure 1A ) . The time-dependent increase in cytotoxicity induced by OMV was not dependent on Cif expression in OMV , given that Δcif OMV did not produce a statistically significant difference in cytotoxicity compared to wild-type OMV ( Figure 1C ) . To determine if intact OMV are required for cytotoxicity , purified OMV were lysed with 0 . 1 M EDTA and the lysate was applied to airway epithelial cells for 8 h ( Figure S2 , Figure 1D ) . This method was previously employed by Horstman et al . to effectively lyse E . coli OMV [21] . The lysed OMV did not have a cytotoxic effect on airway epithelial cells , demonstrating that cytotoxicity is mediated by virulence factors delivered into the host cell cytoplasm by bacterial-derived OMV ( Figure 1D ) . In the next series of experiments we began to examine the mechanism whereby OMV deliver virulence factors into the cytoplasm of the host airway epithelial cell . Previously we reported that purified , recombinant Cif , a virulence factor secreted in OMV by P . aeruginosa , is necessary and sufficient to reduce apical membrane expression of CFTR and P-glycoprotein ( Pgp ) in human airway epithelial cells [16] , [22] , thus reducing mucociliary clearance and xenobiotic resistance of the host cells , respectively . In the current study we use Cif as a model protein to investigate how OMV deliver virulence factors into the cytoplasm of human airway epithelial cells . First , experiments were conducted to confirm our previous observation that Cif is secreted in purified OMV and second , to determine if Cif is an intravesicular component of OMV . Cif was detected in OMV derived from P . aeruginosa expressing the cif gene , but not in OMV derived from P . aeruginosa in which the cif gene was deleted ( Figure S3 ) . The inability of Proteinase K ( 0 . 1 µg/ml ) , which does not enter the lumen of OMV , to degrade OMV-associated Cif indicates that this virulence factor is an intravesicular component of OMV ( Figure S3 ) . Thus , these studies demonstrate that Cif maintains an intravesicular localization in purified OMV . Next , studies were conducted to determine if the Cif virulence factor packaged in OMV was functional when delivered to airway epithelial cells . Cif function was measured by examining the ability of Cif to reduce apical plasma membrane CFTR abundance in airway epithelial cells . Purified OMV containing Cif reduced apical plasma membrane CFTR in a time-dependent manner ( Figure 2A ) , whereas purified OMV from P . aeruginosa deleted for the cif gene had no effect on CFTR membrane expression ( Figure 2B ) . Taken together these studies confirm and extend our previous observations that OMV-packaged Cif reduces plasma membrane CFTR . If OMV modulate host physiology of lung cells without direct bacteria-host contact , we would predict that OMV secreted by bacteria should overcome barriers such as the mucus overlying human airway cells [23] . OMV containing Cif reduced CFTR in the apical plasma membrane of airway epithelial cells that have a thick layer of mucus on the apical surface ( Calu-3 cells , Figure 2C ) . A delay in the Cif-mediated reduction of apical membrane CFTR abundance was observed in Calu-3 cells , compared to airway cells that lack a overlying mucus layer , suggesting the mucus only delays OMV from diffusing to host airway epithelial cells . Thus , OMV allow the long distance delivery of secreted bacterial factors to the host cell in the absence of direct bacteria-host contact . We next explored the mechanism whereby OMV deliver bacterial proteins into the host cell using the Cif virulence factor as a model . Based on a recent study showing that Filipin III disrupts E . coli OMV association with host cells [15] , we hypothesized that OMV deliver secreted bacterial proteins to host cells by fusing with lipid raft microdomains . Five minutes after addition of OMV to epithelial cells , Cif , as well as a protein documented to be associated with OMV , Omp85 [20] , were detected in membrane lipid raft fractions ( Figure 3A ) . The lipid raft ( i . e . detergent insoluble membranes ) fractions were separated with density gradient fractionation and characterized by labeling with the flotillin-1 antibody . The fusion of OMV with membrane rafts was observed visually by confocal microscopy using cholera toxin B subunit ( labeled with FITC ) , a documented lipid raft marker [24] , which co-localized with rhodamine-R18 labeled OMV five minutes after OMV were added to the apical side of airway cells ( Figure 3B ) . The rhodamine-R18 dye is quenched when loaded in bilayer membranes at a high concentration and is subsequently dequenched , fluorescing in the red channel upon membrane fusion , which allows dilution of the probe and fluorescence detection . Pearson's correlation and Mander's overlap coefficients demonstrated a high degree of co-localization of cholera toxin B subunit and OMV ( 0 . 771+/−0 . 018 and 0 . 952+/−0 . 012 versus control levels of 0 . 153+/−0 . 026 and 0 . 259+/−0 . 026 , respectively , p<0 . 0001 ) . In further support that Cif-containing OMV fuse with lipid rafts , Cif co-immunoprecipitated with the glycosylphosphatidylinisotol-anchored protein p137 [25] , a documented lipid raft-associated protein , from the lipid raft fraction of airway cells that had been treated with OMV for five minutes ( Figure 3C ) . To determine if host cell lipid raft microdomains are required for OMV fusion , the cholesterol-sequestering agent Filipin III complex was used to disrupt lipid raft domains and OMV fusion was assessed . The rhodamine-R18 dye was utilized to allow visualization and quantitation of OMV fusion with host cells . Rhodamine-R18 only fluoresces upon OMV fusion to host cells , thus an increase in fluorescence is interpreted as an increase in OMV fusion . Rhodamine-R18 labeled OMV applied to the apical membrane of airway epithelial cells produced a time-dependent increase in fluorescence ( Figure 4A ) . In contrast , the fluorescence did not increase above background levels in samples containing only airway epithelial cells or only rhodamine-R18 labeled-OMV ( Figure 4A ) . Filipin III eliminated the fusion of OMV with epithelial cells , indicated by a lack of fluorescence detected when compared to control epithelial cells ( Figure 4B ) . Microscopy studies were confirmed by the quantitative , fluorescence-based assay , described in Figure 4A , which also demonstrated that OMV fusion to the host cell was blocked with Filipin III pretreatment of the host cells ( Figure 4C ) . We next tracked the Cif virulence factor biochemically to determine if lipid raft microdomains are required for virulence factor delivery and function in host cells . Five minutes after OMV are exposed to host airway epithelial cells , the Cif virulence factor is detected by Western blot analysis in the endosomal sub-fraction of the host cell lysate , but not in the cytoplasmic fraction ( Figure 4D ) . The Filipin III complex prevented the appearance of Cif in the endosomal fraction of host cells ( Figure 4D ) and blocked the ability of Cif to reduce apical membrane CFTR ( Figure 4E ) , demonstrating a requirement for the host lipid raft machinery for Cif delivery and function in host cells . Furthermore , disruption of lipid raft microdomains with Filipin III , and thus blocking OMV fusion with airway epithelial cells , reduced the cytotoxicity induced in the airway epithelial cells with 8 h of OMV treatment ( Figure 4F ) . Thus , lipid raft microdomains are required for OMV-mediated delivery and function of secreted virulence factors in host cells . The actin cytoskeleton , in particular the neuronal WASP ( N-WASP ) –initiated actin assembly , is critical to the internalization of select lipid raft-associated cargo [26] , [27] . Based on these previous studies , we investigated the role of the actin cytoskeleton , in general , and N-WASP-mediated cytoskeletal rearrangements specifically , in OMV fusion to the plasma membrane of the airway epithelial cell . Both cytochalasin D ( an actin monomer-sequestering agent ) and wiskostatin ( an inhibitor of neuronal WASP ( N-WASP ) induced actin polymerization ) disrupted the actin cytoskeleton in airway epithelial cells ( Figure 5A ) , resulting in a loss of OMV fusion ( Figure 5B , C ) , as measured by a reduction in OMV-dependent fluorescence in airway cells pretreated with cytochalasin D or wiskostatin . Thus , wiskostatin and cytochalasin blocked OMV fusion to human airway epithelial cells , demonstrating a need for N-WASP induced actin polymerization for OMV fusion to host cells . Wiskostatin and cytochalasin D were also utilized to determine if the actin cytoskeleton is required for OMV delivery of Cif to airway epithelial cells . Cytoplasmic and endosomal fractions were purified from airway epithelial cells pretreated with vehicle , cytochalasin D or wiskostatin in the presence or absence of Cif-containing OMV . In control cells Cif localized to the endosomal fraction , as described above , whereas cytochalasin D and wiskostatin ( Figure 5D ) blocked the entry of Cif into the endosomal and cytoplasmic fractions . Furthermore , wiskostatin pretreatment blocked the Cif toxin-mediated reduction of CFTR from the apical membrane of airway epithelial cells ( Figure 5E ) . Because cytochalasin D changes the rate of CFTR endocytosis , we cannot assess the effects of this inhibitor on Cif virulence factor-mediated reduction of CFTR . In addition , the purified Cif virulence factor alone did not induce morphological changes to the actin cytoskeleton ( data not shown ) . These results reveal that Cif does not alter the cytoskeleton and establishes the requirement for an intact actin cytoskeleton , specifically N-WASP-mediated actin polymerization , for OMV fusion and virulence factor delivery to the host airway epithelial cells . The data above strongly suggest that OMV deliver the Cif virulence factor into the interior of the host cell and allow this virulence factor to associate with an endosomal compartment ( Figures 1A , 4D and 5D ) . To more precisely identify which endosomal compartment was the target of Cif , OMV were applied apically to airway cells for ten minutes , the airway cells were lysed and endosomes were purified by differential centrifugation . From the purified endosomal fraction , Cif co-immunoprecipitated with Rab5 GTPase , a marker of early endosomes , and the early endosomal antigen ( EEA ) -1 ( Figure 6A ) . In contrast , Cif did not co-immunoprecipitate with Rab4 ( a marker of sorting endosomes ) , Rab7 ( a marker of late endosomes ) or Rab11 ( a marker of recycling endosomes ) ( Figure S4 ) . Proteinase K , which does not degrade luminal endosomal proteins but can degrade proteins on the cytoplasmic face of endosomes , eliminated Cif from the endosomal fraction ( Figure 6B ) . As expected , proteinase K did not affect the endosomal association of the transferrin receptor , a luminal endosomal protein that is resistant to proteinase K treatment [28] . However , the transferrin receptor was not resistant to proteinase K degradation in the presence of 0 . 1% Triton X-100 , which disrupts the endosomal membrane and allows proteinase K access to luminal endosomal proteins ( Figure 6B ) . These data reveal that OMV-delivered Cif is localized to the cytoplasmic face of the early endosomes after entry into the epithelial cell . To determine if OMV-delivered Cif enters the host cell cytoplasm by penetrating the membrane of endosomal vesicles , cells were treated with ammonium chloride , a lysosomotropic drug that inhibits vesicle acidification , and thereby inhibits the movement of virulence factors from endosomal vesicles into the cytoplasm [3] . Ammonium chloride had no effect on the ability of Proteinase K to decrease the amount of Cif in the endosomal fractions ( Figure 6C ) , indicating that the Cif virulence factor does not reach the cytoplasm via penetrating intracellular vesicular membranes . Some intracellular bacteria and virulence factors move through the retrograde pathway from endosomes , to the Golgi apparatus and endoplasmic reticulum , from which they enter the host cytoplasm . However , Brefeldin A , a pharmacologic inhibitor of retrograde transport , had no effect on the entry of the Cif into the airway cell and the appearance of Cif in the endosomal fraction ( Figure 6D ) , or the Cif-mediated reduction in apical membrane CFTR abundance ( Figure 6E ) . Thus , our data demonstrate that OMV deliver Cif directly to the host cytoplasm rather than requiring passage across an endosomal membrane or through the retrograde transport pathway . Interestingly , PlcH and alkaline phosphatase also localized to the endosomes after entry into the airway epithelial cells , whereas β-lactamase was detected in the cytoplasmic fraction , as determined by subcellular fractionation and Western blot analysis ( data not shown ) . Thus , virulence factors with differing functions are distributed to different subcellular locations after entry into the host cytoplasm . We propose that rather than secretion of virulence factors into the surrounding medium , OMV are a physiologically- and clinically-relevant mechanism utilized by Gram-negative bacterium , in particular P . aeruginosa , to deliver secreted products into the host cell . In support of this hypothesis , Cif packaged in OMV was 17 , 000-fold more effective than purified , recombinant Cif in reducing plasma membrane CFTR , with 3 ng of Cif in OMV- reducing plasma membrane CFTR expression as effectively as 50 µg of purified , recombinant Cif protein ( Figure 7A ) . Cif was detected in lysates of airway epithelial cells exposed to OMV ( 3 ng Cif ) and 50 µg of recombinant Cif ( Figure 7B ) , but Cif was not detected in cells exposed to up to 10 ng of recombinant Cif , correlating the presence of the virulence factor inside the host cell with virulence factor function . Moreover , airway epithelial cells treated with lysed OMV ( Figure S2 ) showed a dramatic reduction in the ability of the Cif toxin to reduce apical membrane CFTR , as compared to cells treated with intact OMV ( Figure 7C ) . Therefore , OMV-mediated delivery of virulence factors to airway epithelial cells increases the efficacy of these virulence factors in altering host cell physiology .
We have demonstrated that P . aeruginosa OMV deliver multiple virulence factors , simultaneously , into host airway epithelial cells via a mechanism of OMV fusion with host cell lipid raft machinery and trafficking via an N-WASP induced actin pathway to deliver OMV cargo directly to the host cytoplasm . The OMV-delivered Cif virulence factor is then localized to the cytoplasmic face of the early endosomal compartment ( Figure 8 ) . E . coli OMV association with host cells had previously been shown to be sensitive to Filipin III treatment , and thus was proposed to be lipid raft-dependent , but whether the OMV actually delivered cargo into the host cells and the mechanism by which this occurred was not characterized [15] . Fiocca et al . demonstrated that VacA packaged in OMV from H . pylori was internalized into a cytoplasmic vacuole in gastric epithelial cells , but did not investigate a mechanism [29] . We propose the first mechanism for the entry and intracellular fate of OMV-delivered bacterial virulence factors . Considering the pioneering work of Beveridge to characterize OMV and our current mechanistic studies , we propose that OMV-mediated virulence factor delivery should be considered for designation as a secretion system [4]–[7] . Like the T3SS , OMV can deliver bacterial proteins directly to the host cell cytoplasm without releasing the naked bacterial proteins into the extracellular environment where they could be degraded by secreted proteases [30]–[32] . OMV deliver fully-folded , enzymatically-active secreted virulence factors into host cells , ready for immediate action upon delivery . By delivering multiple , active OMV-packaged virulence factors , the pathogen may be able to impact the host on multiple levels . For example , simultaneously altering epithelial cell function by perturbing surfactant abundance or tight junction integrity , and the innate immune response to bacteria by stimulating pro-inflammatory cytokine production [14] , [33]–[36] . Based on our studies in P . aeruginosa and published reports of OMV production by E . coli , H . pylori , A . actinomycetemcomitans , V . cholerae and N . meningitidis , it is likely that other bacteria package multiple secreted virulence factors in OMV for efficient transfer to host cells and thus , the studies proposed here likely represent a general strategy utilized by Gram-negative bacteria in their interactions with the host [10]–[12] , [37] , [38] . In contrast to known secretion systems , OMV-mediated direct delivery of bacterial proteins to the host can occur at a distance , and in the absence of bacteria , thus obviating the need for the pathogen to interact directly with the host cell to cause cellular cytotoxicity and alter host cell biology to promote colonization . Furthermore , OMV can deliver bacterial factors across host barriers , such as mucus layers . We believe that our work should prompt those studying bacterial pathogens to reconsider how secreted virulence factors impact host cells . That is , our data suggest that secreted virulence factors are not released individually into the surrounding milieu where they may randomly contact the surface of the host cell , but are released in a strategic manner , packaged with multiple virulence factors in OMV for coordinated delivery directly into the host cell cytoplasm . It is also possible that OMV provide a mechanism for delivering a concentrated bolus of virulence factors to the host , instead of individual toxins being delivered one at a time to the host cell . Moreover , OMV-mediated , long distance delivery of virulence factors might help explain observations such as , bacterial colonization of catheters causing systemic symptoms in kidney dialysis patients , ocular keratitis occurring in patients who do not have cultivatable pathogens , and the significant lung damage in cystic fibrosis , bronchiectasis , and chronic obstructrive pulmonary disease patients resulting from chronic infections with P . aeruginosa suspended in mucus above the airway epithelium . This mechanism of OMV-mediated protein secretion is reminiscent of the long distance delivery of signaling proteins between and among eukaryotic cells via exosomes [39] , and may represent a general protein secretion strategy used by both pathogen and host .
The antibodies used were: rabbit anti-Cif antibody ( Covance Research Products , Denver , Pa [16] ) ; rabbit anti-OprF antibody ( a generous gift from Nobuhiko Nomura , Graduate School of Life and Environmental Sciences , University of Tsukuba ) ; rabbit anti-pilus antibody ( a generous gift from Michael Zegans , Dartmouth Medical School ) ; goat anti-phospholipase C-H antibody ( a generous gift from Michael Vasil ) ; mouse anti-human CFTR C-terminus antibody ( clone 24-1; R&D systems , Minneapolis , MN ) ; mouse anti-CFTR antibody ( clone M3A7; Upstate Biotechnology , Lake Placid , NY ) ; mouse anti-EEA1 antibody , mouse anti-ezrin antibody , mouse anti-flotillin-1 antibody , mouse anti-Rab5 antibody , mouse anti-actin antibody ( BD Biosciences , San Jose , CA ) ; cholera toxin B subunit-FITC ( Sigma-Aldrich , St . Louis , MO ) ; rabbit anti-GPIp137 antibody ( Abgent , San Diego , CA ) ; Alexa 647-conjugated phalloidin ( Molecular Probes , Carlsbad , CA ) ; rabbit anti-Rab4 antibody , rabbit anti-Rab7 antibody ( Santa Cruz Biotechnology , Santa Cruz , CA ) ; mouse anti-Rab11 antibody , mouse anti-transferrin receptor antibody ( Zymed , San Francisco , CA ) ; mouse anti-β lactamase antibody ( Novus Biologicals , Littleton , CO ) ; rabbit anti-alkaline phosphatase antibody ( GeneTex , Inc . , San Antonio , TX ) and horseradish peroxidase-conjugated goat anti-mouse and goat anti-rabbit secondary antibodies ( Bio-Rad , Hercules , CA ) . Other reagents include: Filipin III complex , ammonium chloride , Optiprep , proteinase K , and cytochalasin D ( Sigma-Aldrich ) , wiskostatin ( Calbiochem , San Diego , CA ) , Triton X-100 ( Bio-Rad , Hercules , CA ) . All antibodies and reagents were used at the concentrations recommended by the manufacturers or as indicated in the figure legends . Two airway epithelial cell lines were studied to examine outer membrane vesicle fusion and toxin delivery to host epithelial cells . First , human bronchial epithelial CFBE cells ( ΔF508/ΔF508 ) were stably transduced with WT-CFTR ( generous gift from Dr . J . P . Clancy , University of Alabama at Birmingham , Birmingham , AL; hereafter referred to as airway epithelial cells ) [40] . CFBE WT-CFTR cells were polarized on 24-mm transwell permeable supports ( 0 . 4-µm-pore size; Corning , Corning , NY ) coated with vitrogen plating medium containing human fibronectin , as described previously [41] . Second , human airway epithelial cells ( Calu-3 ) were obtained from the American Type Culture Collection ( Manassas , VA ) and polarized on 24-mm transwell permeable supports , as described previously [42] . Lysogeny broth ( LB ) was inoculated with P . aeruginosa strain UCBPP-PA14 ( PA14 ) [43] and cultures were prepared as previously reported [16] . OMV were purified using a differential centrifugation and discontinuous Optiprep gradient protocol adapted from Bauman et al . [14] OMV were lysed , when noted , with 100 mM EDTA at 37°C for 60 minutes . To study the localization of the Cif toxin after OMV fusion with the airway epithelial cell , differential centrifugation and fractionation techniques were used to isolate cytosolic and early endosomal compartments . Early endosomes were isolated using a protocol adapted from Butterworth et al . [44] . To characterize proteins interacting with the Cif toxin in lipid raft microdomains , Cif was immunoprecipitated from airway epithelial cell lipid raft fractions by methods described previously [45] . To determine if OMV fuse with lipid raft microdomains of the host , detergent-resistant membranes were purified from airway epithelial cells that had been exposed to OMV . These studies were performed using a discontinuous Optiprep gradient in a protocol adapted from Pike et al . [46] . To monitor the fusion of OMV with airway epithelial cells , OMV were fluorescently labeled with a probe that fluoresces upon membrane fusion . OMV purified with the method described above were resuspended in labeling buffer ( 50 mM Na2CO3 , 100 mM NaCl , pH 9 . 2 ) . Rhodamine isothiocyanate B-R18 ( Molecular Probes ) , which integrates in the membrane of the OMV , was added at a concentration of 1 mg/ml for 1 hour at 25°C , followed by ultracentrifugation at 52 , 000×g for 30 min at 4°C . Rhodamine isothiocyanate B-R18 fluorescence is quenched at high concentrations in bilayer membranes , and fluorescence is dequenched when the probe is diluted upon vesicle fusion . Subsequently , rhodamine labeled-OMV were resuspended in PBS ( 0 . 2 M NaCl ) and pelleted at 52 , 000×g for 30 min a 4°C . After a final centrifugation step , the labeled-OMV were resuspended in 1 ml PBS ( 0 . 2 M NaCl ) containing a protease inhibitor cocktail tablet ( Complete Protease Inhibitor Tablet , Roche ) . Labeled-OMV were applied to the apical side of airway epithelial cells at 1∶4 dilution of labeled-OMV to Earle's Minimal Medium ( MEM , Invitrogen ) and fluorescence was detected over time as indicated on a fluorescent plate reader ( Ex 570 nm; Em 595 nm ) . Fluorescence intensity was normalized for fluorescence detected by labeled-OMV in the absence of airway epithelial cells at the indicated time points . To visualize the fusion and localization of OMV with airway epithelial cells , rhodamine R18-labeled OMV ( see OMV Fusion Assay method ) were applied to the apical membrane of cells and confocal sections were captured over time . Airway epithelial cells were seeded at 0 . 1×106 on collagen-coated , glass-bottom MatTek dishes ( MatTek , Ashland , MA ) and grown for 6–7 days in culture at 37°C . For wheat germ agglutin ( WGA , which labels the plasma membrane ) studies , nonpermeabilized cells were incubated for 5 minutes with Alexa-647 WGA ( 1 µg/ml , 37°C; Molecular Probes ) following 15-minute vesicle incubation . Z-stacks of all labeled cells were acquired with a Nikon Sweptfield confocal microscope ( Apo TIRF 60× oil immersion 1 . 49 NA objective ) fitted with a QuantEM:512sc camera ( Photometrics , Tuscon , AZ ) and Elements 2 . 2 software ( Nikon , Inc . ) . For OMV fusion experiments , a single confocal section ( 0 . 4 µm ) at the apical membrane of the airway epithelial cells is presented . Experiments were repeated three times , with five fields imaged for each experiment . To examine the effect of OMV on the apical membrane expression of CFTR , cell surface biotinylation was performed as described in detail previously by our laboratory [41] . Protein band intensity was analyzed as described previously using NIH image software , version 1 . 63 ( Wayne Rasband , NIH , USA; http://rsb . info . nih . gov ) . To determine if P . aeruginosa OMV are cytotoxic to airway cells , cells were incubated with OMV in serum-free media for the indicated time points . Cytotoxicity was measured using the CellTiter 96 AQueous One Solution Reagent ( Promega , Madison , WI ) , according to the manufacturer's protocol . Statistical analysis of the data was performed using Graphpad Prism version 4 . 0 for Macintosh ( Graphpad , San Diego , CA ) . Means were compared using a Students t-test or one-way ANOVA , followed by a Tukey-Kramer post hoc test using a 95% confidence interval . Data are expressed as means+/−SEM . Cif ( PA2934 , NP 251624 . 1 ) ; PlcH ( PA0844 , YP 792433 . 1 ) ; alkaline phosphatase ( PA3296 , YP 789857 . 1 ) ; β-lactamase ( PA1797 , YP 791446 . 1 ) ; N-WASP ( NP 003932 ) ; Omp85 ( PA3648 , YP 789516 ) ; GPIp137 ( NP 005889 ) . | Gram-negative pathogens are responsible for 2 million annual hospital-acquired infections , adding tremendously to U . S . healthcare costs . Pseudomonas aeruginosa , an opportunistic human pathogen , is commonly associated with nosocomial infections , particularly ventilator-associated infections and pseudomonal pneumonia in immunocompromised patients with cystic fibrosis , chronic obstructive pulmonary disease , ventilator-associated pneumonia , community-acquired pneumonia , and bronchiectasis . We have identified the mechanism for a secretion system that Gram-negative bacteria use to strategically deliver toxins to the host to promote bacterial virulence and host colonization , a pathway that we hope to target to develop new therapies to treat P . aeruginosa infections . Our findings have significant implications for the study of Gram-negative bacterial pathogenesis . We propose that secreted virulence factors are not released individually as naked proteins into the surrounding milieu where they may randomly contact the surface of the host cell , but instead bacterial-derived outer membrane vesicles ( OMV ) deliver multiple virulence factors simultaneously and directly into the host cell cytoplasm in a coordinated manner . This long-distance bacterial communication to the host via OMV is reminiscent of the delivery of signaling proteins and miRNA between eukaryotic cells via exosomes , and may represent a general protein secretion strategy used by both pathogen and host . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/nosocomial",
"and",
"healthcare-associated",
"infections",
"respiratory",
"medicine/respiratory",
"infections",
"cell",
"biology/membranes",
"and",
"sorting",
"infectious",
"diseases/bacterial",
"infections",
"cell",
"biology/cytoskeleton"
] | 2009 | Long-Distance Delivery of Bacterial Virulence Factors by Pseudomonas aeruginosa Outer Membrane Vesicles |
Traditional genetic association studies are very difficult in bacteria , as the generally limited recombination leads to large linked haplotype blocks , confounding the identification of causative variants . Beta-lactam antibiotic resistance in Streptococcus pneumoniae arises readily as the bacteria can quickly incorporate DNA fragments encompassing variants that make the transformed strains resistant . However , the causative mutations themselves are embedded within larger recombined blocks , and previous studies have only analysed a limited number of isolates , leading to the description of “mosaic genes” as being responsible for resistance . By comparing a large number of genomes of beta-lactam susceptible and non-susceptible strains , the high frequency of recombination should break up these haplotype blocks and allow the use of genetic association approaches to identify individual causative variants . Here , we performed a genome-wide association study to identify single nucleotide polymorphisms ( SNPs ) and indels that could confer beta-lactam non-susceptibility using 3 , 085 Thai and 616 USA pneumococcal isolates as independent datasets for the variant discovery . The large sample sizes allowed us to narrow the source of beta-lactam non-susceptibility from long recombinant fragments down to much smaller loci comprised of discrete or linked SNPs . While some loci appear to be universal resistance determinants , contributing equally to non-susceptibility for at least two classes of beta-lactam antibiotics , some play a larger role in resistance to particular antibiotics . All of the identified loci have a highly non-uniform distribution in the populations . They are enriched not only in vaccine-targeted , but also non-vaccine-targeted lineages , which may raise clinical concerns . Identification of single nucleotide polymorphisms underlying resistance will be essential for future use of genome sequencing to predict antibiotic sensitivity in clinical microbiology .
Recent research aimed at finding the genetic causes of beta-lactam resistance in Streptococcus pneumoniae has been focused on laboratory mutagenesis [1]–[4] , sequence comparison [1] , [5] , [6] , and identification of interspecies sequence transfer that promotes penicillin non-susceptibility [7]–[10] . Though these studies have increased our understanding , their resolution is limited , and a whole-genome systematic search in real population settings is still lacking . Indicative of this limited resolution is the frequent use of the term “mosaic genes” to describe pneumococcal resistance alleles [7] . Although recombinational mosaics are clearly identifiable as regions of several hundred nucleotides in resistance genes , it is likely that only a subset of the observed alterations are important in causing resistance . Genome wide association studies ( GWAS ) have been used to identify genetic loci associated with complex diseases ranging from cancer to mental illness in human populations [11]–[13] . While the method should , in theory , be applicable to bacterial populations , its use has been inhibited by significant difficulties . These are primarily due to the clonal population structure , and generally limited recombination within bacteria , which make the causal SNPs indistinguishable from other linked SNPs , effectively creating very large haplotype blocks [14] , [15] . Attempts have been made to take this clonal structure into account in association analyses [16] , [17] , but strong linkage disequilibrium will always restrict the resolution of the approach . To overcome this , studies will require either populations with elevated recombination , a large diverse sample , or both , to make the statistical analysis robust . The confounding effect from clonal population structure may be less problematic in highly recombinogenic bacteria . Homologous recombination brings genetic admixture into bacterial populations in a manner akin to sexual reproduction in humans , although it does not occur every generation , and only affects a small part of the genome in each occurrence . In S . pneumoniae , homologous recombination involves , on average , 2 . 3 kb of chromosomal DNA [18] , about twice the size of an average pneumococcal gene , suggesting that large numbers of recombinational events must accumulate in order to break up linkage blocks smaller than this size . However , the recombination signals left after the action of natural selection are not uniformly distributed across the genome but are concentrated at particular loci , which are commonly known as recombination hotspots . These hotspots are coincident with genes involving the bacterial response to selection pressure , which includes host immune responses and antibiotic utilization , particularly beta-lactams [19] , [20] . We hypothesized that the frequency of recombination at these sites would therefore be sufficient to allow the identification of causal SNPs associated with resistance to beta-lactams , given a large enough sample size . Continuing reduction in sequencing costs has allowed the scale of whole-genome bacterial population studies to increase [20] , [21] , and this should provide more robust statistical power for association analyses . The availability of multiple large bacterial population studies allows a replication of such association studies , providing stronger evidence for common causal SNPs as well as the potential to identify rarer causal SNPs , some of which might only be detected in unique population settings . Here we conducted an association study using the pneumococcal populations from carriage cohorts in Maela refugee camp , Thailand [20] , and Massachusetts , USA [21] , two recent species-wide pneumococcal studies from which large numbers of whole genome sequences and phenotypes for beta-lactam susceptibility are available . Given the high recombination frequency in S . pneumoniae generally , the observed recombination hotspots covering antibiotic resistance genes , and the relatively large sample sizes of both studies , we hoped to overcome the challenges in performing bacterial association studies discussed above . Therefore this study aimed to more precisely identify the sets of variants associated with resistance , where they were located in the genome , and how they were distributed across the population .
We conducted an association study on whole genome SNPs and insertions or deletions ( indels ) to identify variants associated with beta-lactam non-susceptibility . Two rounds of analyses were performed separately using 3 , 085 genomes from pneumococcal strains collected from a carriage cohort in Maela [20] , and 616 genomes from a carriage cohort from Massachusetts [21] . Based on the Clinical and Laboratory Standard Institute guidelines ( CLSI , 2008 ) , strains with penicillin minimum inhibitory concentration ( MIC ) ≤0 . 06 µg/ml were classified as susceptible; applying these cutoffs to our data gave 1 , 729 case ( non-susceptible ) and 1 , 951 control ( susceptible ) samples for our study ( with 21 unknown ) . The Maela and Massachusetts populations comprise strains from multi-lineage backgrounds . Therefore , taking the population stratification into account is essential to separate the clonal population signals from true phenotypic associations . The population substructures utilized were those defined previously [20] , [21] , which in both cases were determined using a Bayesian clustering approach ( see Methods ) . Based on this clustering information , the Cochran-Mantel-Haenszel ( CMH ) association statistic was employed to test for associations between beta-lactam non-susceptibility and specific variants , conditional on the population cluster . For each population , we screened for common alterations with minor allele frequency >0 . 01 and reported sites with a p value<0 . 01 , incorporating a Bonferroni adjustment for multiple comparisons ( Figure 1 , Tables S1–S3 ) . We found 858 and 1 , 721 SNPs associated with beta-lactam non-susceptibility in the Maela and Massachusetts populations respectively ( Figure 2 , Tables S2–S3 ) . Among these , 301 SNPs were found to be associated with non-susceptibility in both populations . Considering that the two settings have different population structures that have evolved independently , these co-detected SNPs represent a set of candidates in which we can have more confidence . Rather counter-intuitively , more candidate SNPs were identified in the smaller dataset from Massachusetts than in the larger dataset from Maela . There are several potential explanations for this observation; one being that it is due to different linkage structure within the two datasets . Not all of the candidate SNPs will necessarily play a causative role; some may be tightly linked to causative SNPs , with insufficient recombination in the dataset to separate them ( here called “hitchhiking” SNPs ) , and hence form part of the same haplotypes . To test this , we estimated the size of the haplotype blocks that harbor candidate SNPs in both the Maela and Massachusetts populations , using the criteria described in Gabriel et al . [22] , [23] ( see Methods ) . Haplotype block sizes detected in the Maela data are significantly smaller than the Massachusetts data ( Mann Whitney test p value 6 . 53×10−9 , Figure S1 ) , indicating that many of the candidate SNPs detected in the Massachusetts data are potential hitchhikers , thereby generating some false positive results . The second potential explanation is due to the population stratification defined previously [20] , [21] . As the clustering analyses were performed separately on each dataset , it is possible that the clustering information from the two datasets are not equivalent in their stringency , leading to a more strict control over population stratification in one population than the other . Nevertheless , 51 candidate loci , comprising a total of 301 discrete and linked SNPs , were co-detected in both the Maela and Massachusetts data; using these should provide a high-stringency data set that overcomes these population-specific effects . The co-detected loci include three intergenic SNPs , and 298 SNPs detected in coding sequences . The latter can be divided into 71 non-synonymous and 227 synonymous SNPs . Of these 51 loci , nine were single SNPs , and 42 were in linkage blocks of between two and 19 SNPs , of which 12 contained only a single non-synonymous SNP . Based on assembled sequences , we also investigated whether or not indels found in associated genes could contribute to the resistant phenotype . None of the identified indels showed significant association with resistance , after Bonferroni correction , at a p-value of <0 . 01 . To estimate how much of the phenotypic resistance in our samples could be explained by the identified SNPs , we performed cross-prediction tests using only the SNPs co-detected in both the Maela and Massachusetts association studies , tested back against each population separately . We found that close to 100% of the resistance in each population could be explained by all of the co-detected SNPs ( Figure S2 ) . Unlike human polygenic traits where each locus contributes only a small effect on the phenotype , each of these bacterial loci appears to have a much stronger effect , and indeed some have been shown experimentally to change the phenotype with only a single variant [1] . This can be demonstrated using odds ratios , which indicate the size of the effect of each associated SNP . While human GWAS studies report a median odds ratio of 1 . 33 per SNP [24] , [25] , our analysis gives a median odds ratio of 11 . 09 per SNP , indicating a stronger effect size . For both the Maela and Massachusetts populations , the percentage of resistance explained plateaued after the addition of approximately 10 loci in any order . This suggests that , at most , about 10 loci are required to make a susceptible strain non-susceptible and that multiple different combinations can achieve this . However , in each resistant isolate , combinations of more than ten loci are commonly detected , perhaps indicating that not all loci are involved in conferring resistance , but that some may play a compensatory role in reducing the fitness cost of resistance variants . In total , the co-detected variants are present in 100% and 98% of the Maela and Massachusetts resistant strains respectively , highlighting that a large proportion of possible resistance variants are captured in our study . For both population settings , loci found to be associated with beta-lactam non-susceptibility show higher enrichment in genes compared to intergenic regions than would be expected by chance ( Fisher's Exact Test p-value<0 . 0001 ) . Candidate loci are not randomly distributed across the whole genome , but clustered within certain genes ( Figure 1 ) . Co-detected loci in both datasets are localized in genes participating in the peptidoglycan biosynthesis pathway , including penicillin binding proteins ( pbp2x , pbp1a , pbp2b ) and two transferases required for cell wall biogenesis ( mraW , mraY ) , the cell division pathway ( ftsL , gpsB ) , heat shock protein and chaperones ( clpL , clpX ) , the recombination pathway ( recU ) and a metabolic gene known to confer resistance to co-trimoxazole ( dhfR ) . Some of these sites , particularly in the pbp genes , matched those previously reported to play an important role in beta-lactam resistance in the literature ( Table S1 ) . To our knowledge , out of 71 non-synonymous SNPs reported here , 43 SNPs are novel and potentially contribute to beta-lactam non-susceptibility in addition to those identified in previous studies . Since most beta-lactam antibiotics work by inhibiting cell wall biosynthesis , it is not surprising to see significant associations between non-susceptible phenotypes and variants in genes participating in the peptidoglycan biosynthesis pathway , including pbp2x , pbp1a , pbp2b , mraW and mraY . Many single amino acid alterations in pbp2x , pbp1a and pbp2b have been previously demonstrated experimentally to increase pneumococcal resistance to beta-lactams . Mutations within or close to the active sites of the transpeptidase domain in penicillin binding proteins have been reported to be associated with penicillin resistance [26]–[28] . By interfering with the formation of a covalent complex between the active site serine and antibiotic molecules , these mutations help reduce the binding affinity of beta-lactam rings to the transpeptidase enzyme . This allows the pneumococci to form a functional cell wall , and thereby become non-susceptible . We observed many predicted loci co-localizing with or surrounding the transpeptidase active sites . These are recognized as three conserved amino acid motifs , SXXK , SXN and KT ( S ) G [1] , [29] and are highlighted as vertical dotted lines in the bottom panel of Figure 1 . Many known structurally characterized alterations in pbp genes have been rediscovered in our analysis , providing independent validation of some of our results . In pbp2x , we observed an association at T338A , which is located next to the active site 337 . The side chain of T338 is required for hydrogen-bonding , and the T338A substitution results in a distortion of the active site [1] . In pbp1a , an alteration from TSQF to NTGY at position 574 , which has been shown to have a lower acylation efficiency in vitro [1] , [30] , was also observed in this analysis . In addition to candidates known to confer structural changes that lead to resistance , we also observed association with E285Q in pbp1a which might contribute to a fitness compensation mechanism caused by resistance in pbp2b [31] . Other functional conformational changes , as well as variants that matched previous observations from sequence comparison , are tabulated in Table S1 [1] , [2] , [5] , [6] , [30]–[38] . Moreover , we observed substitutions outside pbp genes that could potentially affect antibiotic susceptibility , or represent compensatory mutations that interact epistatically with changes associated with resistance . These include the genes mraY and mraW , which encode transferases . Both function upstream of the pbp genes in the peptidoglycan biosynthesis pathway . The genome-wide screen provided us with an opportunity to identify associations outside the pbp genes and the peptidoglycan biosynthesis pathway , which are the direct target of beta-lactams . In both the Maela and Massachusetts datasets , nine independent loci comprising 31 SNPs were detected outside of these pathways . These include amino acid alterations in a major heatshock protein , clpL . Mutants lacking clpL have been previously reported to be more susceptible to penicillin [1] . The effect was attributed to the ability of clpL to interact with and stabilize the Pbp2x protein . In the Massachusetts data alone , we observed associations between resistance and ciaH , a histidine kinase sensor , and ciaR , its response regulator , consistent with previous studies reporting a large increase in resistance due to ciaH mutations . The mutations in the ciaH kinase sensor resulted in hyperactivation of the ciaR regulator , which in turn leads to increased beta-lactam resistance [1] , [39] . Association signals from ftsL and gpsB genes were detected in both datasets . These two genes function in cell division and are essential for complete cell wall formation . Depletion of GpsB leads to cell deformation with a similar phenotype to that observed when the Pbp2x protein is inhibited by methicillin [40] . These identified candidates potentially interact with pbp genes , either directly or indirectly through regulation or participation in cell wall formation; however , experimental characterization will be required to explore the mechanisms of how these alterations influence beta-lactam susceptibility . Interestingly , strong discrete associations were also found in genes where specific variants are known to confer resistance to co-trimoxazole , an antibiotic targeting the bacterial DNA synthesis pathway [41] . We detected associations in dhfR ( encoding dihydrofolate reductase ) and folP ( dihydropteroate synthase ) , which are required for folate synthesis and are essential for nucleotide biosynthesis . Given that beta-lactam and trimethoprim resistance arise from different mechanisms and that the dhfR and folP loci are not genetically linked to any other detected loci , it is curious as to why we observed these signals . A possible explanation could be the contemporaneous use of both beta-lactam and trimethoprim antibiotics in the host populations studied , which would drive co-selection for resistance to the two unrelated classes of antibiotics . In both the Maela and Massachusetts datasets , strains that are phenotypically resistant to beta-lactams are more likely to be phenotypically resistant to co-trimoxazole , suggesting that the two phenotypes did not occur independently ( Fisher's exact test p-value<2 . 2×10−16 , Table 1 ) . Clinical records from Thailand listed beta-lactams and co-trimoxazole as the first and second most frequently prescribed antibiotics for upper respiratory infection treatments [42] , indicating that co-selection pressures may have been possible if the two antibiotics were frequently used together . As some of the variants detected by our study are known to affect the binding affinity for beta-lactams , we looked to see whether the effect would be equivalent across all classes of beta-lactam antibiotics , or if resistance due to specific variants would be greater for certain classes of antibiotic . To test this , we replicated the analysis on the candidate loci using the continuous phenotype of the minimum inhibitory concentration ( MIC ) value for two classes of beta-lactam antibiotics; penicillins and cephalosporins ( here represented by ceftriaxone ) . Penicillins and cephalosporins both possess a characteristic beta-lactam ring , but while the beta-lactam ring is fused to a 5-membered thiazolidine ring in penicillins , it is fused to a 6-membered dihydrothiazine ring in cephalosporins . The drugs also differ in side chains that differentiate their kinetic properties [43] . Figure 3 plots the differential association of each locus to the two beta-lactam antibiotics . Loci with stronger association towards penicillin are distributed along the positive y-axis , while those showing a stronger preference towards cephalosporins are distributed along negative y-axis . We found that loci do not contribute equally to different classes of beta-lactam antibiotics ( Kruskal-Wallis rank sum test , p value<2 . 2×10−16 ) , with a strong association of some loci towards resistance to either penicillins or cephalosporins . Given the known pneumococcal population structure in both the Maela and Massachusetts datasets [20] , [21] , we sought to explore and compare the prevalence of candidate beta-lactam resistance alleles identified as loci co-detected in the two populations . The two populations are composed of multiple pneumococcal lineages , many of which are present in one population but absent in the other . This difference in population structure has a large influence on the types of resistance alleles detected in each setting . Therefore , an unbiased comparison between the Maela and Massachusetts populations can only be made using the lineages common to both locations . PMEN-14 , the globally dispersed multidrug resistant lineage , was detected in both populations ( Maela: 2007–2010 , Massachusetts 2001–2007 ) , and thus allows a comparative view between the two datasets . PMEN-14 isolates from Maela and Massachusetts have significantly different beta-lactam resistance allelic profiles ( Mann-Whitney U test , p value = 4 . 68×10−12 ) . Though the local beta-lactam resistance profiles are different , the pattern of their distribution across the Maela and Massachusetts pneumococci is similar . In both populations , the distribution of resistance alleles is not uniform ( Figure 4 ) . The multidrug resistant lineages PMEN-14 and PMEN-1 , along with other vaccine target lineages appear to carry predicted resistance alleles at a higher frequency . This reflects the vaccine's design to target serotypes associated with antibiotic resistance [21] , [44] . However , levels of beta-lactam resistance have generally remained stable post-vaccine introduction [21] , [45] , [46] , which has resulted from the success of resistant non-vaccine lineages with a high frequency of resistance alleles ( e . g . 35B in Massachusetts , NT in Maela; Figure 4 ) and serotype switching by resistant-vaccine type lineages such as PMEN-14 [21] . In pneumococcal populations dominated by NT lineages such as in Maela , the higher rate of recombination observed in these lineages [20] , and the fact that they are not targeted by current vaccines , may allow them to act as both source and sink for resistance alleles , generating more combinations that are then seeded into the wider population .
The power of phenotype-genotype association studies in bacteria is limited by the clonal population structure and limited recombination within these organisms . One approach to overcoming this is to explicitly account for population structure in the analysis , and this was recently attempted for studies of host-association in Campylobacter [16] and Staphylococcus aureus [17] , but the sensitivity of both was limited by a relatively small sample size . Our analysis used a S . pneumoniae data set of much larger size , enhancing statistical power in detecting associated variants . Our approach also took advantage of the higher level of recombination in the genes participating in the peptidoglycan biosynthesis pathway , some of which are known to be recombination hotspots [19] , [20] significantly reducing the effect of long haplotype blocks within these important genes . Together , this enabled us to identify specific nucleotide variants underlying beta-lactam resistance in this organism , some of which were previously known , but many of which are novel . Our analysis allows the refinement of the understanding of resistance beyond “mosaic genes” to identify likely causative variants , and shows that there are multiple loci which may contribute to resistance . We have also been able to show that , while some loci likely contribute universally to all beta-lactam resistance , some can demonstrate a stronger association with resistance to certain classes of antibiotics more than others . We used this resistance variant dataset to examine the allele distribution within the sampled population . While specific lineages can vary between populations in the resistance loci present , a general finding was that a high frequency of resistance alleles could be found in both vaccine and non-vaccine lineages , a potential explanation for why vaccination has not reduced beta-lactam resistance within the population . Some non-vaccine target lineages with a high frequency of resistant alleles can act as both a source and a sink of resistance alleles within the population . A limitation of our approach is that it is sensitive to recombination frequency and requires a non-clonal population and a large sample size . Although the recombination frequency of different bacteria is relatively fixed , current sequencing technologies do now allow very high sample sizes within bacterial populations , and this may increase the applicability of this approach in the future . The sensitivity of detection of association will also be enhanced by the occurrence of de-novo mutations conferring resistance ( homoplasy; [19] ) , representing convergent evolution . However , this will only be possible using whole-genome sequencing , and these could not be detected by eukaryotic-style marker-SNP-based association studies . Future use of whole-genome sequencing for antibiotic resistance/sensitivity prediction in clinical practice will rely on the ability to assign function to specific variants , rather than mosaic blocks , and this kind of study will be essential to enable these future applications . Nevertheless , results reported from this genome-wide association study are hypothesis-generating and will require further functional validation .
The test populations represent the largest datasets for which whole genome sequences and antibiotic-resistance phenotypes are available - Maela [20] and Massachusetts [21] . Beta-lactam susceptibilities were determined in both datasets by disk diffusion following the CLSI 2008 guidelines [47] . Our analyses contained 1 , 501 non-susceptible , 1 , 568 susceptible and 16 unknown phenotypes in the Maela data; 228 non-susceptible , 383 susceptible and 5 unknown phenotypes from the Massachusetts data . The minimum inhibitory concentrations ( MIC ) of non-susceptible isolates were confirmed by the E-test method ( bioMerieux , Marcy L'Etoile , France ) . Strain names and a full list of MICs from the Maela dataset [20] are given in Table S4 . Strains and metadata for the Massachusetts dataset were given as supplementary data in ref [21] . Samples were previously sequenced as multiplexed libraries on Illumina Hiseq 2000 machines using 75-nt or 100 nt paired-end runs as described in [20] , [21] . Short reads from both studies have previously been deposited in the European Nucleotide Archive under study number: Maela data - ERP000435 , ERP000483 , ERP000485 , ERP000487 , ERP000598 and ERP000599; and Massachusetts data as listed in Table S1 of [21] . Reads from both datasets were mapped onto a single reference genome , S . pneumoniae ATCC 700669 ( European Molecular Laboratory ( EMBL ) accession FM211187 ) [48] using SMALT 0 . 5 . 7 . ( http://www . sanger . ac . uk/resources/software/smalt/ ) . Bases were called from mapped sequences using the methods described in [49] , resulting in 392 , 524 and 198 , 248 SNP calls from the Maela and Massachusetts data respectively . Genomes from Maela were de novo assembled using Velvet [50] with combinations of SSPACE [51] , GapFiller [52] , BWA [53] and Bowtie [54] as in [20] and genomes from Massachusetts were assembled with Velvet exclusively as described in [21] . Assembled sequences allowed variations from insertions and deletions ( indels ) to be incorporated for a deeper analysis at each locus . The Maela and Massachusetts populations represent species-wide data sets; they respectively consist of 65 and 46 different capsule types , and at least 277 and 154 known multilocus sequence types . The population substructures as determined in [20] [21] were used in this analysis . Briefly , whole genome-mapped sequences and concatenated core genome sequences were used in the Maela [20] and Massachusetts data [21] , respectively , as input to the BAPS software [55]–[57] . BAPS was used to define the clonal population structure by estimating the structure based on non-reversible stochastic optimization . The method has successfully been applied to bacterial populations of several different species [58] , [59] . Individual strains in Maela and Massachusetts data were first partitioned into clusters based on multiple runs of the estimation algorithm ( Methods in [20][21] ) . This resulted in 33 and 16 initial clusters for the Maela and Massachusetts data , respectively . Due to the large sample size of the Maela dataset , BAPS was additionally run in a hierarchical manner . As described in [20] , data from each of the primary clusters identified in the Maela data were re-analyzed to obtain secondary clusters within each primary cluster , and these were used to represent the population structure of Maela pneumococci . The haploid bacterial SNP information was treated as human mitochondrial sequence in PLINK v . 1 . 07 [60] and controlled for missing rate and allele frequency . We excluded variants with minor allele frequency <0 . 01 , missingness by strain >0 . 1 and missingness by variants >0 . 1 . For each site , the top two most common variants were parsed to the next analysis to reduce complexity in the test statistic . Intrinsic noise from genetic variation alone can lead to false positive signals . To estimate basal false positive rates and decide a suitable cut-off for each dataset , we separately ran 100 GWAS permutations with true genotypes but randomized binary phenotypes ( Figure S3 ) . None of the permutations of either the Maela or Massachusetts datasets achieved any significant association at p-value 0 . 01 with a Bonferroni correction for multiple testing , therefore validating a Bonferroni-adjusted cut-off at p-value 0 . 01 as our conservative threshold . We first determined SNPs associated with beta-lactam resistance with binary phenotypes: susceptible or non-susceptible . However , the intrinsic clonal population structure of bacteria can result in high false positive rate in GWAS . The tests were thus performed conditioned on the population structure generated by BAPS in previous publications [20] , [21] and controlled for genomic inflation factor . Based on known cluster information , the Cochran-Mantel-Haenszel ( CMH ) test for 2×2×K binary phenotype x variants | population cluster was employed with sites corrected for multiple testing using the Bonferroni correction at a p-value of 0 . 01 . The application of the CMH test reduced the genomic inflation factor from 80 . 16 ( mean chi-squared statistic = 68 . 99 ) to 2 . 56 ( mean chi-squared statistic = 3 . 05 ) in the Maela data , and 13 . 18 ( mean chi-squared statistic = 14 . 17 ) to 3 . 76 ( mean chi-squared statistic = 4 . 73 ) in the Massachusetts data . The reductions in genomic inflation factor seen in both datasets suggest a decrease in false positive rates due to underlying population structure . However , the genomic inflation factors observed here are relatively high compared to those observed in human nuclear chromosome GWAS , suggesting that intrinsic clonal population structure is still an issue for bacterial association studies . Genome wide association studies are sensitive to population stratification . While a stringent stratification helps reduce false positives , it potentially increases false negatives . Due to the size of the Maela dataset , we had available the ( more relaxed ) primary BAPS clusters , and the ( more stringent ) secondary BAPS clusters , and we therefore used these to investigate the effect of clustering size with respect to the number of discovered variants and false positive rate in our data . We separately repeated the Cochran-Mantel-Haenszel ( CMH ) test as described above using information on primary and secondary BAPS clusters as previously defined [20] . We detected greater numbers of variants with significant associations when stratified by primary clusters compared to secondary clusters ( 10 , 451 SNPs compared to 858 SNPs ) . Also , a higher false positive rate was observed in the analyses using the primary clusters than the secondary clusters ( genomic inflation factor of 6 . 58 compared to 2 . 56 ) . This result is consistent with what is expected , reflecting a trade-off between false positives and false negatives , and will be dependent on the sample size and underlying population structure . A high genomic inflation factor indicated that some of the candidate alleles were influenced by population structure and were likely to be hitchhikers . We explicitly tested for linkage disequilibrium between candidate SNPs using Haploview version 4 . 2 [61] . The information was treated as male human X-chromosome to retain its haploidy . Haploview was devised for human genetics where linkages between distant sites are disrupted by crossing-over . Unlike human , bacterial recombination does not necessarily break long distance linkage . We therefore set Haploview to consider all pairwise comparisons under 2 , 200 kb , which is the size of the whole S . pneumoniae genome , thus incorporating all possible linkage predictions into our analysis . Using 95% confidence bounds as described in [22] , a haplotype block was identified as a region with a low recombination rate ( Figure S4 ) . These linkage blocks were used to show the context of the predicted alleles and thus potential limitations of our study . We also compared physical linkage size ( Figure S1 ) detected in the Maela and Massachusetts data . The smaller linkage blocks found in the Maela data suggest a higher likelihood of capturing recombination in the larger dataset and thus separating causative SNPs from hitchhiking SNPs . We plotted the proportion of resistance in the population that could be explained by the co-detected loci in each of the test populations , using only combinations of variants observed in both the Maela and Massachusetts datasets ( Figure S2 ) . The order of loci added was permutated to accommodate all possible combinations . To test whether or not there were variants conferring more specific resistance to certain classes of beta-lactam antibiotics , we repeated the analysis on co-detected candidate SNPs in both datasets and replaced the binary phenotypes with continuous phenotypes: penicillin MIC values and ceftriaxone MIC values . P-values calculated from penicillin MIC and ceftriaxone MIC for each SNP were grouped by the linkage structure . For each BAPS cluster in both the Maela and Massachusetts data , we calculated the mean prevalence of candidate loci by averaging the frequency of linked SNPs detected in each locus per cluster size . All statistical analysis were performed in PLINK version 1 . 07 and R version 2 . 11 . 1 . Graphical representations were created in R . | Streptococcus pneumoniae is carried asymptomatically in the nasopharyngeal tract . However , it is capable of causing multiple diseases , including pneumonia , bacteraemia and meningitis , which are common causes of morbidity and mortality in young children . Antibiotic treatment has become more difficult , especially that involving the group of beta-lactam antibiotics where resistance has developed rapidly . The organism is known to be highly recombinogenic , and this allows variants conferring beta-lactam resistance to be readily introduced into the genome . Identification of the specific genetic determinants of beta-lactam resistance is essential to understand both the mechanism of resistance and the spread of resistant variants in the pneumococcal population . Here , we performed a genome-wide association study on 3 , 701 isolates collected from two different locations and identified candidate variants that may explain beta-lactam resistance as well as discriminating potential genetic hitchhiking variants from potential causative variants . We report 51 loci , containing 301 SNPs , that are associated with beta-lactam non-susceptibility . 71 out of 301 polymorphic changes result in amino acid alterations , 28 of which have been reported previously . Understanding the determinants of resistance at the single nucleotide level will be important for the future use of sequence data to predict resistance in the clinical setting . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"biology",
"and",
"life",
"sciences",
"population",
"biology",
"microbiology"
] | 2014 | Comprehensive Identification of Single Nucleotide Polymorphisms Associated with Beta-lactam Resistance within Pneumococcal Mosaic Genes |
Genome rearrangements often result from non-allelic homologous recombination ( NAHR ) between repetitive DNA elements dispersed throughout the genome . Here we systematically analyze NAHR between Ty retrotransposons using a genome-wide approach that exploits unique features of Saccharomyces cerevisiae purebred and Saccharomyces cerevisiae/Saccharomyces bayanus hybrid diploids . We find that DNA double-strand breaks ( DSBs ) induce NAHR–dependent rearrangements using Ty elements located 12 to 48 kilobases distal to the break site . This break-distal recombination ( BDR ) occurs frequently , even when allelic recombination can repair the break using the homolog . Robust BDR–dependent NAHR demonstrates that sequences very distal to DSBs can effectively compete with proximal sequences for repair of the break . In addition , our analysis of NAHR partner choice between Ty repeats shows that intrachromosomal Ty partners are preferred despite the abundance of potential interchromosomal Ty partners that share higher sequence identity . This competitive advantage of intrachromosomal Tys results from the relative efficiencies of different NAHR repair pathways . Finally , NAHR generates deleterious rearrangements more frequently when DSBs occur outside rather than within a Ty repeat . These findings yield insights into mechanisms of repeat-mediated genome rearrangements associated with evolution and cancer .
Human structural variation contributes to phenotypic differences and susceptibility to disease [1] . Recent studies suggest that many structural variants are mediated by non-allelic homologous recombination ( NAHR ) between dispersed repetitive DNA elements [2]–[5] . While the importance of NAHR in shaping genome structure is becoming more apparent , the mechanism of NAHR remains poorly understood . NAHR ( also known as ectopic recombination ) utilizes the molecular pathways that mediate allelic homologous recombination ( AHR ) between sister chromatids or homologs . AHR and NAHR are both initiated by a double-strand break ( DSB ) that is processed by 5′-3′ DNA resection to generate 3′-OH tailed single-stranded DNA ( ssDNA ) intermediates [6] . The resected ssDNA , called the recipient , is activated to search for homologous sequences , called the donor , to be used as a template for repair . If the recipient is unique DNA , then the donor will be the homolog or sister chromatid , and AHR ensues . However , if the recipient is repetitive DNA , it may choose a non-allelic repeat as a donor , leading to NAHR and potentially a chromosome rearrangement . The establishment of this basic recipient-donor partnership during homologous recombination ( HR ) defines four fundamental parameters for NAHR that we address here . The first parameter is the position of a DSB relative to repetitive and unique sequences . DNA resection starts from the DSB ends and is thought to activate break-proximal sequences before break-distal sequences [7] . Based on this model , break-proximal recipients ( sequences at or near the break site ) direct homology searches before break-distal recipients ( sequences distal from the break site ) . Therefore , a DSB near or in a repetitive element should activate that repeat as a recipient , which may search for a non-allelic donor repeat to promote NAHR . Alternatively , a DSB in a large track of unique sequences should preferentially activate break-proximal unique sequences as recipients . In a diploid , these break-proximal recipients can repair efficiently using allelic donors on the sister chromatid or homolog . Therefore it has been assumed , but never tested directly , that a DSB in unique sequences in a diploid will rarely induce NAHR . However , a few studies in haploid yeast have observed a preference for recombination using more distal sequences over break-proximal recipients , suggesting that break-distal recipients can participate in homology searches [8]–[10] . The second important parameter of NAHR is the percent and length of identity shared between a recipient and potential donors . Introduction of ∼1% sequence divergence between model repeats decreases recombination rates 9- to 25-fold [11] , [12] , suggesting that even very limited divergence may significantly affect NAHR rates . The minimum length of uninterrupted identity between two sequences needed for efficient recombination is called the minimal effective processing segment ( MEPS ) [13] . Using model repeats , the MEPS necessary for efficient NAHR is about 250 bp [14] , [15] . This suggests that small retroelements , such as Alus ( ∼300 bp ) and long terminal repeats ( LTRs; ∼330 bp ) , are potentially sufficient to promote efficient NAHR . However , how homology between natural repeats relates to usage for NAHR has never been assessed at a genome-wide scale . The third important parameter of NAHR is genomic position of a recipient and potential donors . Recipients and donors are more likely to recombine when they are on the same chromosome than when they are on different chromosomes [16]–[18] . Interchromosomal recombination between model repeats can also be influenced by their proximity to centromeres and telomeres [19] , [20] . However , these NAHR position preferences have not been tested with natural repeats in an unbiased system , where the unrestricted choice of repair partners and pathways is allowed . Finally , which HR pathway acts upon a recipient and donor may impact whether NAHR occurs . Single-strand annealing ( SSA ) can occur when resection from a DSB proceeds through flanking direct repeats , exposing complementary sequences that anneal to generate a deletion product [6] . In contrast , Rad51-dependent HR pathways involve strand invasion events where Rad51 polymerizes onto resected recipient DNA to mediate invasion into a homologous duplex donor . When recipient sequences on both sides of the DSB invade the same donor , repair can occur by gene conversion ( GC ) . However , if the recipient shares identity with the donor on just one side of a DSB , then one-ended strand invasion events can repair through break-induced replication ( BIR ) . GC is faster and more efficient at repairing DSBs than BIR [21] . In addition , GC competes effectively with SSA [22] , [23] . While the competition between SSA , GC , and BIR can influence NAHR outcomes , little is known about the relative usage of these pathways during NAHR with natural repeats . Thus the efficiency and outcome of NAHR are potentially influenced by its ability to compete with AHR , the sequence identity between recipients and donors and their genomic position , and the usage of HR pathways . Yet these potential influences remain untested or unresolved , particularly in the context of a family of naturally repeated sequences . To address these fundamental issues , we developed a new genome-wide system to study NAHR between the dispersed and divergent families of Ty retrotransposons in purebred and hybrid diploids of budding yeasts . We exploit this system to provide insight into the most important parameters that control NAHR in a eukaryotic diploid genome .
Ty1 and Ty2 represent the most abundant families of dispersed repetitive elements in S . cerevisiae . Our system to study Ty-mediated NAHR relies on three components: ( 1 ) knowledge of the sequence and position of all Ty1/Ty2 elements in the genome , ( 2 ) strains with genetic features for the recovery of Ty-mediated NAHR events , and ( 3 ) a protocol to measure these events out of all possible outcomes . Below we provide a brief description of each component . As a first step , we completed the sequence of the S . cerevisiae unannotated chromosome III Ty elements ( Figure S1 ) . With the completed sequence , we generated a map of the distribution of 37 full length Ty1s and 13 full length Ty2s [which includes 98 Ty-associated 5′ and 3′ long terminal repeats ( LTRs ) ] , and 208 solo LTRs ( Figure 1A ) . The sequence and positional information is critical since it defines all potential Ty1/Ty2 recipients and donors in the S . cerevisiae genome , allowing us to determine whether some repeats are used and others are not in NAHR . The potential for Ty elements to act as recipients and donors in NAHR depends in part on their sequence identity . The average percent sequence identity is 95 . 7±2 . 4% between Ty1s , 95 . 9±4 . 8% between Ty2s , and 73 . 9±3 . 4% between Ty1 and Ty2 ( Table S3 ) . Previous work has determined that recombination between model repeats decreases 9-fold with 99% identity and 50-fold with 91–94% identity relative to identical model repeats [11] . Thus the sequence divergence of the Ty1/Ty2 family could dramatically reduce the pool of potential Ty recipients and donors , limiting the number of elements that participate in NAHR . However , if the mismatches are clustered , rather than distributed evenly within the full length of Ty1/Ty2 ( 5 . 9 kb ) , then long stretches of identity may allow efficient NAHR . With this in mind , we analyzed the longest block of uninterrupted identity between all pairwise alignments of Ty1/Ty2 , a parameter that has not been previously assessed for Ty elements . To evaluate the significance of these blocks , we categorized them according to the previously determined MEPS value of about 250 bp for NAHR [14] , [15] . Recombination rates are predicted to significantly drop when lengths are below MEPS and proportionally increase when lengths are above MEPS [13] . Using our binning analysis , 73% of all Ty1/Ty2 alignments ( 891 out of 1225 ) have blocks of identity ≥250 bp ( Figure 1B and Table S4 ) . All pairwise comparisons between repeats within either the Ty1 or Ty2 family are above the MEPS value while 31% of pairwise comparisons between Ty1 and Ty2 repeats have a block of identity ≥250 bp . Thus , for the full length Ty1s and Ty2s , the shared blocks of uninterrupted identity strongly predict that a given Ty1/Ty2 recipient can undergo NAHR with many potential Ty1/Ty2 donors , thereby establishing a competition among donors . In contrast , only 1% of all LTR pairwise comparisons ( 544 out of 46 , 665 ) have a block of uninterrupted identity ≥250 bp ( Figure 1C and Table S5 ) . This limited length of uninterrupted identity between the LTRs predicts that they may be inefficient substrates for NAHR . In addition , sequence identity amongst pairwise comparisons of the 306 LTR elements widely range between 3%–100% , with an average of 59 . 6%±22 . 7% ( Table S6 ) . Thus the poor sequence identity between LTRs suggests that solo LTRs will be unfavorable substrates for NAHR . The second component of our system is the use of specific strains to optimize the recovery of Ty-mediated NAHR events . In order to recover all possible NAHR events , we use diploid yeast where loss of genetic material can be complemented by homologs . In contrast , Ty-mediated rearrangements that occur in haploids may delete genes necessary for viability . Along with S . cerevisiae diploids ( referred hereafter as “purebred” ) , we generated synthetic hybrid diploids by mating S . cerevisiae with a sequenced relative , S . bayanus ( referred hereafter as “hybrid” ) ( Figure 2 ) , which is largely devoid of Ty1/Ty2 elements [24] , [25] . The diploids are genetically marked to allow identification of all cells that suffer an I-SceI site-specific DSB as well as the subset of cells in which the broken chromosome is repaired or lost ( Figure 2 and see below ) . Like the purebreds , viability remains high after induction of an I-SceI-induced DSB in the hybrid diploids ( Figure 2 ) . In addition , the hybrid diploids grow well and are competent in DNA maintenance and repair like the purebred diploids ( Figure S2 ) . Since S . bayanus complements almost all the genes in S . cerevisiae [26] , S . bayanus can also balance S . cerevisiae by suppressing any loss of gene function due to NAHR of the S . cerevisiae genome . However , in contrast to the purebred diploids , the hybrid diploids have three important advantages . The significant sequence divergence between the two genomes ( 62% intergenic , 80% genic ) [27] suppresses AHR , favoring NAHR between the more homologous Ty1/Ty2 elements and thus enhancing the recovery of Ty-mediated NAHR events . The sequence divergence also facilitates the analysis of S . cerevisiae rearrangements by array comparative genomic hybridization ( aCGH ) and PCR . Finally , the comparison of NAHR between the purebred and hybrid diploids allows the assessment of NAHR with and without AHR competition ( Figure 2 ) . The third component of our system is an unbiased clone-based assay to determine the frequencies of NAHR events among all possible outcomes ( Figure 3A ) . An I-SceI recognition sequence [referred to as the I-SceI cut site ( cs ) ] , along with a Hygromycin-resistance gene ( HYG ) , is integrated at different positions on the S . cerevisiae chromosome III homolog . We choose to initiate a DSB on the S . cerevisiae chromosome III since this chromosome has the highest density of Ty1/Ty2 elements relative to all other chromosomes ( see Figure 1A ) , making it a good model for the repetitive-rich chromosomes of higher eukaryotes . We initiate the DSB with the addition of galactose to the media for two hours in exponentially growing cultures to induce expression of the I-SceI endonuclease fused to the galactose promoter . Galactose induction of I-SceI expression leads to formation of a DSB at the 163cs position on one S . cerevisiae chromosome III homolog ( Figure 3B ) , which activates recipient sequences adjacent to the break site to undergo a homology search . The cells are then plated onto nonselective YEPD media for individual colonies ( referred to as clones ) . These clones are then phenotyped to determine whether the I-SceI-induced DSB occurred ( HygS , see Figure S3 ) followed by chromosome repair ( Leu+Ura+ or Leu+Ura− ) or loss ( Leu−Ura− ) . We find that the majority of I-SceI-induced DSBs are repaired in both the purebred ( 99±2% ) and hybrid ( 79±5% ) diploids , although the hybrid diploids exhibit a significant increase in chromosome loss ( 20±5% ) compared to the purebred diploids ( 1±2% ) ( Figure 3C ) . HR mediates almost all of this DSB repair in both diploids since repair is nearly abolished when the essential HR protein Rad52 is absent ( Figure 3C ) . To assess the structure of the repaired chromosome in the two genetic repair classes , a random subset of clones in each class are further analyzed by pulse-field gel electrophoresis ( PFGE ) /Southern analysis ( Figure 3D ) . An I-SceI-induced DSB at the 163cs position that is repaired by AHR results in an unchanged chromosome III size whereas repair by NAHR results in a rearrangement with a changed chromosome III size ( Figure 3D ) . Further aCGH and PCR characterization of the genetic repair classes reveals four types of chromosome III rearrangement structures with Ty elements localized to the recombination junctions ( Figure S4 and see Materials and Methods ) . The Leu+HygSUra+ repair class I contains internal deletions , and the Leu+HygSUra− repair class II includes isochromosomes , rings , and translocations ( see schematics in Figure 3A ) . The recovery of these distinct Ty-mediated NAHR rearrangements from one site-specific DSB reveals a competition between recipient and donor Ty elements for NAHR , validating our system as a means to study NAHR between complex families of natural repeats . A site-specific DSB in unique DNA allows us to assess the likelihood that break-distal repeats are activated as recipients in a homology search to facilitate NAHR . HR events that use a break-distal recipient for recombination are termed here as break-distal recombination ( BDR ) . With 163cs positioned inside 18 . 1 kb of unique DNA on chromosome III ( see map in Figure 3A ) , we tested the possibility for BDR by monitoring three potential Ty recipient loci ( YCRCdelta6 , YCRCdelta7 , RAHS ) at various distances distal from the break site . Because our assay employs no selection , we are able to calculate the frequencies of I-SceI-induced Ty-mediated rearrangements among all possible outcomes after the DSB ( see Materials and Methods ) . Below we highlight the major points from the data compiled in Table 1 and Table 2 . In purebred diploids , 17% of cells after DSB at 163cs undergo NAHR through BDR to mediate rearrangements . Despite a sufficient length of unique sequences that can facilitate AHR with the identical homolog after the DSB , 15±6% of cells use the RAHS recipient , 0 . 3±0 . 3% of cells use YCRCdelta7 , and 2±0 . 7% of cells use the YCRCdelta6 recipient located 11 . 7 kb , 28 . 9 kb , and 47 . 5 kb distal from the DSB , respectively ( Figure 4A ) . To test the robustness of BDR , we changed a number of parameters . We eliminated the nonhomology immediately at the DSB ends ( 1 . 6 kb I-SceIcs/HYG construct ) to test whether BDR is due to the presence of nonhomologous ends , which may inhibit the coordination of two-ended strand invasion events during GC [28] . However , with identity at the DSB ends , BDR is still observed , generating rearrangements ( Figure S5 ) . We further tested if BDR was specific to the 163cs position by moving the position of the DSB more centromere-proximal . With the I-SceI-induced DSB at 147cs , BDR-mediated Ty rearrangements occur in 3±3% of cells after DSB ( Figure 4A ) . Interestingly , the frequency of YCRCdelta6/YCRCdelta7 usage is similar to when a DSB initiates at 163cs , suggesting that the usage of these LTR recipients is not determined by their distance from the break site . Lastly , we tested if BDR occurs when the I-SceI-induced DSB initiates on a different chromosome . BDR still occurs in 8±4% of cells after formation of a DSB on S . cerevisiae chromosome V to generate Ty-mediated rearrangements ( Figure S6 ) . Thus distal repeats mediate BDR despite the presence of break-proximal unique DNA that can effectively facilitate AHR . This result suggests that unique and repetitive recipient sequences at least 47 . 5 kb distal to a DSB can participate in recombination . To test whether AHR competes with BDR , we analyzed BDR in the hybrid diploids . In the hybrid diploids , AHR is mostly suppressed compared to purebred diploids ( 3±4% of cells after DSB in hybrid compared to 82±6% of cells after DSB in purebred , Figure 4B ) , as expected from the extent of divergence between S . cerevisiae and S . bayanus genomes . Under these conditions of suppressed AHR , the frequency of BDR increases 4 . 5-fold compared to purebred diploids ( increasing from 17% to 76% , Figure 4B ) , indicating that BDR competes with AHR . Furthermore , the distribution of different BDR-mediated rearrangements remains the same between hybrid and purebred diploids ( compare Figure 4C to Figure 4A , and Table 1 ) . Thus the presence of a divergent homolog at the break site enhances BDR-mediated rearrangements but does not alter preferences of Ty recipient and donors on chromosome III . This aspect of hybrid diploids makes them an excellent model to investigate the features of the recipients and donors that give rise to their preferred use . To begin to define the parameters that influence the preferred use of recipient sequences to repair a DSB , we determined the largest block of uninterrupted identity between the recipient and its donor . The DSB at 163cs is positioned in the right arm of chromosome III distanced 57 . 4 kb from the centromere and 165 . 6 kb from the right telomere . Thus for AHR in purebred diploids , there is >50 kb of identity with the homolog on both sides of the DSB . In contrast , among the BDR events , the largest block of uninterrupted identity with the donors is 1 , 877 bp for the RAHS recipient , 29 bp for YCRCdelta7 recipient , and 98 bp for the YCRCdelta6 recipient . This reveals that the homology search in purebred diploids can be efficiently directed by 0 . 1% , 0 . 2% , or 3% ( 29 , 98 or 1 , 877 bp out of 57 , 453 bp ) of the potential recipient sequences activated by the DSB , and that this small fraction very distal to the break site generates rearrangements in a total of 17% of cells after DSB . In addition , the smaller and more break-distal solo LTRs , YCRCdelta6 and YCRCdelta7 , compete effectively with the larger and more break-proximal RAHS cluster in both purebred and hybrid diploids ( see Figure 4A and Figure 4C ) . These data are consistent with our analysis of AHR in hybrid diploids , where the recombinant junctions occur both proximal and distal to the break site ( data not shown ) . Moreover , these hybrid allelic junctions do not coincide with the longest length of uninterrupted identity ( 138 bp ) found between potential recipients through S . cerevisiae and S . bayanus chromosome III alignments . Thus the relative effectiveness of repetitive and unique recipient sequences competing next to the DSB is not solely predicted by length of uninterrupted identity or distance from the DSB . Our characterization of Ty-mediated NAHR events also allowed us to investigate the preferred usage of Ty donors with a DSB at 163cs . Intrachromosomal Ty sequences are used as donors in 75±4% of hybrid and 17±6% of purebred cells after DSB at 163cs , generating internal deletions , isochromosomes , or chromosome rings ( intra-NAHR in Figure 5A and Table 1 ) . In contrast , only 1±0 . 7% and 0 . 3±0 . 3% of cells after DSB at 163cs produce Ty-mediated interchromosomal translocations in hybrid and purebred diploids , respectively ( inter-NAHR in Figure 5A and Table 1 ) . Thus despite the greater number of potential inter- than intrachromosomal Ty donors ( see Figure 1A ) , Ty donors on the same chromosome are preferred approximately 50 times more than Ty donors on a different chromosome . Again as a first assessment , we wondered whether the NAHR biases for intra- over interchromosomal donors and amongst the two intrachromosomal donors ( LAHS and FRAHS ) are dictated by sequence identity between the donors with its Ty recipient . We generated a ranked list of sequence homology , comparing the three Ty recipient elements distal to 163cs ( YCRCdelta6 , YCRCdelta7 , RAHS ) with all potential Ty donor elements in the genome . We find that the intrachromosomal Ty donors ( LAHS and FRAHS ) are not among the most identical by either percent sequence identity or the longest block of uninterrupted identity ( Figure 5B and Table S7 , Table S8 ) . Of the intra-NAHR Ty partners , we also find no correlation with the extent of sequence homology between the chosen Ty donors and their frequency of usage . For example , in the hybrid diploids , 61±3% of cells after DSB generate internal deletions between RAHS and YCRWTy1-5 at FRAHS ( 97% identity , 1 , 635 bp largest block of uninterrupted identity ) whereas only 3±1% of cells after DSB generate a chromosome ring between the same RAHS recipient and the LAHS donor ( 97% identity , 1 , 877 bp largest block of uninterrupted identity ) . Furthermore , relaxing the stringency for sequence identity in NAHR using msh2Δ/msh2Δ , msh6Δ/msh6Δ , and sgs1Δ/sgs1Δ mutants in hybrid diploids does not abolish the intrachromosomal donor preference ( Figure 5A ) , further suggesting that the preferred usage of donors is not due to sequence identity [29] , but donor position . Similar to the findings for the usage of recipient sequences for NAHR , the preferred usage of Ty donors is neither dictated nor can be predicted by sequence homology . Thus the primary determinant of Ty donor choice during NAHR is genomic position , with ∼50-fold preference for intrachromosomal over interchromosomal donors . Sequence homology between the Ty1/Ty2 families failed to dictate the recipient and donor competition during NAHR . One explanation is that each Ty-mediated rearrangement requires different genetic factors ( Table 1 ) , suggesting that they are generated through distinct NAHR pathways . Since HR pathways are known to compete after a DSB , we examined how this competition affected recipient and donor choice . In the hybrid diploids with the I-SceI-induced DSB in unique sequences at 163cs , 61±3% of cells form internal deletions between the RAHS recipient and the FRAHS donor ( Table 1 ) . These deletions form independent of RAD51 suggesting they occur through SSA ( Table 1 ) . RAHS also mediates isochromosomes ( 3±1% ) and rings ( 3±1% ) with the LAHS donor , and translocations with interchromosomal Ty donors ( 1±0 . 7% ) , all of which have Rad51-dependencies ( Table 1 ) . Thus the same RAHS recipient mediates internal deletions 20–40 fold higher than isochromosomes , rings , or translocations , suggesting that SSA dominates the NAHR pathway choice to generate Ty-mediated rearrangements when a DSB occurs in unique sequences . With at least four NAHR pathways operating after the DSB at 163cs ( suggested by the different genetic dependencies of the Ty-mediated BDR rearrangements , see Table 1 ) , we then asked if these NAHR pathways were in competition with one another . To address pathway competition , we attempted to abolish or enhance particular NAHR pathways by removing their intrachromosomal donors and/or repositioning the I-SceIcs in the hybrid diploids . We then compared changes in the frequencies of the Ty-mediated rearrangement product as a readout of their NAHR pathway , where compensatory effects indicate competing pathways . In addition , since Rad51-independent SSA and Rad51-dependent pathways have been shown to compensate for each other after a DSB and hence compete [22] , [30] , our analysis groups the NAHR pathways into these two distinct HR mechanisms . We first eliminated the dominant SSA pathway by deleting the FRAHS donor ( FRAHSΔ , B in Figure 6 ) and looked for compensation through the remaining rearrangements . These rearrangements are grouped as Rad51-dependent NAHR since rings show full Rad51-dependency while isochromosomes and translocations have partial Rad51-dependency ( Table 1 ) . While some Rad51-dependent rearrangements show a modest increase ( rings increase 3±1% to 11±3% , Table 2 ) , the majority of cells cannot repair the DSB at 163cs without SSA , resulting in chromosome loss ( 71±3% loss , Figure 6 ) . One possibility for this repair inefficiency is that the DSB is too far from the Ty recipients ( at least 11 . 7 kb from the break site ) to effectively activate the recipients in Rad51-dependent NAHR pathways . This would be consistent with evidence that Rad51 binding is limited to about 5 kb on either side of a DSB [31] . We then repositioned the I-SceIcs at 151cs , within 0 . 1 kb of the RAHS recipient in the FRAHSΔ strain ( C in Figure 6 ) , in order to enhance Rad51 presynaptic filament assembly onto RAHS . Although a modest increase in Rad51-dependent rearrangements was observed , the majority of cells after the DSB at 151cs with FRAHSΔ cannot efficiently repair the chromosome in the absence of SSA ( 58±2% loss , Figure 6 ) . These data reveal that Rad51-dependent NAHR pathways induced by a DSB in unique sequences ( 163cs or 151cs ) are inherently inept at repairing the DSB using Ty1/Ty2 elements . Taken together , for a DSB in unique DNA , the efficiency of the SSA pathway coupled with the inefficiency of Rad51-dependent NAHR pathways generates the intrachromosomal position bias and preferential usage of Ty recipients and donors . Our findings show that the I-SceI-induced DSB in unique DNA ( 147cs , 151cs , or 163cs ) generates substantial NAHR between Ty repeats , giving rise to a broad spectrum of rearrangements through BDR in the purebred diploids . This is in contrast to current models that propose that break-proximal sequences determine the outcome , where DSBs in unique DNA lead to AHR ( between sisters or homologs ) and DSBs in repetitive DNA can lead to NAHR [32] . To assess the relative consequence of DSBs in unique versus repetitive DNA , we repositioned the I-SceIcs into the RAHS locus ( called RAHScs , Figure 7 ) and used our nonselective assay to measure all possible outcomes after the DSB at RAHScs in hybrid and purebred diploids . From the repair clones generated in our assay , we further characterized two Ty-mediated products that exclusively arise with the DSB at RAHScs , intra-Ty deletions and Ty GC . These Leu+HygSUra+ repair clones are distinguished from each other by assaying RAHS locus size using PFGE/Southern analysis ( Figure S7 ) . In comparison to the wild-type RAHS size , we observe a smaller RAHS size for intra-Ty deletion events and a similar RAHS size ( with only the removal of the small nonhomologous 1 . 6 kb I-Scecs/HYG ends ) for Ty GC events . Similar to results with the DSB at 163cs , SSA dominates the NAHR pathway competition , with 66% and 61% of cells after DSB at RAHScs generating Ty-mediated deletions in hybrid and purebred diploids , respectively ( Table 2 ) . SSA again imposes a strong intrachromosomal position bias , dictating recipient and donor preferences . The internal deletions from RAHScs , however , can be generated between the RAHS recipient and two different Ty donors , sequences within RAHS itself ( referred to as intra-Ty ) and FRAHS ( now referred to as inter-Ty ) . All of the internal deletions in purebred diploids are intra-Ty events ( 61±9% ) whereas in hybrid diploids , 59±9% are intra-Ty and 7±5% are inter-Ty ( Figure 7 and Table 2 ) . This is consistent with previous work describing a proximity effect during SSA using model repeat donors , with break-proximal donors preferred over break-distal donors [7] . In addition to the events observed with a DSB at 163cs , we find that the second most frequent event after DSB at RAHScs is Ty GC . 22±8% and 33±10% of cells after DSB at RAHScs lead to Ty GC events in hybrid and purebred diploids , respectively ( Figure 7 ) . The lower frequency of Ty GC relative to intra-Ty deletions measured in our diploids are in agreement with those events measured using an HO-induced DSB inside Ty1 in S . cerevisiae haploids [33] . Ty GC occurs through the coordination of a two-ended strand invasion event into a Ty donor , which is not a possibility when the DSB initiates in unique DNA ( as for 163cs ) . These GC events in the hybrid diploids must be mediated by a non-allelic Ty donor from the S . cerevisiae genome ( since S . bayanus lacks Ty1/Ty2 ) , which likely occurs in purebred diploids as well [16] . Thus , paradoxically , NAHR efficiently mediates conservative repair when a DSB occurs in repetitive DNA . Having completed our analyses of a DSB within a Ty1 repeat , we can now compare its impact to a DSB in unique DNA on genome integrity . We categorized the outcomes of the I-SceI-induced DSB at RAHScs and at 163cs into two groups: ( 1 ) change in gene copy number ( inter-Ty deletion , isochromosome , ring , translocation , and chromosome loss ) and ( 2 ) no change in gene copy number ( intra-Ty deletion , Ty GC , and allelic ) . This comparison reveals that the DSB in unique DNA is 3 to 5-fold more likely to cause a change in gene copy number than the DSB in repetitive DNA ( increases from 19% to 97% in hybrid diploids and 6% to 19% in purebred diploids , Figure 8 ) . Thus , distinct from models that highlight the role of DSBs inside repeats in mediating genome rearrangements , our results suggest that the relative mutagenic potential of a DSB in the genome actually decreases when the break occurs within repetitive DNA . Furthermore , this finding suggests that DSBs in unique DNA are more likely to lead to mutagenic rearrangements than DSBs in repetitive DNA .
We report a novel genome-wide system in budding yeast to study non-allelic homologous recombination ( NAHR ) between natural repeats . While previous assays isolate aspects of competitive repair addressed here , our system gauges the competition between all parameters concurrently , as what naturally transpires in a cell . The value of this new approach is evidenced by the surprising features of NAHR our system reveals . Remarkably , in purebred diploids , DSBs within a long stretch of unique sequences are not always repaired by allelic homologous recombination ( AHR ) as previously assumed . Rather , 17% of these DSBs repair by NAHR . This NAHR arises because the DSB activates Ty recipients 12 to 48 kb distal from the break site to recombine with non-allelic Ty donor sequences . Robust NAHR through break-distal recombination ( BDR ) is supported by a previous study of bridge-breakage-fusion in diploid budding yeast by Malkova and colleagues [34] . In this and the previous study , competition between BDR-dependent NAHR and AHR occurs after an endonuclease-induced DSB . In diploids , endonucleases can cleave one homolog prior to DNA replication and both its sister chromatids after DNA replication , thereby eliminating the sister chromatid as a donor for AHR . Therefore , the only AHR donor is the uncut homolog . However , a homolog is also the only AHR donor for repair of spontaneous DSBs that occur on unreplicated DNA in G1 or S . Indeed , recent evidence suggests that spontaneous DSBs occur on unreplicated DNA [35] . We suggest that spontaneous DSBs in unique unreplicated DNA are also likely to induce robust BDR-dependent NAHR . The fact that break-distal Ty sequences undergoes frequent NAHR reveals two surprising features of recombination that have important mechanistic implications for current models of recipient activation and choice . The first surprise is that distal Ty repeats are activated as recipients at all ( presumably by becoming single-stranded ) when break-proximal ssDNA can undergo AHR . Indeed , a recent study in diploid yeast suggests that ssDNA is generated at least 10 kb from a DSB before its repair is complete [36] . To explain this extensive break-distal resection , we suggest that a step after resection must be slow , such as the homology search for donor sequences . A slow homology search would provide time for break-distal sequences to be resected and compete with previously resected break-proximal sequences . Such a slow homology search is consistent with studies suggesting the slow diffusion of chromosomal sequences [37] . The second surprise is the disproportionate use of very small break-distal Ty sequences as recipients for NAHR . They would represent only a very small proportion of the entire block of resected DNA , which can all act as a recipient for AHR . We suggest that the smaller Ty recipients encounter their potential Ty donors first because chromosome territories [38] generate a high local concentration of potential intrachromosomal Ty donors . In contrast , the larger allelic recipients must travel further to partner with allelic donors on the homolog . Consistent with this model , almost all NAHR rearrangements through break-distal Ty recipients result from pairing with intrachromosomal Ty donors . Along with recipient usage , our genome-wide system reveals the role sequence homology and genomic position play in NAHR donor choice . We find that the Ty donors chosen by a recipient are not among the most homologous in the genome by the criteria of either percent identity or longest block of uninterrupted identity . Rather the primary determinant of NAHR donor choice is local proximity . We observe a ∼50-fold preference for Ty repeat donors on the same chromosome over different chromosomes . This intrachromosomal NAHR preference is consistent with previous studies [16]–[19] , although the magnitude of this preference differs , possibly due to specific configurations of repeats relative to a break site , as observed in our studies . However , in contrast to previous work , our study shows this intrachromosomal bias occurs under conditions that allow unrestricted choice of repair pathways and partners amongst a natural repetitive family . Interestingly , Ty1/Ty2 elements are preferentially inserted within 750 bp upstream of tRNA genes [39] , and dispersed tRNA genes cluster together [40] . Our results suggest that possible Ty interchromosomal contacts mediated by tRNA clustering is not sufficient to overcome an intrachromosomal bias . It will be interesting to see whether higher-order chromosome organization may influence donor repair choice of natural repeats when only interchromosomal donors are available for NAHR . Our system also provides insights into the preferred repair pathways that act on a family of natural repeats . We show that NAHR occurs mostly by the SSA pathway whether DSBs occur in unique sequences or a Ty repeat . The robustness of SSA is consistent with previous studies using model repeats [18] , [23] , [30] , [41] , [42] . Since repair of a single DSB by SSA will occur through an intrachromosomal donor , the predominance of SSA helps explain the preferential usage of intrachromosomal donors and the resulting preference for intrachromosomal NAHR . Importantly , our pathway analysis of NAHR also helps explain one of the most surprising and striking observations of this study: DSBs that occur outside repeat clusters are more mutagenic than DSBs that occur inside repeat clusters . This seemingly counterintuitive observation arises because DSBs that occur inside a Ty have better options for repair , both in efficiency of pathways and favorably positioned donors . DSBs within the Ty predominately repair through two highly efficient pathways , SSA within the Ty locus or GC with preferred intrachromosomal Ty donors [16] . These types of repair preserve gene copy number since neighboring unique genes are unaffected . Since SSA and GC are compensatory pathways [22] , it is possible that DSBs inside repetitive elements that cannot undergo SSA ( i . e . solo insertion of LINE-1 ) efficiently repair through GC events [43] . A recombination execution checkpoint has been suggested to maintain genome integrity by ensuring the coordination of two-ended strand invasion events during GC for conservative repair [28] . Consistent with this , our results suggest that NAHR through GC between natural repeats is a major mechanism that limits changes in genome structure . In contrast , DSBs in unique sequences that repair predominately through GC with the homolog is not as effective in limiting detrimental rearrangements . As the search for the interchromosomal homolog allows for more time to activate a break-distal Ty as a recipient , BDR occurs more frequently through SSA between distinct Ty loci or one-ended events through the BIR pathway . In this situation , SSA always , and BIR often times , change the copy number of neighboring unique genes . Hence , this opens up the possibility that DSBs in unique sequences , rather than repeats , may generate spontaneous or irradiation-induced NAHR-dependent rearrangements observed in yeast [32] , [44] . Similarly , NAHR-dependent rearrangements in the human genome may also occur by a DSB in the surrounding unique DNA followed by BDR-dependent NAHR . If so , then the recombinant junction would not coincide with the site of the initiating lesion . Therefore , analysis of NAHR junctions alone may miss underlying mechanisms for genome rearrangements . Examining broad regions around NAHR junctions could potentially identify fragile sites that predispose a locus to recurrent instability , contributing to genetic diversity and disease .
Standard yeast genetic and molecular biology methods were used [45] . All S . cerevisiae strains were derived from BY4700 ( MATa ura3Δ0 ) , BY4716 ( MATα lys2Δ0 ) , or BY4704 ( MATa ade2Δ::hisG his3Δ200 leu2Δ0 lys2Δ0 met15Δ0 trp1Δ63 ) [46] . All S . bayanus strains were derived from a S . bayanus prototroph received as a gift from Ed Louis . Deletion of the HO gene and auxotrophic markers were introduced by transformation to generate a number of haploid S . bayanus strains for laboratory use , including MH3399 ( MATa hoΔ::hisG ura3Δ::NAT leu2Δ::NAT ade2Δ::hisG ) , YZB9-4B1 ( MATa hoΔ::KAN ura3Δ::NAT leu2Δ::NAT ) , YZB5-102 ( MATα hoΔ::KAN lys2-1 ) ( this study , [47] ) . Since S . bayanus is sensitive to high temperatures , the following modifications were made to the high efficiency yeast transformation protocol [48] for S . bayanus and hybrid diploids strains: room temperature incubation of transformation mix for 30 minutes , 5 minute heat shock at 42°C , and 5 minute rest at room temperature following heat shock . Except for some noted below , insertion/knockout constructs were generated through one-step transformation of a PCR amplified linear construct . Each primer for these constructs included ∼50 bp of homology to target for genomic integration and ∼20 bp that anneal to a plasmid template for the amplification of a selectable marker [pAG32-hphMX4 ( Hygromycin B ) , pAG25-ClonatMX4 ( Clonat ) , pFA6a-kanMX4 ( Kanamycin ) , or pMPY-ZAP ( hisG-URA3-hisG pop-in/pop-out construct ) ] . One primer of each of the I-SceI cut site primer pairs also included the 30 bp I-SceI recognition sequence from [49] . For RAHScs , the primers included linkers to amplify an AgeI-I-SceIcs/HYG-ClaI fragment , which was digested and ligated into AgeI-ClaI site of pFT1 ( derived from p150Ty , this study ) . The resulting plasmid , called pFT1-SceIcs , was double-digested with NotI and KpnI and a 10 . 2 kb purified NotI-KpnI fragment was used for transformation to create RAHScs . For FRAHSΔ::hisG , three primer pairs ( FRAHSΔ-left , FRAHSΔ-middle , FRAHSΔ-right ) were used to generate three overlapping fragments that were co-transformed . Sequences for gene knockout primers are available upon request . All other strain construction primers included in Table S2 . All genome manipulations were performed in haploid strains , and all constructs were verified by Southern blot analysis . Pairs of S . cerevisiae and S . bayanus haploids were mated to generate the desired purebred and hybrid diploids , and then transformed with the I-SceI expression plasmid ( see below ) . All experiments in this study were performed at 23°C unless noted otherwise . Yeast strains were grown in YEP , SC-ADE , SC-ADE-URA media supplemented with 2% dextrose ( D ) , 2% lactic acid 3% glycerol ( LAG ) , 0 . 3 mg/ml Hygromycin B ( HYG ) , as indicated . YEPD media was supplemented with 10 µg/ml adenine . Glucose and glycerol was purchased from EMD Biosciences , lactic acid ( 40% v/v stock , [pH 5 . 7] ) from Fisher Scientific , and Hygromycin B ( HYG ) from Roche . SC dropout powders were homemade from amino acids purchased from Sigma-Aldrich . The GALp-I-SceI construct was from pWJ1320 [49] , a gift from Rodney Rothstein . pMH5 was derived from pWJ1320 ( 2 micron-based ) by deleting a 2 . 0 kb EcoO109I fragment containing URA3 marker . pMH6 ( 2 micron-based ) and pMH7 ( CEN-based ) were created by ligating the 2 . 0 kb SalI fragment from pWJ1320 ( containing the GALp-I-SceI expression construct ) into the unique SalI site of pRS422 and pRS412 , respectively . pMH6 and pMH7 were generated to include a larger promoter sequence for the ADE2 marker , however , all plasmids yielded similar results . A single colony from SC-ADE-URA+D+HYG plates [to select for GALp-I-SceI expression plasmid ( Ade+ ) and no DSB ( HygRUra+ ) ] was used to inoculate SC-ADE-URA+D for a 5 ml starter culture that was grown to saturation . A small volume of the starter was used to inoculate SC-ADE+LAG cultures and these cultures were grown for more than two doubling to exponential phase [OD ( 600 ) ∼1 . 0] . For the uninduced control , immediately before DSB induction , an aliquot was appropriately diluted in water and plated onto YEPD for individual colonies ( uninduced frequencies are subtracted out of induced frequencies , see below ) . To induce the DSB , galactose ( 20% v/v stock ) was added to a final of 2% and after two hours , the cultures were diluted in water and plated onto YEPD for individual colonies ( referred to as clones ) . Plates were incubated at 23°C for 3–5 days . YEPD platings from uninduced and induced were first replica plated onto YEPD or 2% agar plates . This replica plate was then immediately used on a fresh velvet to replica onto YEPD+HYG , SC-URA+D , and SC-LEU+D plates . These marker plates were incubated at 23°C for 2–4 days . Each colony from the original YEPD plate was scored for the presence or absence of chromosome III markers ( LEU2 , HYG , URA3 ) by growth or no growth on marker plates . Assessment of the heterozygous markers ( present on the S . cerevisiae homolog with the I-SceIcs ) determines whether the founding cell had experienced an I-SceI-induced DSB ( leading to the HygS phenotype ) followed by chromosome repair [HygS and Leu+Ura+ ( class I ) or HygS and Leu+Ura− ( class II ) ] or chromosome loss [HygS and Leu−Ura− ( class III ) ] . The HygS phenotype most likely occurs through the removal of the nonhomologous ends ( 1 . 6 kb I-SceIcs/HYG construct ) , which is a natural and efficient step during HR repair [50] , [51] . The following three steps were used to calculate frequencies of repair and loss events . First , the numbers of clones that fell into each genetic class ( I , II , III ) out of the total number of clones scored were calculated as percentages for both uninduced and induced cultures . Second , uninduced percentages were subtracted from induced percentages to eliminate events that occurred before galactose addition . Occasionally , cultures with high background frequencies ( >50% of clones were HygS in uninduced cultures ) were observed and not used . HygS phenotypes before galactose induction are due to leakiness of the galactose promoter during nonrepressive growth ( see Figure S3 ) . Third , the total percentage ( class I + class II + class III ) was normalized to 100% . A third potential repair class , HygS and Leu−Ura+ , arose so infrequently ( <1% in wild-type purebred and hybrid diploids ) that it was omitted from these calculations . Single repair clones ( class I and II ) from SC-LEU+D marker plates were restruck for individual isolates onto fresh SC-LEU+D plates to ensure clonality ( i . e . possible mixing during replica plating process ) . One isolate from this restreak was used to inoculated YEPD media and grown to saturation for the subsequent isolation of genomic DNA for PFGE/Southern analysis using a LEU2 probe ( see below ) . Hybridization that resulted in wild-type chromosome III size ( purebred diploids at 341 kb , hybrid diploids at 320 kb ) was identified as AHR and those with an altered chromosome III size , indicative of a rearrangement , were classified as potential NAHR . The structures of the chromosome III rearrangement structures were first determined in wild-type hybrid diploids ( MH3360 ) due to the advantage of no signal from an uncut homolog . Frequencies were calculated in three steps . 1 ) Frequencies of genetic classes ( I , II , III ) of uninduced cultures were subtracted from frequencies of induced cultures to eliminate events that occurred prior to galactose addition ( described in more detail above , frequency of chromosome loss determined here ) . 2 ) For the repair events , the fraction of each type of repair ( i . e . allelic , internal deletion , etc ) among the total PFG plugs analyzed from its corresponding genetic class ( I or II ) was calculated . 3 ) For the repair events , the genetic class frequency ( step one ) was multiplied by the fraction of each repair type in that genetic class ( step two ) . For example , in wild-type purebred diploids ( MH3359 ) , 85 . 7% of HygS clones ( n = 1062 ) were class I ( Leu+HygSUra+ ) . 5 out of 32 random repair clones of class I were classified as internal deletions by PFGE/Southern analysis , so the frequency of internal deletions in MH3359 is 5/32 ( 85 . 7% ) = 13 . 4% . Yeast genomic DNA was prepared in 1% low-melting agarose plugs ( SeaPlaque 50100 ) as previously described [52] and resolved on 1% agarose gel ( Bio-Rad 162-0138 ) in 0 . 5XTBE using Bio-Rad CHEF-DR III System . To optimize resolution between S . cerevisiae and S . bayanus chromosome III the following parameters were used: 6 V/cm , 120° angle , 1–25 s switch times , 24 hours at 14°C . To assess yeast whole genome karyotypes ( i . e . for translocations ) , the parameters were the same except for 60–120 s switch times . Gels were blotted using GeneScreen Plus membrane ( Perkin Elmer NEF988 ) and probed with a 1 . 3 kb fragment from the S . cerevisiae LEU2 locus amplified using the U2-FOR/U2-REV primer pair ( Table S2 ) . To calculate SEMs for the repair outcomes , the following numbers were used: ( a ) average frequency of Leu+HygSUra+ genetic class I , ( b ) average frequency of Leu+HygSUra− genetic class II , ( c ) total number of Leu+HygSUra+ ( class I ) plugs analyzed by PFGE/Southern analysis , ( d ) total number of Leu+HygSUra− ( class II ) plugs analyzed using PFGE/Southern analysis , ( e ) number of Leu+HygSUra+ ( class I ) plugs of a particular repair outcome ( i . e . allelic , internal deletion ) , ( f ) number of Leu+HygSUra− ( class II ) plugs of a particular outcome ( i . e . ring , translocation , isochromosome ) . SEM was calculated in two steps . First , the initial SEM was calculated using the formula SQRT ( pq/n ) , where p = fraction of a particular repair outcome observed by PFGE/Southern analysis over total analyzed from that class ( e or f divided by c or d , respectively ) , q = 1-p , and n = total number of repair clones analyzed by PFGE/Southern analysis from that corresponding class ( c or d ) . Second , the final SEM was calculated by weighting the SEM with the corresponding genetic class frequency ( initial SEM multiplied by a or b ) . The rationale for this method was to be most stringent by using the smallest n ( d or e ) . In the following cases e or f was assigned the number 1: ( 1 ) when all Leu+HygSUra+ plugs were deletions ( i . e . in hybrid diploids ) , ( 2 ) no products appear in any plugs analyzed ( i . e . rings in rad51Δ/rad51Δ mutant ) , ( 3 ) genetic class is 0 ( i . e . Leu+HygSUra− class II in rad52Δ/rad52Δ hybrid diploids ) , ( 4 ) when no plugs analyzed ( i . e . Leu+HygSUra− class II in rad52Δ/rad52Δ purebred diploids ) . For case 1 , the error was estimated by assuming the next plug would not be that particular outcome . For case 2 , 3 , and 4 , the upper bound was estimated by assuming the next plug would be that particular outcome . In the case where repair outcomes came from both the Leu+HygSUra+ and Leu+HygSUra− genetic classes ( i . e . other , allelic in purebred diploids ) , “final SEMs” were calculated as described above and then “final SEMs” from each class was added together for the reported SEM . To calculate SEMs for chromosome loss , the formula SD/SQRT ( n ) was used where SD ( standard deviation ) = SD of the frequency of Leu−HygSUra− clones from different isolates and/or DSB-inductions ( same experiment used to generate numbers for a and b above ) and n = total number of different DSB-inductions performed for that particular strain ( ranging between 2 to 8 ) . Exponential cultures in –ade +2% lactic acid +3% glycerol were appropriately diluted in water and the same volume was plated on –ade +2% galactose and –ade +2% glucose . Plates were incubated at 23°C . Percent viability was calculated as the number of colony forming units on galactose divided by the number of colony forming units on glucose . aCGH methods were performed as previously described [53] . S . cerevisiae/S . bayanus hybrid microarrays were custom designed and printed by Lewis-Sigler Institute Microarray Facility at Princeton University . Numerous studies have brought to light unannotated Ty elements on chromosome III [34] , [44] , [54]–[56] , with a few studies publishing a limited restriction digest map of the Ty structure in these regions [44] , [54] , [55] . These unannotated Ty clusters were sequenced here . Each cluster was cloned from strain MH3303 ( MATa lys2Δ0 ura3Δ0 , derived from BY4716 [46] ) by gap repair to create p85Ty , p150Ty , and p169Ty ( see Figure S1 ) . Each plasmid was subjected to transposon bombing using the Finnzymes Template Generation System ( TGS ) . For each plasmid , 192 clones with different random transposon insertions were picked and sequenced with a pair of primers located at the edges of the TGS transposon to produce pairs of oppositely directed reads . 384 attempted reads were performed per yeast clone . Sequence data were processed , assembled and edited using the Phred/Phrap/Consed suite of programs [57] . Each assembly was reviewed and edited to ensure there were no discrepancies due to misplaced reads or low quality regions . The automated assembler resulted in collapses of repeats , and these were manually resolved . 16 . 8 kb of sequence at LAHS , 14 . 5 kb at RAHS , and 14 . 7 kb at FRAHS were deposited into GenBank with accession number GU224294 , GU220389 , and GU220390 , respectively . The sequence included five additional full length Ty1s and a solo LTR , complementing the LAHS reference sequence in SGD and almost entirely replacing the RAHS and FRAHS reference sequence . The new sequence changes chromosome III size from 316 , 617 bp ( in SGD ) to 341 , 823 bp . Sequences for all previously described Ty1 , Ty2 and LTRs ( delta ) elements were obtained from the SGD “Non-ORF dataset” ( http://downloads . yeastgenome . org/ , timestamp January 5 , 2010 ) . Several corrections were made based on our resequencing and analysis: ( 1 ) addition of five Ty1 elements on chrIII ( Ty1–1 through Ty1–5 ) ( 2 ) addition of nine delta elements on chrIII ( delta16 through delta24 ) ( 3 ) removal of three delta elements on chrIII ( YCRWdelta8 , YCRWdelta9 , and YCRWdelta10 ) ( 4 ) addition of one unannotated Ty1 element on chrXII ( encompassing YLR035C-A ) ( 5 ) addition of two unannotated delta elements on chrIV ( LTRs for YDRCTy1-2 ) . The “Overall Identity ( % ) ” between two sequences was determined by creating a global sequence alignment using the Needleman-Wunsch algorithm ( gapopen = 10 , gapextend = 0 . 5 ) as implemented in needleall v6 . 2 . 0 [58] . The “Longest Block of 100% Identity ( nt ) ” was determined by first creating a local sequence alignment using the NCBI BLAST algorithm ( match = 1 , mismatch = −3 , gapopen = −1 , gapextend = −1 ) as implemented in bl2seq v2 . 2 . 18 [59] . Custom Perl scripts using BioPerl v1 . 6 . 1 iterated through each set of hits to identify the longest contiguous block of matching nucleotides [60] . Finally , the contribution of sequence similarity to donor usage is likely more complex than either overall identity or longest block of perfect identity . We therefore calculated bit scores using the BLAST heuristic , which attempts to balance length and perfect identity when searching for a shared region between two sequences that has the “most” similarity . This “Local Identity ( bitscore ) ” was determined using blastall . Source code and data files can be found at: http://dl . getdropbox . com/u/547386/code . zip | The human genome is structurally dynamic , frequently undergoing loss , duplication , and rearrangement of large chromosome segments . These structural changes occur both in normal and in cancerous cells and are thought to cause both benign and deleterious changes in cell function . Many of these structural alterations are generated when two dispersed repeated DNA sequences at non-allelic sites recombine during non-allelic homologous recombination ( NAHR ) . Here we study NAHR on a genome-wide scale using the experimentally tractable budding yeast as a eukaryotic model genome with its fully sequenced family of repeated DNA elements , the Ty retrotransposons . With our novel system , we simultaneously measure the effects of known recombination parameters on the frequency of NAHR to understand which parameters most influence the occurrence of rearrangements between repetitive sequences . These findings provide a basic framework for interpreting how structural changes observed in the human genome may have arisen . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/recombination",
"molecular",
"biology/chromosome",
"structure",
"molecular",
"biology/dna",
"repair"
] | 2010 | Competitive Repair by Naturally Dispersed Repetitive DNA during Non-Allelic Homologous Recombination |
Synaptic communication is a dynamic process that is key to the regulation of neuronal excitability and information processing in the brain . To date , however , the molecular signals controlling synaptic dynamics have been poorly understood . Membrane-derived bioactive phospholipids are potential candidates to control short-term tuning of synaptic signaling , a plastic event essential for information processing at both the cellular and neuronal network levels in the brain . Here , we showed that phospholipids affect excitatory and inhibitory neurotransmission by different degrees , loci , and mechanisms of action . Signaling triggered by lysophosphatidic acid ( LPA ) evoked rapid and reversible depression of excitatory and inhibitory postsynaptic currents . At excitatory synapses , LPA-induced depression depended on LPA1/Gαi/o-protein/phospholipase C/myosin light chain kinase cascade at the presynaptic site . LPA increased myosin light chain phosphorylation , which is known to trigger actomyosin contraction , and reduced the number of synaptic vesicles docked to active zones in excitatory boutons . At inhibitory synapses , postsynaptic LPA signaling led to dephosphorylation , and internalization of the GABAAγ2 subunit through the LPA1/Gα12/13-protein/RhoA/Rho kinase/calcineurin pathway . However , LPA-induced depression of GABAergic transmission was correlated with an endocytosis-independent reduction of GABAA receptors , possibly by GABAAγ2 dephosphorylation and subsequent increased lateral diffusion . Furthermore , endogenous LPA signaling , mainly via LPA1 , mediated activity-dependent inhibitory depression in a model of experimental synaptic plasticity . Finally , LPA signaling , most likely restraining the excitatory drive incoming to motoneurons , regulated performance of motor output commands , a basic brain processing task . We propose that lysophospholipids serve as potential local messengers that tune synaptic strength to precedent activity of the neuron .
Activity-dependent plasticity of neuronal networks refers to the adaptive changes in their properties in response to external and internal stimuli . In a prominent form of central nervous system ( CNS ) plasticity , synaptic strength results in an increase ( potentiation ) or decrease ( depression ) of transmission efficacy , depending on the neuron’s precedent activity ( activity-dependent synaptic plasticity ) . Short-lived processes that modify synaptic strength occur in practically all types of synapses [1] , and short-term synaptic plasticity is essential in regulating neuronal excitability and is central to information processing at both cellular and neuronal network levels [2] . Homeostatic adjustment of synaptic weights counteracts neuronal rate disturbances that affect self-tuning neuronal activity within a dynamic range via Ca2+-dependent sensors [3] . The number of receptors in the surface membrane and at synaptic sites , and the size of the readily releasable pool ( RRP ) of synaptic vesicles ( SVs ) , are important determinants of synaptic strength , short-term plasticity , and intersynaptic crosstalk [4–8] . Unmasking the feedback mechanisms that are believed to sense neuron activity and adjust synaptic strength ( i . e . , activity-dependent , coupled messenger synthesis and/or release ) would help to explain how circuits adapt during synaptic homeostasis , experience-dependent plasticity , and/or synaptic dysfunctions that underlie cognitive decline in many neurological diseases . The ligand-gated ionotropic channels—A-type GABAA receptors ( GABAARs ) and AMPA-type glutamate receptors ( AMPARs ) —mediate fast synaptic transmission at the vast majority of inhibitory and excitatory synapses , respectively , in the mammalian brain [4 , 5 , 9] . Cell surface stability of receptors is further regulated by post-translational phosphorylation , palmitoylation , and/or ubiquitination . In particular , AMPAR and GABAAR phosphorylation modulates the receptor’s biophysical properties and membrane trafficking . Hence , the coordinated activity of kinases and phosphatases plays a pivotal role in controlling synaptic strength and neuronal excitability . Key residues within the intracellular domains of diverse AMPAR and GABAAR subunits are targeted by a number of kinases , including protein kinases A and C , calcium/calmodulin-dependent kinase II , and tyrosine kinases of the Src family . Generally , phosphorylation stabilizes the receptor on the surface and , conversely , dephosphorylation appears to be important for receptor endocytosis [4 , 9] . Lysophosphatidic acid ( LPA ) is a strong candidate to function as a local messenger that rapidly affects synaptic strength . A membrane-derived bioactive phospholipid that affects all biological systems , LPA is an important intermediary in lipid metabolism and has a vital role in de novo biosynthesis of membrane phospholipids [10] . The nervous system is markedly modulated by LPA signaling . LPA , autotaxin ( the main LPA-synthesizing enzyme ) , and many subtypes of LPA-specific G-protein-coupled receptors ( LPA1–6 ) are enriched in the brain [10–12] . Downstream signaling cascades mediating LPA signaling include mitogen-activated protein kinase ( MAPK ) activation , adenylyl cyclase inhibition or activation , phospholipase C ( PLC ) activation/Ca2+ mobilization and/or protein kinase C ( PKC ) activation , arachidonic acid release , Akt/PKB activation , and the activation of small GTPase RhoA and subsequent Rho kinase ( ROCK ) stimulation [10] . Many subtypes of LPA receptors ( LPARs ) are expressed in the brain; in particular , LPA1 is highly expressed and is the most prevalent receptor subtype in both the embryonic and adult brains [13–15] . Accordingly , LPA targets all CNS cell types to modulate developmental processes including neurogenesis , migration , differentiation , and morphological and functional changes [10] . However , little is known about how LPA signaling influences neuron physiology and neuronal connectivity or integrates incoming synaptic drive . Presynaptic LPA2 at glutamatergic synapses mediates neuronal network hyperexcitability in an epileptic mouse model [16] . In addition , LPA1-deficient mice manifest alterations in managing diverse neurotransmitters [17–20] . Endogenous ROCK activity , an intracellular partner in LPA signaling , is necessary to maintain afferent AMPAergic and GABAAergic synaptic strength in motoneurons [8] . As a conventional link in synaptic plasticity , activity-dependent LPA production occurs downstream of noxious activation of glutamate receptors in models of neuropathic pain [21] . However , whether LPA signaling is actually able to modulate synaptic strength and mediate activity-dependent synaptic plasticity remains unresolved . The aim of this study was to investigate whether LPA regulates synaptic strength and plasticity of motoneuron excitatory and inhibitory synapses . Here , we show that LPA—mainly via LPA1—induced rapid and reversible depression in synaptic strength ( short-term depression [STD] ) , and operated as an autocrine messenger mediating activity-dependent STD at inhibitory synapses . At glutamatergic synapses , presynaptic LPA signaling reduced the size of the RRP of SVs . At GABAergic synapses , postsynaptic LPA action mediated dephosphorylation and endocytosis-dependent internalization of the GABAAγ2 subunit . Strikingly , LPA signaling regulated the performance of motor output commands in vivo . Therefore , LPA seems to have important implications for synaptic plasticity , pathology , and information processing in the brain .
To explore a possible role of LPA in shaping the normal motor output of the HN , it was necessary to determine the predominant isotype of its main target receptors expressed in this motor nucleus . Assessment of the expression levels of mRNAs for LPA1–6 receptors in microdissected HN from neonatal ( P7 ) rats revealed that lpa1 mRNA was 1 . 5 to 12 . 5 times more abundant than lpa2–6 transcripts ( Fig 1A ) . Subsequently , confocal analysis of double immunolabeled HN from P7 pups showed LPA1-immunoreactive ( ir ) puncta , patches , and fiber-like structures colocalizing with SMI32-positive HMN perikarya and dendrite-like structures ( Fig 1B–1D ) . Three-dimensional reconstructions agreed with a cytoplasmic and membrane localization of LPA1 in perikarya and main dendritic branches of HMNs ( Fig 1E and 1F ) . Triple immunofluorescence for LPA1 , SMI32 , and the vesicular glutamate ( VGLUT2 ) or GABA ( VGAT ) transporters as synaptic markers confirmed that LPA1-ir puncta were colocalizing with excitatory ( VGLUT2-ir ) or inhibitory ( VGAT-ir ) presynaptic structures ( Fig 1G , 1H , 1J , and 1K ) . Both excitatory and inhibitory inputs were also found apposed to SMI32-ir neuropil or somata coexpressing LPA1 ( Fig 1H and 1I ) . Although LPA1 expression in other neural cell types is not excluded , this expression pattern supports pre and/or postsynaptic roles of LPA1 at the main excitatory and inhibitory inputs on HMNs , suggesting a potential contribution of LPA to motoneuron physiology . Next , we investigated the functional effects of LPA on glutamatergic and GABAergic synaptic currents by whole-cell patch-clamp recordings of HMNs ( slices from P6–P9 rats ) . Electrical stimulation of the ventrolateral reticular formation ( VLRF ) evoked postsynaptic currents ( ePSCs ) in HMNs ( Fig 2A ) . The AMPAR- or GABAAR-mediated components of ePSCs ( excitatory [eEPSCsAMPA] or inhibitory postsynaptic currents [eIPSCsGABAA] , respectively ) were isolated and recorded as described in S1 Text . The two major species of LPA ( approximately 70% ) found in the brain [26] , monounsaturated ( 18:1 , or LPA ) and saturated ( 18:0 , or s-LPA ) , were used in this study . While LPA activates LPA1–3 , s-LPA has high affinity for LPA1/2 , but is a comparatively poor agonist against LPA3 [27] . Unless stated otherwise , LPA was used at a similar concentration ( 2 . 5 μM ) to that found in serum ( 1–5 μM ) [28] . In general , unsaturated LPAs are more potent than s-LPA in activating LPARs and inducing biological activities [29] . Accordingly , a higher concentration was used for s-LPA ( 40 μM ) than for LPA ( 2 . 5 μM ) to achieve a similar effect on neurotransmission . Both phospholipids , added for 10 min to the bath solution , strongly attenuated the amplitude of eEPSCsAMPA and eIPSCsGABAA ( Fig 2B ) . The effects were reversed after 10 min of washing . Thus , LPA modulated rapidly and reversibly the strength of AMPAR- and GABAAR-mediated synaptic transmission in motoneurons . The tested dose ( 2 . 5 μM ) of LPA had a proportionately higher effect on inhibitory than on excitatory inputs ( Fig 2B and 2C ) . Further , differential sensitivity to LPA was studied by applying various concentrations , ranging from 1 nM to 20 μM . After subtracting vehicle-induced changes ( S1 Text ) , an effect on both currents was detectable at concentrations as low as 10 nM and increased with LPA concentration to a similar maximum reduction in both currents ( approximately 70% ) at 10–20 μM ( Fig 2C ) . Dose-response relationships were well fitted ( p < 0 . 001; r2 > 0 . 99 ) by biphasic ( two slopes ) five-parameter logistic equations , suggesting that LPA affects synaptic neurotransmission by multiple mechanisms . It remains to be determined whether this is the consequence of the recruitment of diverse isoreceptors and/or downstream signaling pathways . In any case , from the nanomolar to first-order micromolar range , LPA diminished inhibitory inputs ( IC50 = 1 . 0 ± 0 . 17 μM ) in greater proportions ( p < 0 . 001 , Kolmogorov-Smirnov test ) than excitatory ones ( IC50 = 1 . 8 ± 0 . 08 μM ) , but at higher concentrations , LPA affected both synaptic systems similarly ( Fig 2C ) . As in our previously published study [8] , a combined electrophysiological analysis was performed to identify the LPA synaptic site of action . LPA signaling on AMPAR-mediated transmission is likely not attributable to changes in postsynaptic sensitivity to glutamate . LPA did not alter the amplitude in both the miniature quantal EPSCsAMPA ( mEPSCsAMPA ) and postsynaptic currents evoked by exogenous glutamate pulses ( S1 Text; S1 Fig ) . For that reason , we sought evidence for a presynaptic mechanism by recording spontaneous AMPAergic synaptic currents under facilitated spontaneous glutamate release ( sEPSCsAMPA ) . In this condition , synaptic activity was a mixture of action potential-dependent and -independent events . After LPA treatment , the sEPSCsAMPA amplitude , but not frequency ( 10 . 8 ± 1 . 0 Hz , p = 0 . 761 ) , reversibly decreased to a value similar to that recorded for mEPSCsAMPA in control condition ( before: 36 . 0 ± 3 . 8 pA; LPA: 24 . 0 ± 2 . 0 pA; Fig 3A–3C ) . This agrees with a LPA-induced full inhibition of action potential-dependent events . In addition , we evaluated eEPSCsAMPA facilitation using paired-pulse and repetitive afferent stimulation protocols as in our previously published study [8] . Under repetitive stimulation , a change in the amount of facilitation is considered to be attributable to a presynaptic change in the release probability of neurotransmitter quanta [1] . In the control condition , paired-pulse stimulation displayed a strong facilitation of eEPSCsAMPA over the entire range of interstimulus intervals tested , but this was more pronounced at shorter interstimulus intervals ( Fig 3D; S2 Fig ) . Facilitated PPR ( paired-pulse ratio ) showed a marked and reversible increase at 25 ms and 50 ms intervals after application of either s- or LPA ( abbreviated as s-/LPA; Fig 3D; S2 Fig ) . On average , LPA and s-LPA increased the magnitude of PPR by 12 . 8% and 29 . 3% at 25 ms , respectively . The finding that LPA also reversibly potentiated the facilitation index of eEPSCsAMPA under repeated VLRF stimulation provided additional evidence in support of these outcomes ( S1 Text; S3 Fig ) . At this point in our study , the attenuation of eEPSCsAMPA induced by LPA was related to a reduction in the glutamate release probability , which is believed to be determined by the number of fusion-competent SVs or the size of the RRP of SVs [6 , 7] . This idea was further strengthened by a subsequent analysis of eEPSCsAMPA amplitude using the minimal stimulation paradigm , designed to stimulate only one fiber and a single or small number of release sites . As in our previous study [8] , the intensity of the stimulation was set to elicit eEPSCsAMPA with 30% to 40% failure ( Fig 3E ) . In this context , LPA treatment evoked a significant reduction of the mean amplitude of eEPSCsAMPA elicited by minimal stimulation and an enhancement of the eEPSCsAMPA failure rate ( Fig 3E; S4 Fig ) . The presynaptic action of LPA on glutamatergic inputs is further supported because LPA1-ir puncta colocalize with Munc13-1 , a presynaptic active zone ( a . z . ) marker [30] , in VGLUT2-containing boutons ( Fig 3F and 3G ) . The LPA1 association with a region of the presynaptic membrane compromised in the fusion of SVs supports that LPA signaling has a direct relationship with the machinery involved in the regulation of neurotransmitter release . The qRT-PCR and immunohistochemical studies , together with additional pharmacological tests ( S1 Text: S5 Fig; S6 Fig ) , robustly point to LPA1 as a pivotal LPAR affecting glutamatergic synapses . In this context , injection of a small interfering RNA ( siRNA ) against lpa1 ( siRNAlpa1; 2 μg/2 μl ) into the fourth ventricle efficiently reduced LPA1 expression in the brain stem ( Fig 4A; S1 Text; S7 Fig ) . siRNAlpa1 robustly diminished , but did not fully avoid , ( s- ) LPA-induced alterations on eEPSCsAMPA amplitude and PPR relative to the administration of control noninterfering siRNA ( cRNA; 2 μg/2 μl ) or vehicle ( RNase-free phosphate buffered saline; 2 μl ) ( Fig 4A–4D ) . Whether the remaining response of eEPSCsAMPA to ( s- ) LPA could be due to residual LPA1 expression or to recruitment of compensatory mechanisms—e . g . , via up-regulated LPA3 in response to LPA1 knockdown—remains to be elucidated . LPA1 couples with and activates three G proteins: Gα12/13 , Gαi/o , and Gαq/11 [10] . Previous findings [8] and pharmacological data ( S1 Text ) did not support Gα12/13 involvement . Alternatively , preincubation for 2 h with the Gαi/o specific inhibitor pertussis toxin ( PTX ) , but not with the noncatalytic B oligomer of PTX ( bPTX ) , prevented ( s- ) LPA-induced STD and PPR increase ( Fig 4E , 4G , and 4H; S8A , S8D , and S8E Fig ) . Cascade downstream of lysophospholipids included PLC activation; the PLC inhibitor U73122 , but not its inactive analog U73343 , reversed—to a control-like state—the changes in amplitude and PPR provoked by ( s- ) LPA ( Fig 4F–4H; S8B , S8D , and S8E Fig ) . Finally , the Gαq/11 inhibitor YM-254890 did not interfere with s-LPA effects on eEPSCsAMPA ( S8C–S8E Fig ) . Altogether , these findings indicate that LPA signaling controls excitatory inputs via presynaptic Gαi/o-protein-coupled LPA1 and PLC ( Fig 4I ) . LPA induces smooth muscle contraction in a PLC-dependent , ROCK-independent manner that involves myosin light chain ( MLC ) phosphorylation by MLC kinase ( MLCK ) [31] . These findings point to MLCK as a potential kinase mediating the presynaptic action of LPA on excitatory neurotransmission . Accordingly , LPA increased the p-MLC:MLC ratio in the HN relative to aCSF-incubated brain stem slices , which was fully prevented by coincubation with the specific MLCK inhibitor ML-7 ( Fig 5A and 5B ) . In concordance , though ML-7 per se did not alter the amplitude of eEPSCsAMPA , as we also recently reported [8] , it fully suppressed LPA-induced alterations on eEPSCsAMPA amplitude and PPR ( Fig 5C–5F ) . This further supports MLCK as a main molecular substrate activated by LPA signaling within excitatory presynaptic terminals . MLC phosphorylation stimulates actomyosin interactions [32] , and presynaptic Ca2+ concentration regulates MLCK activity and modulates the RRP size in the calyx of the Held synapse [33] . Therefore , LPA signaling , through its modulatory control on MLCK and the actomyosin cytoskeleton , might regulate clustering and spatial distribution of SVs within excitatory ( S-type , spherical SVs-containing ) boutons ( S1 Text ) . Electron microscopy analysis , performed as in our previous study [8] , showed that , in a MLCK-dependent way , LPA noticeably reduced the number of SVs near the a . z . in S-type boutons attached to HMNs , compared to control conditions ( Fig 5G–5L; S1 Text ) . In addition , LPA induced a drop ( −20 . 2 ± 6 . 3% ) in the SV population morphologically docked to ( i . e . , in contact with ) the a . z . , which corresponds to the release-ready neurotransmitter quanta [34] that was prevented by coaddition of ML-7 ( Fig 5M and 5N ) . These outcomes robustly support that LPA signaling regulates the size of the RRP of SVs in S-type boutons by a MLCK-dependent mechanism . Together , these data strongly suggest that the depression of synaptic strength induced by LPA treatment is dependent on a reduction in the probability of release from excitatory glutamatergic terminals . This effect is attributable , at least in part , to a reduction in the size of the RRP of SVs . Our results reaffirm that LPA signaling modulates excitatory synaptic transmission through mechanisms modulating the presynaptic component of the synapse . Next , we explored whether LPA modulates GABAergic and glutamatergic synapses by similar mechanisms of action . Amplitude , but not frequency , of miniature quantal IPSCsGABAA ( mIPSCsGABAA ) recorded in HMNs was reduced by LPA , in agreement with a postsynaptic site of action ( Fig 6A; S9 Fig ) . The molecular cascade downstream of LPA is also distinct , since LPA-induced alterations on mIPSCsGABAA were reversed by the ROCK inhibitor H1152 ( Fig 6A; S9 Fig ) . H1152 also returned ( s- ) LPA-induced changes in eIPSCGABAA amplitude to a control-like state ( S10A and S10B Fig ) . In support of a non-presynaptic action of s-LPA on eIPSCsGABAA , the mean PPR remained similar to the control condition in the presence of s-LPA or s-LPA plus H1152 ( S10C and S10D Fig ) . Colocalization in HMNs of LPA1-ir with the postsynaptic marker gephyrin , a clustering protein for GABAARs [35] , strengthened the evidence of a postsynaptic site of action for LPA ( Fig 6B ) . Postsynaptic action and the molecular signaling underlying LPA-induced modulation of GABAAergic system were assessed in primary cultures of spinal motoneurons ( SMNs ) ( S1 Text; S11 Fig ) . The mean amplitude of inward GABAAR-mediated current evoked by exogenous GABA pulses ( −4 . 13 ± 0 . 98 nA; n = 8 SMNs ) was robustly reduced by s-LPA ( −62 . 5 ± 10 . 1% , p < 0 . 001 , one-way ANOVA for repeated measures ( RM-ANOVA ) ) , in a ROCK-dependent way ( s-LPA+H1152: −3 . 23 ± 0 . 49 nA , p = 0 . 345 ) ( Fig 6C ) . In addition , we observed that s-LPA activated the small GTP-binding protein RhoA , the major ROCK activator , in SMNs . This was evidenced by an s-LPA-induced increase ( +78 . 3 ± 25 . 7%; p < 0 . 05 , one-way ANOVA on Ranks ) in the membrane ( M ) :cytosolic ( C ) ratio of RhoA expression relative to the control condition ( Fig 6D ) . Supplementary data support LPA signaling as the activator for the RhoA/ROCK pathway in motoneurons ( S1 Text; S12 Fig ) . Furthermore , pretreatment with siRNAlpa1 prevented the effects of ( s- ) LPA on GABAAR-mediated currents compared to cRNA-treated SMNs , providing conclusive evidence of postsynaptic LPA1 involvement ( Fig 6E and 6F; S1 Text; S13 Fig ) . Phosphorylation of serine 327 on the GABAAγ2 subunit ( pGABAAγ2 ) regulates GABAAR clustering and synaptic strength at inhibitory synapses [36 , 37] . Therefore , we investigated whether LPA1-ROCK signaling regulates phosphorylation of GABAAγ2 . Contrary to expectations of a direct interaction between ROCK and GABAAγ2 , s-LPA induced a robust reduction ( −83 . 3 ± 5 . 2% ) of the pGABAAγ2:GABAAγ2 ratio in SMNs that was prevented by coaddition of H1152 ( +1 . 6 ± 6 . 0% ) ( Fig 6G ) . This was also observed in the HN ( S1 Text; S14 Fig ) . Strikingly , direct binding of the phosphatase calcineurin ( CaN ) to GABAAγ2 subunits dephosphorylates Ser327 [37 , 38] , which leads to a reduction in inhibitory postsynaptic current amplitude [37] . Therefore , recruitment of CaN ( also named Ca2+/calmodulin-dependent phosphatase 2B ) , was proposed as a potential link between LPA1-ROCK signaling and GABAAγ2 dephosphorylation . Preincubation of SMNs with CaN autoinhibitory peptide ( Cap; 50 μM ) also prevented ( +4 . 8 ± 16 . 5% ) s-LPA from inducing a reduction in pGABAAγ2:GABAAγ2 ratio ( Fig 6G ) . Expression of GABAAγ2 remained unchanged regardless of treatment ( Fig 6G ) . s-LPA also had no effect on the GABA-evoked currents in SMNs pretreated with Cap for 30 min ( Cap: 2 . 2 ± 0 . 3 nA; Cap+s-LPA: 2 . 1 ± 0 . 3 nA; n = 5 SMNs ) ( Fig 6H ) . s-LPA-induced alterations in mIPSCsGABAA and eIPSCsGABAA in HMNs were also CaN-dependent ( S1 Text; S15 Fig ) . Additionally , CaN activity strongly increased in SMNs after incubation with s-LPA , but not with s-LPA plus H1152 or H1152 alone ( Fig 6I ) . Altogether , these data show that ( s- ) LPA , specifically acting through postsynaptic LPA1-RhoA/ROCK-CaN signaling pathway , regulate GABAAR-mediated neurotransmission , by a mechanism involving dephosphorylation of GABAAγ2 subunit at Ser327 . It is generally accepted that dephosphorylation appears to be important for receptor endocytosis [4 , 9] . As a next step , we investigated whether LPA-triggered dephosphorylation was accompanied by further subunit internalization . We found that s-LPA ( 15 min ) led to a strong reduction ( −99 . 9 ± 0 . 01% ) in the amount of GABAAγ2 allocated in M fraction in SMN cultures . A proportional increase ( +109 . 4 ± 14 . 1% ) in the quantity of GABAAγ2 was observed in the C fraction relative to total GABAAγ2 ( Fig 7A ) . These outcomes suggest a translocation of at least this subunit from the SMN membrane to the cytosol triggered by s-LPA . The s-LPA-induced translocation was prevented by coincubation with either the ROCK inhibitor H1152 or the CaN inhibitor Cap ( Fig 7A ) . GABAAγ2 compartmentalization in SMNs was maintained after treatment with H1152 or Cap per se ( Fig 7A ) . To explore whether internalization is actually required for LPA-induced GABAAergic STD , and given that GABAAR endocytosis is dynamin-dependent [39] , we added the dynamin inhibitor dynasore to the bath to block GABAAR endocytosis . Dynasore ( 80 μM for 30 min ) fully prevented both a reduction in the GABAAγ2 M:T ratio and an increase in the C:T ratio induced by s-LPA , which was not altered by vehicle ( −84 . 5 ± 5 . 8% ) . Dynasore per se did not modify GABAAγ2 location ( −6 . 4 ± 18 . 8% ) relative to the vehicle condition ( 100 . 0 ± 36 . 7% ) ( Fig 7B ) . Interestingly , electrophysiological recordings showed that preincubation with dynasore had no effect on s-LPA-induced changes in GABA-evoked currents ( −48 . 1 ± 8 . 7%; n = 4 SMNs ) ( Fig 7C ) . These outcomes support that GABAAγ2 internalization by endocytosis is not required for the attenuation in GABAAergic neurotransmission induced by LPA signaling . CaN-dependent dephosphorylation of Ser327 at the GABAAγ2 subunit is involved in the increase of lateral diffusion and cluster dispersal of surface GABAARs in the dendrites of cultured hippocampal neurons [36 , 40] . Therefore , we investigated whether s-LPA-induced STD under endocytosis inhibition conditions would involve GABAAR cluster disarrangement . Double immunolabelling for GABAAγ2 and the postsynaptic scaffolding protein , gephyrin , confirmed GABAAγ2-ir clusters at the surface of SMNs , most of them colocalized with gephyrin-ir clusters ( Fig 7D ) . In consonance with phospholipid-evoked GABAAR internalization , treatment with s-LPA ( 10 min ) reduced mean fluorescence intensity , but not area , per cluster for these two postsynaptic proteins ( Fig 7E–7G ) . However , the size of surface GABAAγ2-ir clusters increased in parallel with a reduction in fluorescence when s-LPA was added after pretreatment with dynasore ( Fig 7E–7G ) . This agrees with s-LPA-induced lateral diffusion and cluster dispersal of GABAARs . In addition , the mean area of GABAAγ2-associated clusters of gephyrin was unaltered , but fluorescence was reduced by s-LPA under endocytosis inhibition ( Fig 7E–7G ) . These results are compatible with s-LPA-induced disorganization of GABAAR clusters that concludes in receptor internalization . Effects under the presence of dynasore support that this GABAAR disarrangement might involve previous lateral diffusion and cluster dispersal of surface GABAARs like that reported previously for cultured hippocampal neurons [36 , 40] . In summary , our data highlight a pathway by which , via recruitment of RhoA/ROCK signaling , postsynaptic LPA1 evokes CaN-dependent dephosphorylation at Ser327 of the GABAAγ2 subunit , which is followed by GABAAR cluster dispersion and its concomitant translocation from the plasma membrane to the cytosol ( Fig 7H ) . The latter does not seem to be required for the reduction in GABAAergic synaptic strength triggered by LPA . Phospholipid-induced synaptic strength depression seems to be mainly supported by GABAAγ2 dephosphorylation and subsequent GABAAR cluster dispersal . Next , the role of LPA signaling in short-term , activity-dependent synaptic plasticity was explored . N-methyl-D-aspartate receptor ( NMDAR ) activation causes a rapid , local , surface dispersal of synaptic GABAARs leading to an inhibitory synaptic depression [36 , 37] . We directly examined the role of LPA1-mediated signaling in NMDAR-induced STD of GABAAergic signaling in SMNs . In cRNA-treated SMNs , perfusion of glutamate and glycine ( Glut/Gly ) for 4 min caused a rapid and reversible depression in GABA-induced current ( −59 . 6 ± 5 . 3% , p < 0 . 001 ) in the presence of TTX , d-tubocurarine , strychnine and NBQX . This activity-dependent plastic event was absent in SMNs precultured with siRNAlpa1 ( −15 . 2 ± 8 . 7%; Fig 8A ) , in untreated cells under zero extracellular Ca2+ ( −9 . 3 ± 11 . 5%; n = 6 SMNs ) , or in the presence of APV ( −5 . 4 ± 13 . 9%; n = 6 SMNs ) , demonstrating Ca2+- and NMDAR-dependence . LPA1 knockdown reduced by approximately 40% the magnitude of activity-dependent STD at inhibitory synapses . From an extrapolation of these values to the dose-response curve in Fig 2C , it could be indirectly estimated that local concentrations of phospholipids achieved in response to those levels of motoneuron activity were first order micromolar , assuming all synthesized and released phospholipids were the monounsaturated form of LPA ( 18:1 ) . Glut/Gly also caused a drastic decrease in the pGABAAγ2:GABAAγ2 ratio in untreated or cRNA-incubated SMNs , which was prevented by siRNAlpa1 ( Fig 8B ) . Altogether , these data indicate that NMDAR-driven GABA-current depression was spike-independent and essential to extracellular Ca2+ entry via NMDARs and LPA1 activation , which downstream induces Ser327GABAAγ2 dephosphorylation . Findings from activity-dependent synaptic plasticity experiments agree with the notion that motoneurons are potential sources for Ca2+-dependent , spike-independent synthesis and release of lysophospholipids , which in turn might stimulate autocrine signaling pathways ( to modulate inhibitory synapses ) , at least by way of the LPA1 receptor . These outcomes also strongly point to lysophospholipids as paracrine retrograde messengers that act on presynaptic LPA1 to regulate excitatory synapses; however , further research is needed to confirm this possibility . Finally , physiological involvement of LPA signaling in performance of motor output commands was investigated . In vivo , most HMNs exhibit rhythmic inspiratory-related bursting discharges driven by glutamatergic brain stem afferents , mainly acting on AMPARs , with little or no contribution of inhibitory inputs [22 , 23] . We began by analyzing the level and pattern of expression of the LPA1 receptor within the HN of the adult rat . qRT-PCR analysis showed that disparity between lpa1 and lpa2–6 transcripts in the HN was even more accentuated in adults than at the neonatal stage ( Fig 9A ) . Interestingly , mRNA and protein levels for LPA1 at adulthood were approximately 150% and 140% , respectively , higher than in neonatal animals ( Fig 9A and 9B ) . These results suggest a gain in relevance of LPA1-mediated signaling in the HN during postnatal development , supporting previous observations in the murine brain [41] . Immunohistochemistry revealed LPA1-ir puncta-like structures all along the HN ( Fig 9C ) and colocalization between VGLUT2- and LPA1-ir puncta ( Fig 9D , 9E , and 9H ) . A high proportion of VGLUT2-ir inputs ( 47 . 9 ± 3 . 4%; n = 55 HMNs ) apposed to the perikarya of SMI32-identified HMNs were colocalizing with LPA1-ir puncta ( Fig 9E and 9F ) . This also supposed an increase of approximately 150% during postnatal maturation . LPA1-ir appeared to border and colocalize with SMI32-ir structures ( Fig 9G ) , supporting cytoplasmic and membrane location of LPA1 in adult HMNs . Therefore , the molecular machinery to support a role of LPA1 in modulating excitatory neurotransmission is also present in adults . Additionally , in vivo decerebrated rats maintain respiratory activity [22 , 23] . To look for a role of LPA signaling in processing motoneuron inspiratory activity , LPA1/3 inhibitors VPC 32179 ( 0 . 5 mM ) , VPC 32183 ( 1 mM ) , and Ki16425 ( 2 mM ) or its vehicle ( 10% DMSO ) were microiontophoretically applied to antidromically-identified HMNs subjected to unitary extracellular recordings ( Fig 9I ) . The effect of these drugs on the unitary basal firing inspiratory-related activity of HMNs in basal conditions ( end-tidal CO2 = 4 . 8%–5 . 2% ) was evaluated . The time course of the mean firing rate averaged over the duration of the inspiratory burst ( mFR/burst ) was measured by applying increasing currents ( −20 to −140 nA , 30 s duration ) through the drug barrels ( Fig 9J–9L ) . A current-dependent increase in the mFR/burst of HMNs was observed for all drugs but not when current was applied to the vehicle solution ( Fig 9J–9L ) . In summary , these data point to a physiological role for LPA signaling in motor output performance by restraining the inspiratory-related activity driven by glutamatergic inputs to HMNs .
The present study showed that bioactive membrane-derived phospholipids evoke rapid and reversible synaptic depression and mediate activity-dependent synaptic plasticity , mainly via LPA1 . Phospholipids likely operate as local messengers in activity-dependent GABAergic STD in a Ca2+-dependent , spike-independent manner . Strikingly , at physiological concentrations of nanomolar to first order micromolar , LPA has a greater effect on inhibitory than excitatory inputs . Finally , LPA signaling regulates brain-elemental processing tasks such as performance of motor output commands . These data open a new scenario in which the membrane-phospholipid metabolism actively participates in controlling synaptic strength , and then affects neuronal excitability in physiological and pathological states . Important determinants of synaptic strength , short-term plasticity and intersynaptic crosstalk mainly involve fine-tuning of the number of neurotransmitter receptors and the RRP size of SVs [4 , 8] . LPA depresses the main excitatory and inhibitory synaptic systems , affecting both by different degrees , loci , and mechanisms of action . At glutamatergic synapses , and by way of presynaptic Gαi/o-protein-coupled LPA1 and PLC-MLCK activation , LPA results in MLC phosphorylation , which might stimulate the actomyosin contractile apparatus [32] to reduce the bulk of the RRP of SVs ( Fig 10 ) . Depletion of some RRP of SVs usually underlies short-term forms of synaptic depression [1 , 2] . Ultrastructural correlates for LPA-induced STD further supported that functional synaptic changes are partly explained by a reduction in the size of the RRP of SVs . Changes in the actin cytoskeleton are a prerequisite for exocytosis , enabling docking and fusion of SVs with the plasmalemma [32] . As in our results , LPA-dependent contraction of smooth muscle cells involves activation of PLC and MLCK , followed by MLC phosphorylation [31] that promotes actomyosin interactions [32] . In this context , a physical relationship between p-MLC and glutamatergic synapses on adult and neonatal motoneurons has been recently reported [42] . At the calyx of Held synapse , MLCK controls the size of the fast-releasing pool of SVs [43] . In addition , ROCK regulates p-MLC levels via MLCK inhibition to maintain basal RRP ordering of SVs at excitatory inputs [8 , 42] . Therefore , presynaptic LPA-dependent and ROCK signaling seem to converge onto a common molecular mechanism , namely MLC phosphorylation and size of the RRP at excitatory synapses . It is interesting , then , that the ROCK inhibitor did not actually enhance LPA-induced depression of AMPAR currents . These outcomes suggest that the antagonistic functional actions of ROCK and LPA1-signaling , converging on MLCK , results in a push–pull mechanism that regulates the size of the RRP of SVs at excitatory synapses . At GABAergic synapses , LPA dephosphorylates Ser327 of GABAAγ2 subunits and favors GABAAγ2 internalization via postsynaptic Gα12/13-coupled LPA1/RhoA/ROCK signaling and subsequent CaN activation ( Fig 10 ) . The cell surface stability of GABAARs is regulated by post-translational modifications such as phosphorylation . GABAAR phosphorylation is involved in the modulation of receptor biophysical properties and membrane trafficking [44] . Phosphorylation stabilizes the GABAAR on the surface and , conversely , dephosphorylation is important for receptor endocytosis [4] . NMDAR activation causes GABAAR cluster dispersal and lateral diffusion by CaN activation and dephosphorylation of Ser327GABAAγ2 [36 , 40] , leading to long-term depression at CA1 inhibitory synapses [37] . Dispersal could involve receptor clustering at clathrin-coated sites at the plasmalemma , which invaginate and pinch off to form clathrin-coated vesicles . Internalized receptors are then either subject to rapid recycling or are targeted for lysosomal degradation [4] . Our results indicated that the LPA1-RhoA/ROCK-CaN pathway dephosphorylates the GABAAγ2 subunit , which undergoes lateral diffusion , dispersal of clusters , and subsequent endocytosis ( Fig 10 ) . However , endocytosis does not seem to be crucial for LPA-induced functional depression at GABAAergic neurotransmission , which seemed to be mainly supported by GABAAγ2 dephosphorylation and subsequent clusters dispersal of surface GABAARs . The kinetic recovery suggests rapid replenishment of the synaptic GABAAR content , given that re-establishment of inhibitory synaptic strength occurred with 7 to 10 min washing after LPA-induced depression . The coordinated action of kinases and phosphatases , downstream of LPA1-triggered signaling , then plays a pivotal role in controlling neuronal excitability by modulation of GABAAγ2 phosphorylation and receptor recycling . The present results seem controversial in relation to our previous findings demonstrating a presynaptic role for endogenous baseline ROCK activity in the regulation of AMPAergic and GABAAergic neurotransmission [8]; here , we describe that ROCK also acts postsynaptically to mediate LPA-induced depression of the GABAAergic transmission . Whether presynaptic baseline ROCK activity in inhibitory inputs depends on membrane-derived bioactive lipid mediators , such as LPA and/or sphingosine 1-phosphate , remains to be elucidated . Nevertheless , at glutamatergic synapses , ROCK activity is likely independent of LPA1/3 signaling , because inhibitors of these receptors did not mimic AMPAergic STD induced by ROCK inhibition . However , we cannot discard the involvement of another LPAR in maintaining baseline ROCK activity in the synaptic terminals . Interestingly , although presynaptic ROCK is active in our experimental conditions [8] , postsynaptic endogenous activity of ROCK , if any , is even below the level required to reveal its impact on synaptic strength and membrane properties [8] of motoneurons . This could be explained by the differential expression of ROCK isoforms at the two compartments , ROCKα in the postsynaptic site and ROCKβ in the presynaptic one , and/or the lower concentration of ROCKα in motoneurons relative to synaptic structures [8] . Anyway , data suggest that when motoneuron activity is low , presynaptic ROCK activity maintains inhibitory synaptic strength by stabilizing the size of the RRP of SVs . However , after exogenous addition of LPA or when motoneuron activity rises , and subsequent coupled LPA synthesis and/or release occurs , postsynaptic LPA1 stimulates ROCK . This leads to deinhibition by GABAAγ2 dephosphorylation and receptor endocytosis . In the rat , the highest LPA concentration in tissue is found in the brain [12] . Cultured cortical neurons produce LPA at nanomolar concentrations [45] , but LPA levels increase up to 10 μM after injury , trauma , or hemorrhage involving blood–brain barrier damage [46] . Here , physiological concentrations ( nanomolar to first order micromolar ) of LPA affected GABAergic to a greater degree than glutamatergic inputs , achieving maximal and similar affectation at 10 μM . Thus , it is possible that LPA signaling maintains neuronal excitability around a dynamic range , promoting deinhibition at low levels of neuronal activity and depressing excitatory inputs when activity increases , perhaps as part of a homeostatic mechanism that prevents excitotoxicity . Any candidate for coupling synaptic strength to neuronal activity must be regulated by activity at the postsynaptic site . Interestingly , noxious stimulation of primary afferent neurons induces LPA production in the dorsal horn in a glutamate-dependent manner [21] . Here , LPA signaling , mainly via LPA1 , was essential in STD of inhibitory inputs triggered by precedent activity of the neuron . Autocrine LPA signaling was essential for NMDAR-driven GABA-current depression , which depends on extracellular Ca2+ entry passing through NMDARs . Activity-dependent synaptic plasticity occurred independently of the generation of action potentials at the postsynaptic neuron . Postsynaptic [Ca2+] increase and LPA signaling dependence for activity-dependent STD in cultured motoneurons strongly support that this cell type is a potential source for activity-dependent LPA synthesis and/or release . Despite the apparent lack of endogenous LPA signaling affecting synaptic strength in our in vitro model , local iontophoretic application of three LPA1/3 inhibitors increased , in a dose-dependent manner , the baseline inspiratory-related activity of HMNs in the adult rat . This rhythmic inspiratory-related bursting discharge of HMNs is driven mainly by glutamatergic brain stem afferences , with little or no contribution of inhibitory inputs [22 , 47] . There is an apparent gain in relevance of LPA1-mediated signaling in the HN during postnatal development , to the detriment of LPA2–6-triggered pathways , as well as excitatory inputs apposed to adult HMNs express LPA1 . Taken together , these findings support that phospholipids , most likely activating LPA1 at glutamatergic synapses , controlled physiological inspiratory-related activity of HMNs , presumably by restraining their AMPAergic input drive [22] . Thus , endogenous LPA signaling physiologically contributes in the performance of normal patterns of motor output commands in adult animals . Alterations in phospholipid homeostasis affect various pathological conditions , thus attracting increased diagnostic and pharmacological interest [48] . The exquisite balance between excitatory and inhibitory inputs is critical for the proper functioning of the brain , and its imbalance leads to the cognitive impairment associated with neurodegenerative diseases and metabolic syndromes related to obesity , dyslipidemia , lipodystrophy , insulin resistance , and alcoholism [49–51] . In particular , LPA production and/or autotaxin are increased in obesity-associated metabolic diseases [52] , induced hypercholesterolemia [53] , congenital lipodystrophy [54] , as well as in ethanol-fed mice [55] and in patients with Alzheimer disease [56] or multiple sclerosis [57] . In addition , phospholipids uptake in mammalian cells depends on their activation status , a critical support for cellular incorporation of nutrition-derived fatty acids . Imported phospholipids are utilized for production of bioactive lipids , such as LPA [58] , and thereby modify synaptic transmission . Therefore , we can point to LPA as a promising candidate in coupling brain function , by modulating synaptic strength and plasticity , to the metabolic condition of the organism across physiological and pathological states .
Brain stem coronal sections ( 30 μm thick ) and SMNs were processed by immunohistochemistry against vesicular glutamate ( VGLUT2 ) , GABA ( VGAT ) transporters , GABAAγ2 subunit , gephyrin and/or Munc13-1 as synapse-related markers , LPA1 , and/or the nonphosphorylated form of neurofilament H ( SMI32 ) as a motoneuron marker , following standard protocols . Brain stem slices ( 300 μm thick ) incubated for 10 min ( approximately 22°C ) , with aCSF alone , 0 . 2% DMSO ( vehicle ) or with various drug treatments were immediately fixed and processed for electron microscopy analysis . Ultrathin sections ( 70–80 nm thick ) were analyzed at high magnification ( 43 , 000x ) . Only boutons , contacting with motoneurons at the level of the nucleolus , evidencing at least an a . z . were included in this study [8] . Neonatal rats ( P4 ) received an acute injection of siRNAlpa1 , or nontargeting siRNA ( cRNA ) , ( 2 μg/rat ) in 2 μl of RNase-free PBS into the fourth ventricle . The target sequence for the siRNAlpa1 was UCAUUGUGCUUGGUGCCUU . A group of animals was infused with 2 μl of RNase-free PBS ( vehicle ) as an additional control . Primary cultures of SMNs were incubated with 2 . 5 μl of either cRNA or siRNAlpa1 ( each 100 μM ) for 72 h at 37°C . Cells were then collected for qRT-PCR analyses or used for electrophysiological studies . Total RNA was extracted from the HN or cultured SMNs using TRIzol , and 0 . 5 μg of RNA was used for cDNA synthesis with iScript cDNA synthesis . The PCR primers were as indicated in S2 Table . Total protein was extracted from microdissected HNs , NSC34 cells , and membrane and cytosol fractions of NSC34 cells and SMNs . Membranes were blotted with specific antibodies against GABAAγ2 , pSer327GABAAγ2 , LPA1 , p-MLC , MLC , or RhoA . Membranes were also probed with anti-α1-tubulin or anti-β-actin antibodies as control for the total amount of protein contained in each well . Data are expressed as the mean ± standard error of the mean ( SEM ) . The number of analyzed specimens per experimental condition is indicated in figure legends or in the result section . Data were obtained from at least three animals per experimental condition . In ROCK activity , western blotting and qRT-PCR experiments , each individual assay was performed by using tissue samples collected from at least six animals per experimental condition . Quantitative data from ROCK and CaN activity assays , western blot , and qRT-PCR represent the average of , at least , three independent experiments . Applied statistical tests per experimental condition are indicated in figure legends or in results . Post hoc Holm Sidak or Dunn tests were applied for ANOVA for repeated measures or on Ranks , respectively . In all cases , the minimum significance level was set at p < 0 . 05 . | Neuronal networks are modules of synaptic connectivity that underlie all brain functions , from simple reflexes to complex cognitive processes . Synaptic plasticity allows these networks to adapt to changing external and internal environments . Membrane-derived bioactive phospholipids are potential candidates to control short-term synaptic plasticity . We demonstrate that lysophosphatidic acid ( LPA ) , an important intermediary in lipid metabolism , depresses the main excitatory and inhibitory synaptic systems by different mechanisms . LPA depresses inhibitory synaptic transmission by reducing the number of postsynaptic receptors at inhibitory synapses; whereas it depresses excitatory synaptic transmission by decreasing the size of the ready-to-use synaptic vesicle pool at excitatory terminals . Finally , we demonstrate that LPA signaling contributes to the performance of motor output commands in adult animals . Our data documents that synaptic strength and neuronal activity are modulated by products of membrane phospholipid metabolism , which suggests that bioactive phospholipids are candidates in coupling brain function to the metabolic status of the organism . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Membrane-Derived Phospholipids Control Synaptic Neurotransmission and Plasticity |
cis-Regulatory DNA elements contain multiple binding sites for activators and repressors of transcription . Among these elements are enhancers , which establish gene expression states , and Polycomb/Trithorax response elements ( PREs ) , which take over from enhancers and maintain transcription states of several hundred developmentally important genes . PREs are essential to the correct identities of both stem cells and differentiated cells . Evolutionary differences in cis-regulatory elements are a rich source of phenotypic diversity , and functional binding sites within regulatory elements turn over rapidly in evolution . However , more radical evolutionary changes that go beyond motif turnover have been difficult to assess . We used a combination of genome-wide bioinformatic prediction and experimental validation at specific loci , to evaluate PRE evolution across four Drosophila species . Our results show that PRE evolution is extraordinarily dynamic . First , we show that the numbers of PREs differ dramatically between species . Second , we demonstrate that functional binding sites within PREs at conserved positions turn over rapidly in evolution , as has been observed for enhancer elements . Finally , although it is theoretically possible that new elements can arise out of nonfunctional sequence , evidence that they do so is lacking . We show here that functional PREs are found at nonorthologous sites in conserved gene loci . By demonstrating that PRE evolution is not limited to the adaptation of preexisting elements , these findings document a novel dimension of cis-regulatory evolution .
cis-Regulatory DNA elements are essential for the correct activation , repression , and maintenance of gene expression . These elements typically contain multiple short DNA motifs , which are recognised by sequence-specific DNA binding proteins , that either themselves act as activators and repressors of transcription , or recruit other proteins that do so [1 , 2] . One class of cis-regulatory DNA elements is enhancers , which establish gene expression states . Another important class is Polycomb/Trithorax response elements ( PREs ) , first identified in the Drosophila homeotic ( hox ) gene complexes [3 , 4] , where they maintain the transcriptional states of hox genes that have been determined earlier on in development by embryonic enhancers [5–7] . The hox PREs preserve the transcription patterns of their associated genes stably over many cell generations , long after the proteins that bind the enhancers have disappeared . Thus , hox PREs are epigenetic memory elements [8] . Although PREs are similar to enhancers in many ways , the most important functional difference between these two types of elements is that enhancers respond to local differences in concentration of the transcription factors that bind them , whereas the Polycomb group ( PcG ) and Trithorax group ( TrxG ) proteins are ubiquitously expressed; thus , the PRE element responds to the transcriptional state of the promoter [3 , 4] . Since their initial discovery in the hox complexes , it has become clear that PREs regulate several hundred other genes in addition . In both flies and vertebrates , the targets of Polycomb regulation include genes involved in major cell-fate decisions , and in several differentiation and morphogenetic pathways [9–15] . Consistent with the nature of these target genes , the PcG proteins are essential to the correct identities of both stem cells and differentiated cells [16 , 17] . In D . melanogaster , many PRE elements that have similar functional properties in transgenic assays are enriched in preferred pairs of motifs , enabling the identification of a subset of Drosophila PREs by computational prediction [18 , 19] . However , these same elements show no preferred order or number of motif pairs , suggesting that the design of PREs in terms of linear arrangement of motifs is flexible [18] . Furthermore , fly PREs can act many tens of kilobases upstream , downstream , or in the introns of the genes they regulate [9 , 10] , suggesting that their position relative to their cognate promoter is also flexible . This diversity of design among D . melanogaster PREs raises the question of whether these differences are important for function , and whether PRE position at each gene is conserved across different Drosophila species . The bithoraxoid ( bxd ) PRE , which regulates the hox gene Ultrabithorax ( Ubx; FBgn0003944 ) , shows large blocks of conserved sequence across several Drosophila species , supporting the idea that PRE position is evolutionarily constrained [7 , 20] . However , the conservation of the several hundred other PREs in the D . melanogaster genome has not been evaluated , and it is not known whether these PREs are also evolutionarily constrained . The effects of evolutionary changes in enhancers and promoters have been well studied for several individual genes in diverse organisms [21 , 22] . Starting from a known cis-regulatory element in one species , the orthologous sequences in other species have been analysed in terms of evolutionary changes and their impact on regulatory function . These studies have demonstrated that many cis-regulatory elements show rapid motif turnover [23 , 24] . In cases in which function has been evaluated , these studies have shown that some enhancers tolerate evolutionary change without large differences in function [25–27] . On the other hand , there are also many examples of evolutionary differences in enhancer sequences that lead to major phenotypic changes [21 , 22 , 28–30] . Thus , cis-regulatory elements are a potential source of phenotypic diversity , and it has been proposed that positive selection acts primarily on cis-regulatory sequences rather than protein-coding sequences [31–33] . The genomic sequencing of several closely related species has enabled the study of cis-regulatory evolution on a genome-wide scale [34–36] . To overcome the inherent difficulties in identifying cis-regulatory elements in genomic sequence , much effort has been invested in comparative genomic approaches , based on the idea that in closely related species , functional elements will be more conserved than nonfunctional DNA [36–39] . Thus , to date , both gene-specific and genome-wide evaluation of cis-regulatory evolution have been limited to the examination of local changes within elements that are otherwise conserved between species . These studies have given rise to the view that cis-regulatory evolution operates on existing elements , in which small changes create novel functions [22] . However , there is also evidence that argues against local motif turnover as the only source of cis-regulatory evolution . First , although conservation certainly does imply function [1 , 35 , 36] , it does not necessarily follow that all functional elements must be conserved , nor that nonconserved DNA has no function [2 , 37 , 40] . Indeed , it has been shown theoretically that new elements may arise at a certain frequency from nonfunctional sequences [41 , 42] , generating functional elements that reside at nonorthologous positions in the genomes of related species . Consistent with this prediction , a recent genome-wide chromatin immunoprecipitation on chip ( ChIP-chip ) study in D . melanogaster embryos has demonstrated that many transcription factor binding sites are not evolutionarily conserved , suggesting that comparative genomics has limited ability to identify true functional cis-regulatory elements [2] . By definition , computational approaches based on genome alignment alone cannot identify cis-regulatory elements whose sequence and genomic position is not conserved . Thus , with this approach , it has not been possible to evaluate any aspect of cis-regulatory evolution beyond local motif turnover . An alternative means to ascertain whether more radical types of cis-regulatory evolution do indeed occur would be to begin by analysing single genomes using computational prediction tools , and subsequently , to compare results across several genomes . Since all computational predictions are prone to false-positive and false-negative results , an essential final step would be to validate predictions experimentally . In this paper , we use a combination of alignment-independent prediction of cis-regulatory elements [18 , 19] , comparative genomics , and experimental validation to examine cis-regulatory evolution beyond motif turnover for PREs in four Drosophila species . This analysis shows that PRE evolution is extraordinarily dynamic . We show both computationally and experimentally that the numbers of PRE elements , their motif composition , and their genomic position change rapidly in evolution . We identify at least two classes of PREs: those whose positions are constrained in evolution ( such as the hox PREs ) , and those that do not have constrained positions . Remarkably , despite the general conservation of the hox PREs , we identify an extra functional PRE in the Bithorax complex of D . pseudoobscura . By demonstrating that PRE evolution is not limited to the adaptation of preexisting elements , these findings document a novel dimension of cis-regulatory evolution . The implications of these findings for evolutionary diversity are discussed .
We have previously developed an algorithm that predicts PREs in the genome of D . melanogaster by scoring for favoured pairs of binding sites for proteins that act on them [18 , 19] . In [18] , 43 predicted PREs were selected for experimental analysis; 29 of these were enriched for PcG proteins in ChIP experiments in S2 cells . A further 12 of those 14 sites that were not enriched in [18] were found to be strongly enriched for PcG proteins in other cell types , or were confirmed in transgenic assays [10 , 11 , 18] . Thus , over 95% of these 43 predictions were functional in one cell type or another , confirming the predictive power of the algorithm for correctly identifying PRE elements . Comparison of the full set of 167 predictions [18] with genome-wide binding profiles of PcG proteins performed in different cell types or in embryos [9–11] revealed a partial overlap . Using the most statistically stringent score cutoff ( a score of 157 , corresponding to an E-value , or expected number of false positives , of 1 . 0 ) , PREs were correctly predicted at 20% ( 37 of 186 ) of experimentally defined binding sites in Sg4 cells [10] . Lower score cutoffs gave higher coverage of ChIP sites [8]; however , it is not clear how many of the detected PcG binding sites in [10] contain functional PREs . Indeed , a recent ChIP-chip analysis of transcriptional regulators in Drosophila embryos demonstrated that many detected binding sites appear not to be functional [2] . In addition , we predict many PREs at sites at which no ChIP enrichment was observed [10] . These include , for example , the well-characterised Fab-7 PRE [43] . For a selection of these predicted sites , ChIP in other cell types ( 9/12 positive ) and transgene analysis ( 3/3 positive ) have confirmed that they are indeed bona fide PRE elements and not false-positive predictions [9 , 18] . The fact that these predicted and verified PREs were not all enriched in any one cell type is consistent with the partial overlap observed between three recent genome-wide Polycomb binding profiles ( 28% to 34% ) generated by ChIP or DNA adenine methyltransferase mapping ( DamID ) on different D . melanogaster cell types [8–11 , 15] . Other studies have also observed discrepancies between genome-wide ChIP data and conserved cis-regulatory elements identified by comparative genomics [2 , 36] . Together , these comparisons show that neither ChIP nor computational analysis provides a comprehensive list of all cis-regulatory elements in the genome: computational analysis can identify sites of potential function , whereas ChIP gives a measure of cell-type– or developmental-stage–specific deployment of these elements . For this reason , in the present study , we combine computational prediction of PRE elements with ChIP and transgenic analysis of specific loci . To assess the evolutionary behaviour of PREs independent of genome alignment , we applied the algorithm to four Drosophila genomes: D . melanogaster , D . simulans , D . yakuba , and D . pseudoobscura . The algorithm was trained on D . melanogaster PRE sequences . Its performance on other Drosophila genomes was confirmed by comparison of PRE predictions in the homeotic Bithorax complexes of all four species , showing that well-characterised PREs in D . melanogaster are also predicted with high significance at orthologous sites in the three other genomes ( Figure 1A ) . In addition , antibodies raised against D . melanogaster PcG proteins were confirmed in the other three species by western blot ( Figure 2F ) and were used for ChIP . This analysis showed that PcG proteins were enriched on the predicted PREs of the Bithorax complex in embryos of all four species ( Figures 1B and S1 , and unpublished data ) . Interestingly , Polycomb protein ( PC ) and Polyhomeotic protein ( PH ) were detected at similar levels on the bxd PRE in D . melanogaster , but at different levels on the bxd PRE in the other species ( Figure 1B ) . Similar behaviour was also detected in other ChIP experiments ( Figures 3C , 4B , and 4D ) . It is unlikely that these differences arise from different antibody affinities in the different species , because both the PC and PH antibodies gave essentially identical results in western blots on embryonic extracts of the four species ( Figure 2F ) . Furthermore , the differences in ChIP enrichments are not consistently higher for a given antibody or species ( see , for example , Figure 4B and 4D ) . We reason that these differences may arise from the fact that we used embryos for the ChIP experiments . The ChIP results represent an average of binding levels for a mixture of cell types , and a range of embryonic stages from 0–16 h . We observed that embryonic development in the four species proceeds at slightly different rates , which would affect the distribution of embryonic stages in a 0–16-h collection , and may therefore affect the observed binding levels of PC and PH . Alternatively , the different binding of PC and PH may reflect different species-specific compositions of PcG complexes at different PREs . In order to measure PRE function by independent means , we used a transgenic reporter assay in which a PRE sequence is linked to the miniwhite gene . The predicted bxd PRE ( Figure 1A ) from all four species showed typical PRE behaviour in this assay in D . melanogaster , giving pairing-sensitive repression and variegation of miniwhite , and response to PcG and trxG mutations ( Figure 1C and 1D ) . Taken together , these results indicate that the DNA sequence criteria for PRE function are essentially identical in all four species , and that the D . melanogaster PRE prediction algorithm is applicable to the other three genomes examined here . For PRE prediction in a single genome , we previously used a stringent score cutoff of 157 , corresponding to an E-value ( expected number of false positives ) of 1 . 0 [18] . This emphasis on specificity had costs for sensitivity: with a score cutoff at 157 , only 20% of sites identified by a later ChIP study were predicted [10] . Aiming to improve sensitivity without costs for specificity , we took steps to adapt the algorithm . We first tested binding sites for other proteins such as DSP1 ( FBgn0011764 ) [44] , GRH ( FBgn0259211 ) [45] , and SP1/KLF ( FBgn0020378; FBgn0040765 ) [46] . However , the inclusion of these sites did not improve the predictive power of the algorithm , but merely lowered the stringency ( M . Rehmsmeier , T . Fiedler , and A . Hauenschild , unpublished data ) . The original motif set [18] was thus used for further experiments . We reasoned that the inclusion of comparative genomic data could increase the predictive power of the algorithm . The presence of a high-scoring hit at an orthologous or close position in a second genome would increase statistical confidence . Thus , for the present study , we employed this principle to calculate a sliding scale of score thresholds ( Figure 2A; Materials and Methods ) which in effect gives a bonus to low-scoring predictions in one genome that are close to high-scoring predictions in another genome . This indeed improved the predictive power of the algorithm , without costs for specificity . At an E-value of 1 . 0 , the overlap between D . melanogaster predictions and published ChIP data [10] was increased from 20% to 34% . In summary , this “dynamic” scoring system increases the sensitivity of the algorithm by taking account of comparative genomic information , but does not exclude elements that occur at nonconserved positions . Using this approach , we performed PRE predictions on the four genomes in all possible pairwise combinations . In each search , the starting point was a set of predictions that scored highly ( above 157 ) in a single genome . Remarkably , in these single-genome analyses , the number of predicted PREs in D . pseudoobscura ( 560 ) was over twice that predicted in any of the other species ( D . melanogaster: 201 , D . simulans: 143 , and D . yakuba: 203 ) , despite almost identical genome size [35] . To evaluate interspecies differences in PRE number by independent experimental means , we examined the distribution of PC by immunofluorescence on polytene chromosomes prepared from third instar larvae of the four species . This analysis detected over twice as many PC bands in D . pseudoobscura as in the other three species , consistent with the prediction of over twice as many PREs ( Figure 2C and 2E ) . Essentially identical results were obtained for PH ( unpublished data ) . The anti-PC and -PH antibodies were raised against the D . melanogaster proteins , but detected the PC and PH proteins equally efficiently in a western blot of all three other species ( Figure 2F ) . Nevertheless , to confirm that differences in band number do not reflect differences in antibody behaviour , polytene stainings were also performed with antibodies against histone H3 trimethylated at lysine 27 ( H3K27me3; Figure 2D and 2E ) . This epitope is identical in all four species and is a hallmark of PcG action that is conserved from flies to vertebrates [8] . The band numbers , calculated from analysis of multiple chromosome spreads , were similar for PC and H3K27me3 in all four species , and were consistently approximately twice as high in D . pseudoobscura as in the other three species ( Figure 2E ) . This analysis confirms that in salivary glands , D . pseudoobscura has at least twice as many binding sites for PC protein as any other species , and is consistent with the results of the prediction . To ascertain how many genes with predicted PREs D . melanogaster and D . pseudoobscura have in common , we compared genes that were in the neighbourhood of the PRE , not farther than 10 kb from its closest end . Including PREs predicted with a fixed genome-wide cutoff and PREs predicted with our dynamic scoring scheme , we thus determined 166 genes unique to D . melanogaster , 349 genes unique to D . pseudoobscura , and 112 genes common to both species . This indicates that not only the numbers of PREs differ between D . pseudoobscura and the melanogaster subgroup , but also the identities of the genes they regulate . Despite these differences in PRE number , we expected that a large proportion of PREs would have conserved genomic position . To ascertain whether this is indeed the case , we compared each predicted PRE in a given genome to its nearest counterpart , identified by dynamic scoring in a second genome . For each PRE hit in the first genome , a BLAST search was performed on the second genome , and the distance between the BLAST hit and the nearest statistically significant PRE was calculated ( Figure 2A ) . Figure 2G shows the distribution of these distances for D . melanogaster versus D . yakuba ( triangles ) and for D . melanogaster versus D . pseudoobscura ( squares ) . Surprisingly , despite the statistical bonus given to PRE pairs with conserved position , this analysis predicts that many PREs do not have conserved position . For example , in the D . melanogaster–D . yakuba comparison , although approximately 140 PRE pairs are within 1 kb of each other , we predict 30 pairs that are separated by over 10 kb . In the D . melanogaster–D . pseudoobscura comparison , PRE positions are less conserved still , with approximately 80 pairs within 1 kb , and approximately 80 that are over 10 kb apart in the two genomes . The PREs that have the highest conservation of position in all four genomes are listed in Table S1 . In summary , these data predict that there are at least two classes of PRE elements: those whose positions are evolutionarily constrained , and those whose positions change rapidly in evolution . To test these predictions experimentally , we performed ChIP on embryos from all four species to evaluate binding of PcG proteins to predicted PRE sites in vivo . We focused on specific examples of two classes of predicted PRE: those that have conserved position , and those that do not . For PREs with conserved position , we selected bxd and spalt major ( salm; FBgn0004579 ) as examples of PREs that have been confirmed in D . melanogaster [4 , 10 , 47] . ChIP analysis in embryos from all four species demonstrated robust PcG binding to these predicted PREs ( bxd , Figure 1B; salm , Figure 3C ) , indicating that these sites do indeed have PRE function in all four species . In the case of the salm PRE , the D . pseudoobscura prediction has a score that is significant only in the context of the double-genome search , and would not have been retrieved in a search of the D . pseudoobscura genome alone , demonstrating the value of the dynamic scoring system . The bxd and salm PREs reside in orthologous regions in all four genomes , enabling us to ask whether the motifs that contribute to PRE function are located in the regions of highest conservation [20] . Unexpectedly , this was not the case ( Figure 3A and 3D ) . The highest conserved regions ( dark-grey boxes on D . melanogaster and D . pseudoobscura diagrams in Figure 3A and 3D ) are typically devoid of PRE motifs . We examined other PREs that have conserved positions , and all showed a similar clustering of motifs in the less conserved regions ( Figure S1 , Table S1 , and unpublished data ) . Where minimal functional PRE fragments have been defined [48–51] , these do not map to the sites of highest conservation ( Figures 3A and S1 , and unpublished data ) . This raises the question of whether these highly conserved regions are important for other functions . Although specific roles have not been reported for these sequences , they may contain promoter targeting sequences , boundary elements , or specific enhancers . Alternatively , they may contain unidentified motifs that are important for endogenous PRE function , but that are not required for minimal PRE function in reporter assays [7] . Furthermore , although each PRE has one or more clusters of motifs , the position and order of motifs within the cluster is not conserved . This is most striking in the D . melanogaster–D . pseudoobscura comparison ( red motifs , Figure 3A and 3D ) , but is also true to a lesser extent for pairs of PREs in more closely related species ( D . melanogaster , D . simulans , and D . yakuba; Figure 3A and 3D ) . Other PREs that have conserved positions showed similar motif turnover ( Figure S1 , Table S1 , and unpublished data ) . This rapid evolutionary turnover of motifs in PREs has been noted for the bxd PRE [7] and is similar to the turnover that has been observed in enhancer and promoter sequences [2 , 26 , 29 , 37 , 38] , which suggests that motif turnover is a general feature of many classes of regulatory elements . We next selected examples of PREs that are predicted not to have conserved position , and used ChIP and transgenic assays to evaluate PRE function of the orthologous and nonorthologous sequences within selected loci . For this analysis , the trachealess ( trh; FBgn0003749 ) , decapentaplegic ( dpp , FBgn0000490 ) , and abdominal-A ( abd-A; FBgn0000014 ) loci were selected ( Figure 4 ) . At the trh locus , a PRE is predicted close to the promoter in the three most closely related species , D . melanogaster , D . simulans , and D . yakuba ( Figure 4A , top three panels , site 2 ) . This predicted PRE was also robustly bound by PcG proteins in embryos of these three species ( Figure 4B , site 2 ) . However , in D . pseudoobscura , although the trh coding region is well conserved , no PRE was predicted at the promoter ( Figure 4A , bottom panel , site 2 ) . Consistent with this prediction , ChIP analysis showed only moderate enrichment for PcG proteins at this site ( Figure 4B , site 2 ) . Instead , the strongest PRE prediction in the D . pseudoobscura trh locus is within the second intron ( Figure 4A , site 1 ) . Higher PcG enrichment at this intronic PRE than at the promoter site was detected in D . pseudoobscura , whereas this site was less enriched than the promoter site in the other three species ( Figure 4B , site 1 ) . This analysis suggests that whereas in D . melanogaster , D . simulans , and D . yakuba , the main site of PRE function is at the promoter , in D . pseudoobscura , PRE function is situated at the intron site some 5 kb away . For dpp , the situation is more complex: there are three predicted PRE sites , which have different scores in different species . Site 1 is approximately 12 kb upstream of the dpp promoter , site 2 is 5 kb upstream , and site 3 is at the promoter ( Figure 4C ) . In D . melanogaster and D . simulans , the predicted PRE score and the enrichment for PcG proteins at site 1 are higher than at sites 2 and 3 ( Figure 4C and 4D , top two panels ) . Of the three sites in D . yakuba , site 2 has the highest PRE score and showed the highest PcG enrichment . In D . pseudoobscura , the highest PRE prediction is at site 3 ( the promoter site , Figure 4C ) . This site is bound by PcG proteins in D . pseudoobscura , but no binding above background was detected in the other three species ( Figure 4D ) . Taken together , these results indicate that , like those of the trh locus , the dpp PREs are at different sites in different species , suggesting that gain or loss of PRE function at orthologous sites has occurred during evolution . In several cases , a PRE was predicted in one species , but had no detectable counterpart in other species . Two such examples are shown in Figure S2 ( in the unpaired 2 locus ) and in Figure 4E–4H ( in the Bithorax complex ) . The Bithorax complex of D . melanogaster contains the best-characterised PREs , which act to maintain expression domains of the three hox genes Abdominal-B ( Abd-B; FBgn0000015 ) , abd-A , and Ubx . In all four species examined , the PREs of the Bithorax complex were predicted at well-conserved positions ( Figure 1A ) , with one notable exception: an extra PRE 10 kb upstream of abd-A is predicted in D . pseudoobscura ( Figure 1A , bottom panel , asterisks; Figure 4E ) . Strikingly , the orthologous sequences in D . melanogaster , D . simulans , and D . yakuba have PRE scores of less than 20 ( Figure 1A , top three panels , asterisks ) and have very few PRE motifs ( Figure 4E ) . The predicted extra D . pseudoobscura PRE was bound by PcG proteins in D . pseudoobscura embryos ( Figure 4F , top panel ) , indicating that it may indeed be a functional element . The orthologous sequences showed no detectable PcG binding in any of the other species , suggesting that this element does not function as a PRE in D . melanogaster , D . simulans , or D . yakuba ( Figure 4F , bottom three panels ) . To test these observations by independent means , we generated transgenic reporter flies carrying either the predicted D . pseudoobscura PRE or the orthologous D . melanogaster sequence ( Figure 4G and 4H ) . Whereas the D . melanogaster sequence did not show any typical PRE behaviour in this assay , the D . pseudoobscura element showed pairing-sensitive silencing , variegation , and response to PcG and trxG mutations ( Figure 4G and 4H ) , all typical features of PRE elements [4 , 52 , 53] . Thus , we conclude that this extra D . pseudoobscura element is indeed a functional PRE . The presence of an additional functional PRE in the D . pseudoobscura Bithorax complex is intriguing , particularly since the positions of other PREs at this locus are so well conserved . This PRE may be a remnant of an ancestral Bithorax complex , which has lost the PRE at that position in some lineages . Alternatively , the D . pseudoobscura PRE may have arisen from nonfunctional sequence and been fixed by positive selection . To evaluate these two possibilities , PRE scores were calculated for the orthologous sequences at this position in eight Drosophila genomes [35] . This analysis showed a statistically significant PRE score for this site in D . ananassae , D . pseudoobscura , and D . persimilis , but not in the melanogaster subgroup . A maximum likelihood analysis suggests that the PRE was present in the common ancestor of the species under consideration and was lost in the melanogaster subgroup ( Figure S3 ) . To gain further insight into global gain and loss of PREs during the evolution of the D . melanogaster lineage , we carried out genome-wide comparisons with eight genomes as described in Materials and Methods . From this analysis , it can be inferred that 33 PREs have been gained in D . melanogaster ( Figure S4 and Table S2 ) . For only one of these 33 PREs , the nearest gene , scribbled ( scrib; FBgn0026178 ) , has another PRE , and gene CG12852 ( FBgn0085383 ) has gained two PREs , without having a further one . Thus , 30 of these PREs are associated with genes that previously had no PRE . Taken together , these data indicate that PREs can arise from nonfunctional sequence , and furthermore suggest that genes can newly acquire PcG regulation .
For the PREs that have changed position , there are several possible mechanisms by which a PRE may be lost from one site and gained at another , all of which may be at play in shifting the PRE landscape between species . For example , PREs may move by a simple microinversion event [54] . However , the evolutionary plasticity that we document here mainly involves the loss or gain of PRE function from orthologous sequences that do not contain inversions , thus other mechanisms must be considered . First , PREs may move by “creeping” from one site to the other . In this model , a sequence adjacent to a PRE may acquire new functional motifs , thus shifting the centre of PRE function to a slightly different location . By accumulation of such small shifts , the PRE could effectively move to a new position . Sequence insertions could accelerate this process . We observe such an insertion in the salm PRE ( Figure 3C and 3F ) , in which a single motif cluster spanning approximately 600 bp in the D . melanogaster PRE has split into two clusters in D . pseudoobscura , which are separated by an insertion of a few hundred base pairs . Second , ancestral PREs may lose their function at different sites in different lineages , resulting in an apparent change of position . Third , a PRE could change its position by de novo evolution from nonfunctional sequence . We infer from comparative genomics that this is the case for at least 35 PREs in D . melanogaster . It has been shown theoretically that enhancers could evolve rapidly from nonfunctional sequence , provided that the DNA motifs are simple , and that there is sufficient raw material in the form of “presites” that differ from functional sites by a single nucleotide [41] . This suggests that , as proposed [55] , nonfunctional sequences may be “elected” to take up a role as PREs by relatively few nucleotide changes . We have examined this possibility for selected Drosophila PREs that occur at nonorthologous positions in different species by allowing single base changes in any motif and plotting sites of “pre-PRE” potential . We find that sites of PRE function in one species correspond to sites of high potential in a second species , so that a new PRE could theoretically emerge with very few nucleotide changes ( Figure S5 ) . What is the evolutionary significance of PRE plasticity ? Many studies of enhancers have shown that small differences in sequence can lead to large phenotypic differences [21 , 22 , 28–30] , thus one may expect the same to be true for PREs . However , it is important to bear in mind one important functional difference between enhancers and PREs , namely that enhancers respond to differences in cellular concentrations of the transcription factors that bind them , whereas PREs respond to the activity state of their cognate promoter , and not to local differences in the concentrations of the PcG and TrxG proteins [8] . Thus , PREs may be more tolerant than enhancers to changes in number of binding sites , and indeed to changes in the number of PREs at a given locus . On the other hand , the only feature of enhancers that has been studied is motif turnover . It remains to be seen whether enhancers display evolutionary plasticity similar to that of PREs . Given the flexible nature of PRE design , we envision several possible effects of evolutionary plasticity , which may operate differently at different PREs . First , many differences in PRE number and sequence between species may be tolerated by the organism without causing large phenotypic differences . Indeed , the body plans of the different species are very similar . Thus , some PREs may work to maintain phenotype in the face of environmental differences . For example , one of the most important environmental constraints on different Drosophila species from different latitudes is temperature . In D . melanogaster , the PcG proteins are profoundly sensitive to the temperature at which the flies are raised [52] , giving more potent silencing at higher temperatures . Thus , for some PREs , the plasticity in design that we observe may play a role in “buffering” the system against different temperatures , such that the transcriptional output of the locus is conserved . In addition , PREs may mediate phenotypic plasticity for thermosensitive traits such as pigmentation . Several of the loci involved in the plasticity of pigmentation ( e . g . , Abd-B ) are regulated by PREs [56] . On the other hand , for some PREs , differences in design may have a direct effect on phenotype . Several studies have documented large effects on PRE function caused by changes in one or a few binding sites [44 , 57 , 58] . Thus , we propose that some of the changes we observe would affect the silencing or activation response of the PRE , thus in turn affecting the level of target gene transcription that is maintained , and giving selectable effects on phenotype . For example , one of the major phenotypic differences between Drosophila species is the male sex combs . The sex comb is one of the most rapidly diversifying organs in Drosophila species , and is important for male reproductive success [59] . Evolutionary diversity in sex comb number is associated with diversity in regulation of the hox gene Sex-combs reduced ( Scr ) , which is a well-characterised target of PcG regulation [60 , 61] . In D . melanogaster , D . simulans , and D . yakuba , a single row of sex comb teeth is present , whereas D . pseudoobscura has two such rows . Interestingly , a microinversion event on the 3′ side of the D . pseudoobscura Scr locus [54] has removed 3′ regulatory sequences , including one of a cluster of three Scr PREs , to a new position . The D . pseudoobscura PREs also show many sequence changes compared to the other three species ( unpublished data ) . Thus , differences in PRE sequence , number , and position at the Scr locus correlate well with phenotypic differences , and will provide an excellent model for further study of the effects of PRE plasticity on phenotype . In summary , PREs act on several hundred genes in Drosophila , many of which are master developmental regulators . We propose that the extraordinary plasticity in PRE design that we observe may provide a rich capacity for transcriptional buffering , phenotypic plasticity , and phenotypic diversity between species .
BLAST search . The BLAST search takes a PRE predicted in one species and determines the orthologous position in another species . Because the PRE will usually not be conserved as a continuous sequence , multiple adjacent high-scoring pairs ( HSPs ) have to be grouped together . The grouping is done according to the following criteria: only HSPs with a BLAST E-value not larger than 0 . 01 are considered . HSPs of one group are on the same strand . The distance between adjacent HSPs of one group is below 1 kb . Groups are maximal in the sense that no HSPs can be added that fulfil these three criteria . From all groups that correspond to one initial PRE , we choose the one with the largest sum of HSP lengths . From several groups with the same length sum , the one is taken that happens to be the first processed ( a case which has not occurred in our analysis so far ) . Starting with 201 PREs in D . melanogaster ( version 4 . 0 ) , this procedure resulted in 190 orthologous regions in D . pseudoobscura ( version 2 . 0 ) , 194 in D . simulans ( version 1 . 0 ) , and 176 in D . yakuba ( version 1 . 0 ) . In D . yakuba , an additional 20 fall into “chr2L_random , ” which contains clones that are not yet finished or cannot be placed with certainty at a specific place on the chromosome . These 20 hits were not included in our analysis . Finding the right locus . To evaluate the validity of the BLAST search procedure , we checked whether orthologous regions were in correct loci . For each PRE from D . melanogaster and its orthologous region in D . pseudoobscura , we compared the distance between the PRE and the two genes closest to it with the distance of the orthologous region and the two genes closest to that . If a PRE was inside a gene , only that gene was included into the comparison . In the majority of cases ( 163 out of 190 ) , this “locus shift” is below 10 kb , although it can become larger than 200 kb . In some cases ( 24 ) , the ortholog of the D . melanogaster PRE and the ortholog of one of the possibly two D . melanogaster genes are found on different chromosomes . In general , there are legitimate doubts about the reliability of the D . pseudoobscura gene annotation . Frequently , one or more exons are missing , which leads to too large a distance between PRE ortholog and closest gene in D . pseudoobscura . Additionally , we can show that the observed rare events of chromosome changes are consistent with the gene rearrangement in the annotation . For example , the gene CG1924 is located on chromosome X in D . melanogaster and on chromosome 2 in D . pseudoobscura , whereas the adjacent genes are on chromosome X in both species . Calculating BLAST distances ( Figure 2D ) . A BLAST distance is calculated as the difference between , first , the distance between the centre of the query sequence ( predicted PRE or random ) and the centre of the BLAST hit in the query genome ( D . melanogaster ) , and second , the distance between the centre of the putative functional analog and the centre of the BLAST hit in the target genome ( D . yakuba or D . pseudoobscura ) . We cannot directly calculate the distance between BLAST hit and analog in the target genome only , since BLAST hits are not necessarily centred around the query sequence . PRE prediction and calculation of dynamic scoring thresholds . PRE prediction was performed using the jPREdictor software [19] , which follows the PREdictor algorithm as described in [18] , except that a step size of 10 bp instead of 100 bp was used . Score cutoffs and E-values were calculated with a nonparametric statistics on random sequence data 100 times the size of the D . melanogaster genome , with the D . melanogaster nucleotide distribution ( 29% A , 21% C , 21% G , and 29% T ) . A score s such that scores of s or better occur r times in the random data , corresponds to an E-value of r/100 in the single D . melanogaster genome . For an E-value of 1 , this score cutoff is 157 . For the dynamic scoring system , cutoffs were calculated similarly , taking into account the smaller search spaces of 1 kb , 10 kb , and 20 kb radius and the fact that about 200 such searches are performed ( see Figure 2A ) . All PREs predicted in D . melanogaster and D . pseudoobscura will be available at http://bibiserv . techfak . uni-bielefeld . de/fly_pres upon publication . Evolutionary gain and loss of PREs . We performed a maximum likelihood analysis of 73 D . melanogaster PREs in eight Drosophila genomes . Each of these 73 PREs had been genome-wide predicted , its orthologous regions could be determined in all the other seven species , and at least one of the other species had no functionally analogous PRE . A functionally analogous PRE was defined as a hit predicted dynamically within a 10-kb BLAST distance . The eight species comprise those for which the efficacy of our predictive method has been well established ( up to D . pseudoobscura ) . We employed a probabilistic model whose separate gain and loss parameters were estimated with the Mesquite software ( http://mesquiteproject . org ) on the given contemporary character states: 1 for a ( functionally analogous ) PRE being present in the respective species , 0 for no such PRE being present . Subsequently , maximum likelihood ancestral character states were reconstructed based on the estimated parameters . Defining a D . melanogaster PRE whose most ancestral node ( the root of the tree ) has a PRE likelihood of smaller than 0 . 5 as being gained during evolution resulted in 33 such PREs , listed in Table S2 . Figure S4 shows the trees for the 73 PREs . Strains and handling . For polytene chromosomes and ChIP , D . melanogaster wild-type flies ( Oregon R ) were used . For the other species , the strains used for whole-genome sequencing were obtained from http://stockcenter . arl . arizona . edu/ . Stock numbers: D . yakuba 14021-0261 . 01; D . simulans 14021-0251 . 195; and D . pseudoobscura 14011-0121 . 94 . With the exception of D . pseudoobscura , all species were raised on cornmeal food . For D . pseudoobscura , standard banana-Opuntia food was prepared as specified at http://stockcenter . arl . arizona . edu/ . Genomic fragments of 1 . 5 to 1 . 6 kb were amplified by PCR from genomic DNA of each species and cloned using SpeI/NotI sites into the pUZ P-element vector upstream of the miniwhite reporter gene [18] . Embryo injections were carried out by Vanedis Drosophila injection service ( http://www . vanedis . no ) . Chromosomal mapping and crosses to PcG and trxG mutants were performed as described [18] . Primer sequences , constructs , and transgenic fly lines are available on request . Polytene chromosomes were prepared from third instar larvae of all four species and stained with rabbit polyclonal anti-Polycomb antibody or anti-H3K27me3 ( provided by Thomas Jenuwein ) as described in [62] . Western blotting . Protein extracts were made from 0–12-h-old embryos for all four species , as described in [63] . Western blots were probed with antibodies against PC , PH , H3K27me3 , or H3 ( Upstate ) . Chromatin immunoprecipitation ( ChIP ) . ChIP on whole embryos of D . melanogaster , D . simulans , D . yakuba , and D . pseudoobscura was performed using anti-PC and -PH antibodies , as described [64] . Two independent chromatin preparations on 0–16-h-old embryos , and two to four independent ChIP assays were performed for each species . Enrichments of immunoprecipitated DNA over input DNA were quantified by real-time PCR using SYBR green ( Sigma ) . Three technical replicates were performed for each primer pair on each chromatin preparation . Primers were designed to amplify a fragment of 100 to 300 bp within the highest scoring region of each predicted PRE ( or the minimal PRE , if known ) , or of the orthologous region in the species in which no PRE was predicted . Primer sequences are available on request .
The FlyBase IDs for the genes and gene products mentioned in this paper are as follows: abd-A ( FBgn0000014 ) ; Abd-B ( FBgn0000015 ) ; CG12852 ( FBgn0085383 ) ; dpp ( FBgn0000490 ) ; DSP1 ( FBgn0011764 ) ; GAF ( FBgn0013263 ) ; GRH ( FBgn0259211 ) ; H3 ( FBgn0001199 ) ; KLF ( FBgn0040765 ) ; PC ( FBgn0003042 ) ; PH ( FBgn0004861 ) ; PHO ( FBgn0002521 ) ; salm ( FBgn0004579 ) ; scrib ( FBgn0026178 ) ; SP1 ( FBgn0020378 ) ; trh ( FBgn0003749 ) ; Ubx ( FBgn0003944 ) ; upd 2 ( FBgn0030904 ) ; and ZESTE ( FBgn0004050 ) . | The evolution of regulatory DNA plays a crucial role in making species different from one another . One way to study the evolution of regulatory DNA is by genome alignment , which assumes that elements with conserved function will be found in conserved pieces of DNA . Although conservation does imply function , it does not follow that all functional elements must be conserved , nor that nonconserved DNA has no function . However , computational approaches based on genome alignment alone cannot identify any kind of evolution beyond small changes in otherwise conserved elements . We have used a novel computational approach , in combination with experimental validation , to examine how regulatory DNA evolves in four Drosophila species . We focus on Polycomb/Trithorax response elements ( PREs ) , which regulate several hundred developmental genes , and are vital for maintaining cell identities . We find that PRE evolution is extraordinarily dynamic: not only motif composition , but also the total number of PREs , and even their genomic positions , have changed dramatically in evolution . By demonstrating that the evolution of PREs goes far beyond the gradual adaptation of preexisting elements , this study documents a novel dimension of regulatory evolution . We propose that PRE evolution provides a rich source of potential diversity between species . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"computational",
"biology",
"evolutionary",
"biology",
"genetics",
"and",
"genomics"
] | 2008 | Evolutionary Plasticity of Polycomb/Trithorax Response Elements in Drosophila Species |
Trypanosoma cruzi has three distinct life cycle stages; epimastigote , trypomastigote , and amastigote . Amastigote is the replication stage in host mammalian cells , hence this stage of parasite has clinical significance in drug development research . Presence of extracellular amastigotes ( EA ) and their infection capability have been known for some decades . Here , we demonstrate that EA can be utilized as an axenic culture to aid in stage-specific study of T . cruzi . Amastigote-like property of axenic amastigote can be sustained in LIT medium at 37°C at least for 1 week , judging from their morphology , amastigote-specific UTR-regulated GFP expression , and stage-specific expression of selected endogenous genes . Inhibitory effect of benznidazole and nifurtimox on axenic amastigotes was comparable to that on intracellular amastigotes . Exogenous nucleic acids can be transfected into EA via conventional electroporation , and selective marker could be utilized for enrichment of transfectants . We also demonstrate that CRISPR/Cas9-mediated gene knockout can be performed in EA . Essentiality of the target gene can be evaluated by the growth capability of the knockout EA , either by continuation of axenic culturing or by host infection and following replication as intracellular amastigotes . By taking advantage of the accessibility and sturdiness of EA , we can potentially expand our experimental freedom in studying amastigote stage of T . cruzi .
Trypanosoma cruzi is the causative agent of Chagas’ disease , which affects 6–7 million people mainly in Latin America[1] . The parasite is transmitted by a reduviid bug through its contaminated feces , and enters into the mammalian host when the bite site is rubbed or scratched . Chagas’ disease can also be acquired by vertical and perinatal transmissions , blood transfusion or organ transplant , and oral transmission through contaminated food[2 , 3] . Approximately 30 to 40% of infected people develop chronic disease 10 to 30 years after acute infection[1] . Common chronic phase manifestations include cardiomyopathy , arrhythmias , megaviscera , and polyneuropathy . There are currently two drugs available to treat Chagas’ disease; benznidazole and nifurtimox . Both drugs are effective in acute phase of the infection , but efficacy becomes limited once the disease proceeds to chronic phase[4] . Because most parasite carriers do not get a timely diagnosis or have access to the medication , many of them proceed to chronic phase unnoticed or without proper treatment . In addition , aforementioned drugs are known to cause adverse side effects in roughly 40% of the patients[1] . Thus , development of safer new drugs that are effective in chronic phase is a pressing matter . T . cruzi has distinct developmental stages in its life cycle ( Reviewed in [5] ) . In a reduviid bug , the parasite replicates in a form called epimastigote . It differentiates into metacyclic trypomastigote , a non-replicative infectious form , in the insect rectum before being defecated . Trypomastigotes enter a mammalian host and invade the host cell , where they transform into flagella-less amastigotes and replicate intracellularly . Amastigotes differentiate into highly motile blood stream trypomastigotes as they emerge out of the host cells . The parasites then travel through blood stream to infect another host cell , or to be picked up by an insect vector to complete the life cycle . In order to develop a chemotherapeutic agent effective in chronic phase of the disease , it is important to identify a drug target that is essential for the parasite in amastigote stage . Yet , it is difficult to perform a knockout study in this stage of T . cruzi due to inaccessibility of the parasite in the host cells . Accordingly , conventional method to obtain a transgenic amastigote starts with epimastigote transfection . However , this procedure potentially introduces unwanted bias in the resulting amastigote population by selecting for transfectants that are better-fitted in epimastigote or trypomastigote stages . Even though transient transfection of trypomastigote allows to bypass metacyclogenesis step[6] , it still requires active invasion of the host cells and trypomastigote-amastigote differentiation , which still may introduce some bias . To overcome this issue , we focused our attention on extracellular amastigote ( EA ) to utilize it as a tool for direct experimental manipulations . EA can be obtained either by spontaneous appearance in T . cruzi-host co-culture , or by inducing differentiation of trypomastigote in vitro at low pH[7] . It has been reported that EA is morphologically very similar to intracellular amastigote[7–10] and expresses surface glycoprotein SSP-4 , which is a hallmark of amastigote stage parasite[7 , 9 , 11] . EA is capable of infecting culture host cells by inducing the host actin polymerization[12 , 13] , even if the host cell is not a professional phagocyte[8 , 10 , 12–15] . Notably , EA is able to establish infection and kill mice when inoculated intraperitoneally[8] , and free amastigotes can be found in blood stream of infected animals[9] . Traditionally , amastigotes were considered to be non-infectious , but above findings shed light on the important role of EA in T . cruzi dissemination process . Molecular mechanisms of EA take-up by the host cells are beginning to be elucidated as well[16–20] . In the present study , we demonstrate that EA can replicate free of host cells and can be utilized in variety of assays , including exogenous gene expression and CRISPR/Cas9-mediated gene knockout . We also show that susceptibility of proliferating axenic amastigotes to benznidazole and nifurtimox is close to that of intracellular amastigotes in host 3T3 cells . Our strategy to utilize EA expands methodological freedom in amastigote study , and contributes to advance our understanding in this pathogenic stage of T . cruzi .
Epimastigote of T . cruzi Tulahuen strain ( provided by Dr . Takeshi Nara , Juntendo University ) was maintained in liver infusion tryptose ( LIT ) medium supplemented with 10% heat-inactivated FBS at 28°C . Metacyclogenesis was performed in RPMI medium [21] , and differentiated metacyclic trypomastigotes were isolated by DEAE ion-exchange chromatography[22] . Metacyclic trypomastigotes were added to the 3T3-Swiss Albino fibroblast cell culture for infection , and amastigote-containing culture was maintained in DMEM supplemented with 10% FBS and penicillin/streptomycin at 37°C under 5% CO2 in a humidified incubator , until tissue-derived trypomastigotes emerged out to the culture supernatant . Axenic amastigotes were obtained by in vitro amastigogenesis according to the method described by Tomlinson et al . [7] The tissue-derived trypomastigotes were collected from culture supernatant by centrifugation at 2000 ×g for 15 min , and were transformed into amastigotes by incubation in DMEM buffered with 20 mM MES ( pH 5 . 0 ) and supplemented with 0 . 4% bovine serum albumin ( BSA ) for 24 h at 37°C . To obtain intracellularly-derived amastigotes , infected 3T3 cells were detached from a culture flask by trypsin treatment , and centrifuged at 100 ×g for 5 min . The cells were washed by PBS to remove trypomastigotes and EA , and were suspended in 1 mL of Phosphate Saline Glucose buffer ( 1:9 mixture with 1% glucose ) [23] . To release intracellular amastigotes , the cell suspension was passed through a syringe with 27 G needle 40 times to lyse the host 3T3 cells . Unbroken host cells and debris were removed by centrifugation at 100 ×g for 3 min . The intracellular amastigotes in the supernatant were purified by anion-exchange chromatography[23] . All-in-One Fluorescence Microscope BZ-X710 ( Keyence Co . Ltd . , Japan ) was used to capture images of parasites , using appropriate filter sets . For objective lens , S PL FL ELWD ADM 20xC ( NA0 . 45 ) and CFI Plan Apo λ60xH ( NA1 . 40 ) Nikon lenses were used . TRIzol Reagent ( Thermo Fisher Scientific Inc . , USA ) was used to extract total RNA from epimastigote , tissue-derived trypomastigote , intracellular amastigote , EA derived by in vitro amastigogenesis , and axenic amastigotes cultured in LIT medium for 1 , 3 , 5 , and 7 days . Extracted RNAs were treated with DNase I ( Thermo Fisher Scientific ) to eliminate potential genomic DNA contamination , and the samples were purified by phenol/chloroform/isoamyl alcohol extraction and ethanol precipitation . Reverse-transcription was performed by using 1 μg of above RNAs , 0 . 5 μg of Oligo ( dT ) 12-18 Primer ( Thermo Fisher Scientific ) , and Superscript III ( Thermo Fisher Scientific ) . cDNA was synthesized at 50°C for 1 h , and the enzyme was heat-inactivated at 70°C for 15 min . The reaction product was diluted 10 fold before being used for the following qPCR assay . Target gene IDs and sequences of primers used for qPCR are listed in S1 Table . For amplification reaction , TB Green Premix Ex Taq II ( Tli RNaseH Plus ) ( Takara Bio Inc . , Japan ) was mixed with 2 μL of diluted cDNA , 0 . 2 μM each gene-specific forward and reverse primers , and ROX reference dye in the total reaction volume of 20 μL . qPCR was performed according to the manufacturer’s instruction by using StepOnePlus Real-Time PCR System ( Thermo Fisher Scientific ) . Specificity of the reaction was verified by melting curve analysis , and the gene expression was quantitated by relative standard curve method using StepOne Software v2 . 3 . Expression levels were normalized by GAPDH as an internal control . Benznidazole ( Sigma-Aldrich Co . LLC , USA ) and nifurtimox ( Sigma-Aldrich ) were dissolved in DMSO and dispensed into 96 well microplate , and T . cruzi axenic amastigotes in LIT medium ( 1 × 106 cells in 100 μL ) was added to each well . The final concentration of DMSO was 0 . 5% . After 48 h of incubation at 37°C under 5% CO2 in a humidified incubator , 10 μL of resazurin solution ( Sigma-Aldrich ) was added at a final concentration of 3 mM[24] . The plates were incubated for additional 5 h , and the reaction was stopped by addition of 50 μL 3% SDS . Amount of resorufin was quantitated by scanning the microplate by SpectraMax Gemini fluorescent plate reader ( Molecular Devices , LLC . , USA ) at ex . 560 nm/em . 590 nm . EC50 was calculated by fitting the dose response curves with non-linear regression analysis , using " ( inhibitor ) vs . normalized response" model of GraphPad Prism7 software ( GraphPad Software Inc . , USA ) . Host 3T3 cells were seeded onto 96-well black , clear bottom microplate ( Corning Inc . , USA ) at 5 × 103 cells/well in 100 μL of DMEM supplemented with 10% FBS for 4 h to allow cell attachment . Tissue-derived trypomastigotes were added at multiplicity of infection ( MOI ) of 20 , and incubated at 37°C for 24 h . Uninfected trypomastigotes were removed by washing with PBS , followed by addition of 100 μL of DMEM ( 10% FBS ) containing benznidazole or nifurtimox dissolved in DMSO . The final concentration of DMSO was 0 . 5% . After 48 h incubation at 37°C , the media was removed from wells and cells were fixed by addition of 100 μL of 4% formaldehyde for 15 min at room temperature . After fixation , the nuclei were stained by adding 100 μL of 1 . 0 μg/mL Hoechst 33342 ( Thermo Fisher Scientific ) and 0 . 05% Triton X-100 ( Wako Pure Chemical Industries , Ltd . , Japan ) for 15 min , followed by washing wells with PBS 4 times . The plates were imaged by a fluorescence microscopy . Host cells containing more than three amastigotes were considered as infected . EC50 was calculated by fitting the dose response curves with non-linear regression analysis , using “ ( inhibitor ) vs . normalized response" model of GraphPad Prism7 software . The plasmid constructs and the corresponding cell lines are summarized in Supplementary S1 Fig . Expression vector pTREX-attR derived from pTREX-n[25] was provided by Dr . Takeshi Nara . For constitutive expression of EGFP , EGFP gene was amplified by PCR using forward ( 5’-CTCTAGAATGGTGAGCAAGGGCGAGGAGCT-3’ ) and reverse ( 5’- GCTCGAGTTACTTGTACAGCTCGTCCATGCC-3’ ) primers , and ligated into XbaI and XhoI sites of pTREX-attR to generate pTREX-EGFP . For construction of amastigote-specific EGFP expression vector pTREX-EGFP-amastin 3’UTR , the fragment containing amastin 3’UTR upstream of tuzin site[26] ( GenBank: U25030 . 1 ) was amplified using forward ( 5’-GTACAAGTAACTCGAGCGGGTGCATCCACCGTCT-3’ ) and reverse ( 5’-TCGTAAATGGCTCGAGCGCAGGGCGGGCAGCGGC-3’ ) primers . The resulting 800 bp fragment was ligated into 3’ end of the EGFP gene at XhoI site using In-Fusion HD cloning kit ( Takara Bio ) . Streptococcus pyogenes Cas9 sequence ( RefSeq . NC_002737 ) , twice-repeated sequence of the SD40 nuclear localization signals and EGFP sequence were synthesized and ligated into XbaI and XhoI site of pTREX-attR to generate pTREX-Cas9-EGFP . Amastin 3’UTR sequence was inserted into pTREX-Cas9-EGFP as described above to generate pTREX-Cas9-EGFP-amastin-3’UTR . To generate pTREX-mDsRed-Bsd plasmid for expression of mDsRed and blasticidin resistant gene , full-length mDsRed was amplified by PCR using forward ( 5’-TGCTCTAGAATGGCCTCCTCCGAGAACGT ) and reverse ( 5’-CCGCTCGAGCTACAGGAACAGGTGGTGGC ) primers , and resulting fragment was subcloned between XbaI and XhoI sites . Neomycin resistance gene was then replaced by blasticidin selection marker[27] at PspXI and NheI sites . Transfection was carried out using Basic Parasite Nucleofector Kit 2 ( Lonza Inc . Switzerland ) . Briefly , 2 × 107 epimastigote cells in their log phase were suspended in 100 μL of Nucleofector buffer with provided supplement solution , and 20 μg of plasmid was added to the mixture . Electroporation was carried out using program X-14 of Amaxa Nucleofector device ( Lonza ) unless otherwise stated . To generate a stable cell line harboring pTREX-EGFP-amastin-3’UTR ( EGFP-ama ) , pTREX-Cas9-EGFP ( Cas9 ) and pTREX-Cas9-EGFP-amastin-3’UTR ( Cas9-ama ) , the transfectants were selected in LIT medium containing 500 μg/mL G418 for over 4 weeks at 28°C . For transient expression of mDsRed , axenic amastigotes were electroporated with pTREX-mDsRed-Bsd plasmid as described above . The transfected amastigotes were cultured in LIT medium at 37°C under 5% CO2 in a humidified incubator , or applied onto 3T3 cell culture for an infection experiment . To enrich positive transfectants by blasticidin S resistant marker , axenic amastigotes were transferred to LIT medium immediately after electroporation , and 50 μg/mL of blasticidin S ( Wako Pure Chemical Industries ) was added 24 h later . Cells were monitored for the next 6 days for mDsRed expression , and fraction of mDsRed positive amastigotes were quantitated under fluorescence microscopy . Blasticidin-selected transgenic amastigotes were subsequently applied onto 1 × 105 cells of host 3T3 in 24-well plate at MOI of 40 . After 2 days of incubation , the cells were washed 3 times with DMEM to remove amastigotes remained outside of the host cells . Replication of internalized mDsRed-positive amastigotes was monitored for the next 2 days under fluorescence microscopy . gRNA was purchased from IDT ( Integrated DNA Technologies , Inc . , USA ) as two synthetic RNA oligonucleotides , Alt-R CRISPR crRNA and tracrRNA . The target sequences of EGFP , TcCgm1 and mDsRed ( negative control ) were GGTGGTGCAGATGAACTTCA , TAGCCGCGATGGAGAGTTTA and GGACGGCACCTTCATCTACA , respectively . Gene-specific crRNA and universal tracrRNA were annealed to make a complete gRNA , according to company’s protocol . Transfection of gRNA into Cas9-expressing epimastigote was carried out essentially in the same manner as plasmid transfection described above , except 5 μg gRNA was used . For amastigote transfection , 1 × 107 Cas9-ama-expressing EAs were collected immediately after amastigogenesis in pH 5 . 0 , and were resuspended in Nucleofector buffer . After the electroporation , EAs were transferred to 5 mL of LIT medium and incubated at 37°C under 5% CO2 . Cell growth was monitored by counting surviving cell number using Burker-Turk hemocytometer . Propidium iodide was mixed to the EA culture prior to the counting to aid in distinguishing viable amastigotes from dead parasites . For quantitation of EGFP knockout efficiency , percentage of EGFP-positive population was calculated by analyzing the total cell count in bright field images and the EGFP-positive cell count in fluorescent images , using Hybrid Cell Count software of Keyence microscope . Fluorescence intensity of EGFP-positive/negative cut-off was determined by analyzing the images of WT parasites as the background fluorescence . Total of 1000 parasites were analyzed for transfected epimastigotes , and 500 were analyzed for transfected amastigotes . gRNA-transfected Cas9 cells or Cas9-ama cells were harvested at 2 days after electroporation . Cell pellets were stored at -80°C until use . The cells were resuspended in a buffer containing 50 mM Tris-HCl ( pH 7 . 5 ) , 20 mM NaCl , 10% sucrose , 0 . 1% Triton X-100 and 1×cOmplete EDTA-free Protease Inhibitor Cocktail ( F . Hoffmann-La Roche , Ltd . , Switzherland ) for lysis , and cell debris were cleared by a centrifugation for 1 min at 8000 rpm . Guanylyltransferase assay was carried out by incubating 4 μg of cleared lysate in a reaction mixture containing 50 mM Tris-HCl ( pH 7 . 5 ) , 10 mM MgCl2 , 2 mM DTT and 40 μM [α-32P]-GTP for 20 min at 30°C[28] . The reaction was terminated by addition of SDS loading buffer , and the products were resolved on a 10% SDS-PAGE . Radiolabeled enzyme-GMP covalent adducts were visualized by BAS-2500 phosphorimager and quantitated by Image Gauge 4 . 0 software ( Fujifilm Corp . , Japan ) .
We first tested the multiplication capability of amastigote outside of the host cell . EA was obtained by differentiation of tissue-derived trypomastigotes by incubating the parasite in acidic DMEM , buffered with MES ( pH 5 . 0 ) and supplemented with 0 . 4% BSA , for 24 h at 37°C . EA was subsequently cultured in LIT medium ( 10% FBS ) or DMEM ( 10% FBS ) at 28°C or 37°C , and their growth was monitored for the next 10 days . Axenic amastigotes replicated most efficiently in LIT medium at 37°C ( Fig 1A , closed circle ) . In this condition , amastigotes continued to proliferate for a week before replication slowed down and eventually ceased past day 8 . Intracellular amastigotes obtained by host cell rupture also replicated in LIT medium at 37°C , and showed similar growth pattern as in vitro-derived EA ( Fig 1A , closed triangle ) . DMEM at 37°C did not support the growth of axenic amastigote ( Fig 1A closed square ) . At this temperature , amastigotes in all groups retained typical round morphology throughout the observation period ( Fig 1B ) . On the other hand , EAs replicated only for a few days in LIT medium when incubated at 28°C ( Fig 1C , closed circle ) . In addition , some amastigotes started to transform into intermediate morphologies on day 5 , which resembles epimastigote , trypomastigote and spheromastigote , based on their shapes and nuclear staining patterns ( Fig 1D ) . By day 8 , the number of those intermediate forms were up to 30% of the total number of the parasites in LIT 28°C ( Fig 1C , closed and open circles ) . Axenic amastigotes in DMEM at 28°C did not replicate , although their morphology remained as round form throughout the observation period ( Fig 1C , closed square ) . These results are consistent with previously reported observations that EA can replicate free of host cells , given appropriate media condition and temperature setting[10 , 29 , 30] . We also found that amastigotes derived from both in vitro amastigogenesis and from host cell rupture have comparable proliferation capability in axenic environment ( Fig 1A , closed circle and triangle ) . For the rest of experiments in this paper , we used LIT medium at 37°C to maintain axenic amastigote culture . We use the term “EA” for extracellular amastigote in general or amastigote soon after in vitro amastigogenesis , and “axenic amastigote” for amastigote cultured free of host cells for more than two days . Since trypanosomatids transcribe their mRNAs polycistronically , the amount of each mRNA is controlled mainly post-transcriptionally , and 3’UTR plays a major role in differential gene expression in trypanosomatids ( Reviewed in [31] ) . Amastin is a family of surface glycoproteins , which is most abundantly expressed in amastigote stage of T . cruzi[32] . To investigate whether axenic amastigotes maintain amastigote-specific 3’UTR-mediated gene regulation during prolonged cultivation , we generated a cell line that expresses EGFP under the control of amastin 3’UTR[26] . Control cells harboring EGFP construct without stage-specific 3’UTR expressed EGFP in all developmental stages ( S2 Fig , EGFP ) , whereas EGFP-amastin 3’UTR cell line ( EGFP-ama ) expressed EGFP in intracellular amastigote stage but not in epimastigote or trypomastigote stages ( S2 Fig , EGFP-ama ) . When non-fluorescent , tissue-derived trypomastigotes of EGFP-ama cell line were transformed into EA by in vitro amastigogenesis , EGFP signal became apparent as trypomastigotes differentiated into round flagella-less form ( Fig 2A , Extracellular amastigote ) . Proliferating axenic amastigotes in LIT medium continued to express EGFP , and retained the fluorescence at the same level even after 1-week of host-free replication ( Fig 2A , Axenic amastigote ) . These results indicate that in vitro-transformed EA and axenic amastigotes use similar differential gene expression system to that of intracellular amastigote , and that amastigote-specific 3’UTR regulation persists even after 1 week of axenic cultivation . To verify whether the endogenous genes in axenic amastigote are also under stage-specific regulation , the amount of selected mRNAs; amastin , paraflagellar rod protein and TcAc2 , were analyzed by RT-qPCR . Differential expression of the target genes were microarray-identified and qPCR-verified previously[33] and analyzed by RNA-seq more recently[34] by other groups . δ-Amastin is known to be expressed abundantly in amastigote stage[35] . In axenic amastigote , mRNA of δ-amastin remained in similar level to that in intracellular amastigote isolated from host cells throughout 1 week of axenic cultivation , which is around 4–6 fold higher than the level of trypomastigote or epimastigote ( Fig 2B ) . Remarkable upregulation of δ-amastin mRNA during in vitro amastigogenesis is consistent with RNA-seq data , in which peak expression of δ-amastin coincides with the timing of trypomastigote-amastigote transition[34] , and the fact that the surface of EA is already rich in amastigote-specific glycoproteins by the time it finishes differentiation from trypomastigote[7 , 9] . Paraflagellar rod protein , a key component of flagellum , was significantly downregulated as trypomastigote transformed into EA , and remained roughly 10 fold less than that of trypomastigote during axenic cultivation ( Fig 2B ) . This low expression in axenic amastigote was roughly the same level in intracellular amastigote , reflecting their flagellar-less morphology . TcAc2 is a thiol-dependent reductase and is a virulence factor , also known as Tc52[36] . It was significantly upregulated in epimastigote , comparing to trypomastigote , axenic amastigote and intracellular amastigote , as expected from the previous report[33 , 34] . We observed temporal upregulation of TcAc2 during in vitro amastigogenesis , which is presumably due to a stress response of the parasite to acidic environment[36] . In all three target genes examined , expression levels in axenic amastigote was comparable to that in intracellular amastigote , and clearly distinct from trypomastigote and epimastigote , even after 1 week of host-free replication . To further characterize the nature of proliferating axenic amastigotes and to explore its potential usage in drug screening assay , we compared the efficacy of benznidazole and nifurtimox , clinical drugs for Chagas’ disease , against axenic amastigotes and intracellular amastigotes . We employed resazurin redox assay to quantitate the cell viability of axenic culture . For intracellular amastigote assay , host-amastigote co-cultures were fixed and stained by Hoechst to identify infected 3T3 cells . Percent infection was calculated and were normalized to untreated controls to derive the relative inhibition . Intracellular amastigote assay ( Fig 3 , open circle ) provided EC50s of 3 . 20 ( ± 0 . 36 ) μM and 0 . 53 ( ± 0 . 031 ) μM for benznidazole and nifurtimox , respectively , which are comparable to previously reported EC50s[37] . Estimated EC50s of benznidazole and nifurtimox for axenic amastigote ( Fig 3 , closed circle ) were 1 . 32 ( ± 0 . 074 ) μM and 0 . 39 ( ± 0 . 028 ) μM , respectively . Dose response curves of intracellular amastigotes tend to show steeper Hill Slope than axenic amastigotes , both for benznidazole and nifurtimox . Nonetheless , these results indicate that axenic amastigotes and intracellular amastigotes have similar susceptibility to the tested trypanocidal compounds . Taking advantage of the fact that axenic amastigote is fairly robust , we explored the possibility of using EAs for exogenous DNA transfection by standard electroporation method . Nucleofector system was used as electroporation device and reagent , and pTREX-mDsRed-Bsd plasmid was used to visualize transient expression of a fluorescent marker , mDsRed . Expression of mDsRed became visible 1 day after electroporation ( Fig 4A ) , and continued to be detectable at least for the next 6 days . Eight pre-programed pulse settings ( U-06 , U-33 , V-06 , V-33 , X-01 , X-06 , X-14 , and Y-01 ) were tested to determine suitable pulse condition for EA electroporation . Highest transfection efficiency , 7 . 4 ±0 . 8% , was achieved by X-14 program , whereas maximum survival rate was observed with X-01 and X-06 programs ( S2 Table ) . We selected X-14 program as our standard protocol for EA transfection , as it is also suited for epimastigote transfection[38] . Since axenic amastigote culture is sustainable at 37°C for approximately 1 week without major deterioration ( Fig 1 ) , we next subjected the mDsRed-Bsd transfected EAs to blasticidin selection to enrich mDsRed-positive population . After pTREX-mDsRed-Bsd plasmid was electroporated into EAs , the parasites were transferred to LIT medium and cultured at 37°C . Blasticidin S was added 24 h later , and the percentage of mDsRed-expressing amastigotes were monitored for the next 6 days . Without addition of blasticidin , fraction of mDsRed-positive amastigotes gradually decreased after transfection ( Fig 4B , Bsd - ) . On the other hand , percentage of mDsRed-positive amastigotes increased in presence of blasticidin , since many of mDsRed-negative amastigotes died during the selection period ( Fig 4B , Bsd + ) . Effect of drug selection peaked on 5 days after blasticidin addition , or 6 days post electroporation . Further selection beyond 5 days did not seem to benefit the population enrichment . This is primarily due to gradual loss of proliferation capability of axenic amastigote after 1 week of cultivation , as seen in Fig 1A . Next , blasticidin-selected transfectants from above were used to infect 3T3 cells . Axenic amastigotes were allowed to infect host cells for two days , and amastigotes remained outside of the host cells were washed away . The host-parasite co-culture was incubated for additional two days before visualizing and quantitating the prevalence of mDsRed-expressing amastigotes in 3T3 cells . Transfectants were successfully internalized by the host cells and established productive infection , despite the 6-day-long axenic cultivation ( Fig 4C ) . Infection efficiencies of blasticidin-selected ( Bsd + ) and non-selected ( Bsd - ) transfectants were 19 . 6% and 29 . 8% , respectively ( Fig 4D ) . When blasticidin-selected EAs were used for infection , percentage of mDsRed-expressing amastigotes in total intracellular amastigotes was 36 . 0% , whereas that of non-selected amastigotes was only 2 . 0% ( Fig 4D , black bars ) . These proportions are well correlated with the fractions of mDsRed-positive population in initial transfectants applied onto 3T3 cells for infection ( Fig 4B ) . Blasticidin-selected amastigotes differentiated into trypomastigotes and emerged out to culture supernatant 4 days post infection ( S1 and S2 Movies ) . These results suggest that axenic amastigotes can be utilized for electroporation-mediated exogenous gene transfer , and selectable marker is useful to enrich positive transfectants without significantly impairing the ability of amastigotes to infect host cells and to proceed to the next stage of life cycle . Drug target research against T . cruzi entails validation of gene essentiality especially in amastigote stage . Even though CRISPR/Cas9 system offers effective knockout strategy in T . cruzi[27 , 39–44] , it is troublesome to perform this in amastigotes , because intracellular amastigotes are shielded by the host cell and direct access for experimental manipulation is hindered . Utilization of EA as an experimental tool potentially offers an alternative mean to bypass such issue , and allows us to perform knockout studies solely in amastigote stage . To this end , we investigated whether CRISPR/Cas9 system can function in EA to knockout a target gene and yield measurable growth phenotype to allow evaluation of the target essentiality , either as an axenic culture or as intracellular amastigotes followed by host infection . For a proof of concept , we first transfected gRNA against EGFP into the stable Cas9 cell line , which harbors EGFP as a Cas9 fusion protein , to confirm the functionality of the system and to estimate the knockout efficiency ( Fig 5 ) . The fraction of EGFP-positive population dropped to 2% in epimastigote and to 4% in amastigote at 1 day post transfection . We routinely achieved knockout efficiency higher than 95% in both epimastigote and amastigote using conventional electroporation method , based on the fluorescence intensity . Analysis of genomic DNA confirmed that CRISPR/Cas9-mediated knockout introduced mutations in the target DNA . ( S3 Fig ) . We then targeted an endogenous gene , TcCGM1 as a model target . TcCgm1 is T . cruzi homologue of T . brucei mRNA capping enzyme TbCgm1 , which is responsible for cap 0 formation on SL RNA and is essential for the proliferation of T . brucei[28] . We transfected gRNA against TcCGM1 into epimastigote and amastigote stages of T . cruzi . Guanylyltransferase activity of TcCgm1 , along with the other capping enzyme TcCe1 , can be detected in the cell lysate by incubating the total protein with [α-32P]-GTP in presence of metal cofactor . The resulting enzyme-[32P]-GMP covalent intermediate can be visualized as a radiolabeled band in SDS-polyacrylamide gel . After transfection with gRNA against TcCGM1 , signal of TcCgm1-[32P]-GMP became weak comparing to the control cells that received gRNA with unrelated sequence ( Fig 6A and 6B ) . Radiolabeled signal of the other guanylyltransferase , TcCe1 , was relatively unaffected . The amount of TcCgm1-[32P]-GMP was reduced to 32% in epimastigote and to 49% in axenic amastigote 2 days after gRNA transfection . We then monitored the phenotype of TcCGM1 knockout cells after gRNA transfection . In epimastigote , Cas9 cells transfected with TcCGM1-gRNA halted the growth on the day after electroporation ( Fig 6C ) . Deformation of the knockout cells started to appear on day 2 , and became clearly noticeable on day 3 post transfection ( Fig 6D ) . TcCGM1-knockout epimastigotes tended to be large , and often possessed multiple flagella . Nuclear staining revealed that many cells contained abnormal number or size of nuclei or kinetoplasts . Much smaller spots of unknown nature were observed in some cases ( Fig 6D , arrow head ) . In axenic amastigote , growth of Cas9-ama cells transfected with TcCGM1-gRNA was also suppressed ( Fig 6E ) . For the first 3 days , there was no apparent morphological changes in knockout cells , comparing to the amastigotes received control gRNA . However from day 4 , TcCGM1-knockout amastigotes started to display irregular shapes . Deformation became more noticeable on day 5 ( Fig 6F ) . Unlike knockout epimastigotes , not many cells possessed multiple nuclei or kinetoplasts , except occasional large cells that show abnormal Hoechst staining pattern ( Fig 6F , arrow head ) . For intracellular amastigote assay , gRNA-transfected EAs were applied onto 3T3 cells 1 day after electroporation at MOI of 20 , and allowed to infect host cells for 2 days . Amastigotes remained outside of the host cells were washed away , and host-parasite co-culture was incubated for additional 2 days . Infected 3T3 host cells were then fixed and stained by Hoechst to identify intracellular amastigotes ( S4 Fig ) to calculate the percent infection . Fraction of host cells infected by amastigotes transfected with TcCGM1-gRNA and control gRNA were 4 . 6% and 13 . 6% , respectively ( Fig 6G ) . This outcome is in agreement with the transfectants’ cell growth monitored as axenic cultures ( Fig 6E ) . Taken together , these results show that essentiality of a target gene in EA can be analyzed after CRISPR/Cas9-mediated knockout by monitoring growth phenotype of the amastigotes , either as axenic culture or as intracellular amastigotes followed by host invasion .
T . cruzi goes through distinct life cycle stages as they travel between insect vectors and mammalian hosts . Being able to isolate and study individual stage is crucial in understanding the parasite biology . Epimastigote stage of T . cruzi has been routinely used for basic cell biology research and drug development study , because of the easiness of culture maintenance . On the other hand , amastigote had received little attention as a subject of direct experimental manipulation due to complication and inaccessibility in the host co-culture , even though this life cycle stage is most relevant in host-parasite interaction and drug development studies . Here , we demonstrated that EA can be proliferated as axenic culture at least for one week without major morphological change or loss of stage-specific gene expression . Susceptibility of EA and intracellular amastigote to benznidazole and nifurtimox was comparable in terms of EC50 values . We also demonstrated that a plasmid vector can be delivered directly into EA for transient gene expression , and transfectants can infect host 3T3 cells and replicate just like bona fide amastigotes even after 6 days of axenic culturing in presence of selective agent . CRISPR/Cas9 system can function in Cas9-expressing axenic amastigotes when gRNA is transfected by conventional electroporation . These new methodologies open up the possibility to carry out stage-specific experiments in a truly amastigote-specific manner . It is widely accepted that amastigote of T . cruzi is an obligate intracellular parasite . Host metabolic factors such as Coenzyme Q10 and Akt-related pathways including glucose and lipid metabolisms have been implicated as key growth regulators of intracellular amastigote[45 , 46] . Our result indicates that components in LIT medium can compensate for such nutrient needs , at least for a short while , to sustain the growth of axenic amastigote . Although it was previously demonstrated that EA can uptake exogenous glucose[47] , it is unlikely that glucose by itself allows axenic proliferation , because DMEM contains higher concentration of glucose than LIT medium , yet EA kept in DMEM did not replicate at all ( Fig 1A ) . Optimum nutrients required for axenic amastigote cultivation beyond 1 week remain to be investigated . Also , whether the technique of genetic manipulation during temporal axenic cultivation is applicable to other strains of T . cruzi or not must be investigated in future studies . There are few instances in previous literatures that EA increased in number during host-free incubation in Y strain[30] and Brazil strain[10 , 29] , but those observations were not followed up . Recently , several techniques for high-throughput inhibitor screening against T . cruzi host co-culture have been developed ( Reviewed in [37] ) . However , those systems are specialized for phenotypic assays in compound screenings . In order to identify a drug target , to probe into a mode of action of drug candidates , or to investigate the biological role of specific gene , there is still a great need to directly investigate the amastigote itself . Preceding examples of utilization of axenic amastigotes can be found in Leishmania drug screening studies[48–53] . It must be noted , however , that inhibitory compounds identified by axenic assay and host co-culture assay do not perfectly overlap[49 , 52] . This discrepancy originates in part from physiological differences between intracellular and extracellular forms of Leishmania amastigotes , namely proteome[54] and transcriptome[55] . In T . cruzi , some differences between the two forms of amastigotes have also been reported . For example , EA is more resistant to complement-mediated lysis than intracellular amastigote . Hundred percent of intracellular amastigote is lysed in fresh serum but EA is completely resistant in case of Tulahuen strain[56] . Also , EA is more infectious to the cultured host cell than intracellularly-derived amastigotes[56] . Since literature on drug treatment of T . cruzi axenic amastigotes is extremely limited[29] and this present study is the first instance of directly comparing the dose response curves of EA to that of intracellular amastigotes in host cells , it definitely requires further investigations to see whether T . cruzi axenic amastigotes can be utilized for inhibitor screening assays in general . Our data indicate that Hill Slopes tend to be shallower in axenic amastigote comparing to intracellular amastigote ( Fig 3 ) . This might partly be resulted from different counting schemes in our experiments , i . e . , resazurin assay of axenic amastigote reflects redox activity of viable parasites , whereas percent host infection of intracellular amastigote does not account for the population dynamics of the parasite within individual host cell . Alternatively , steepness of the dose response curves may be associated with complexity of the target molecule or pathway , and presence or absence of the host cells could affect “tipping point” of the lethality . Since benznidazole and nifurtimox both affect wide range of cellular machinery by generation of reactive oxidant species , it would be interesting to see whether target-specific trypanocidal compounds also produce similar slope trends in dose response profiles . If significant phenotypic discrepancies are found between EA and intracellular amastigotes in drug sensitivity or selectivity , those inhibitors potentially provide valuable insights into host-parasite interaction and cellular biology of T . cruzi amastigote . Considering the amount of information we can extract , there is no doubt that parasite-host co-culturing is the most relevant system in terms of phenotypic assay[37] . It surely requires further investigation to determine the relevance and practicality of the use of EA in drug screening . Nonetheless , the use of EA can give us an easy and fast evaluation of compound susceptibility of amasitigote itself . In the co-culture system , passing the host cell barrier is one of the top criteria that the compounds are selected for . However , it is not uncommon that subsequent chemical modification significantly improves the permeability of the compound to cell membrane through lead optimization process[57–59] or by drug delivery system[60] . Therefore , the use of naked amastigote in early-stage compound screening may give chance to some candidate compounds that are otherwise dropped out , and expand the options of starting materials to move forward with the next step of drug development . There are some studies utilizing EA in T . cruzi in the past , however their objectives were mostly limited to investigation of signaling factors involved in trypomastigote-to-amastigote differentiation[30 , 61] or host invasion[16–20] . To our knowledge , the present study is the first instance of utilizing T . cruzi EA for direct transfection for exogenous gene expression and endogenous gene knockout . Previously , Padmanabhan et al . demonstrated that trypomastigote can be transfected for later infection and differentiation to produce transgene-expressing intracellular amastigotes[6] . In their report and also in our hands , plasmid transfection efficiency of trypomastigote is about 5% . One advantage of using EA instead of trypomastigote is that electroporation can be followed by proliferation and selection of transfectants to enrich positive population to compensate for low transfection efficiency . In our present study , fraction of mDsRed-positive EA was initially about 4% , but reached to 37% after 5 days of blasticidin selection ( Fig 4B ) . It may be possible to improve the enrichment efficiency by using a selection marker that requires shorter selection period , or by improving the culture medium to allow longer cultivation of axenic amastigote . It is , of course , feasible to take advantage of a fluorescence activated cell sorter prior to the host infection[6] to obtain homogeneous population of transgenic amastigotes instead . Another advantage of EA transfection is that we can bypass active host invasion ( as supposed to “passive” mode of infection by amastigotes ) and trypomastigote-to-amastigote differentiation that may introduce unwanted bias to transfectants , which is crucial when studying stage-specific cellular functions . For example , in a drug target research , one would like to produce knockout parasites in search for a target gene that is essential in clinically-relevant amastigote stage . However , if a candidate gene was essential in trypomastigote stage , we cannot obtain a knockout amastigote by infection and differentiation of the lethal trypomastigote . Our strategy of using axenic amastigotes enables evaluation of essential genes truly in amastigote-specific manner . Knockout study of some target genes may yield different outcomes between axenic amastigotes and host intracellular amastigotes . If so , those cases provide us with opportunities to examine the involvement of host factors in amastigote-specific gene functions . In summary , having direct access to amastigote as experimental tools may greatly expand methodological freedom to investigate basic cellular biology of T . cruzi and potentially provide valuable insights into the drug development study in the future . | We developed an experimental system to study amastigote stage of Trypanosoma cruzi as a proliferable axenic culture . Use of axenic amastigotes allows us to directly introduce exogenous gene into T . cruzi amastigote and select for drug resistant parasite to enrich the transfectants . Our strategy bypasses differentiation steps involved in conventional epimastigote transfection procedure to obtain transgenic amastigotes . Gene knockout can also be performed in amastigote-specific manner , using Cas9-expressing extracellular amastigotes . Drug sensitivity could also be assessed during 1-week axenic growth period . Our method potentially leads to variety of new experimental strategies to make amastigote-stage-specific manipulations and analyses possible . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"transfection",
"viral",
"transmission",
"and",
"infection",
"microbiology",
"parasitic",
"protozoans",
"protozoan",
"life",
"cycles",
"developmental",
"biology",
"trypomastigotes",
"protozoans",
"molecular",
"biology",
"techniques",
"epimastigotes",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"life",
"cycles",
"guide",
"rna",
"molecular",
"biology",
"amastigotes",
"trypanosoma",
"cruzi",
"mechanical",
"treatment",
"of",
"specimens",
"biochemistry",
"rna",
"trypanosoma",
"eukaryota",
"host",
"cells",
"specimen",
"disruption",
"nucleic",
"acids",
"virology",
"biology",
"and",
"life",
"sciences",
"protozoology",
"electroporation",
"organisms"
] | 2019 | Utilization of proliferable extracellular amastigotes for transient gene expression, drug sensitivity assay, and CRISPR/Cas9-mediated gene knockout in Trypanosoma cruzi |
Linked to extreme rates of chronic heart and kidney disease , pyoderma is endemic amongst Aboriginal children in Australia's Northern Territory ( NT ) . Many of those with pyoderma will also have scabies . We report the results of a community-based collaboration within the East Arnhem Region , which aimed to reduce the prevalence of both skin infections in Aboriginal children . Commencing September 2004 , we conducted an ecological study that included active surveillance for skin infections amongst children aged <15 years in five remote East Arnhem communities over a three year period . Screening was undertaken by trained local community workers , usually accompanied by another project team member , using a standard data collection form . Skin infections were diagnosed clinically with the aid of a pictorial flip chart developed for the purpose . Topical 5% permethrin was provided for age-eligible children and all household contacts whenever scabies was diagnosed , whilst those with pyoderma were referred to the clinic for treatment in accordance with current guidelines . In addition , annual mass scabies treatment ( 5% permethrin cream ) was offered to all community residents in accordance with current guidelines but was not directly observed . Pyoderma and scabies prevalence per month was determined from 6038 skin assessments conducted on 2329 children . Pyoderma prevalence dropped from 46 . 7% at baseline to a median of 32 . 4% ( IQR 28 . 9%–41 . 0% ) during the follow-up period – an absolute reduction of 14 . 7% ( IQR 4 . 7%–16 . 8% ) . Compared to the first 18 months of observation , there was an absolute reduction in pyoderma prevalence of 18 cases per 100 children ( 95%CI −21 . 0 , −16 . 1 , p≤0 . 001 ) over the last 18 months . Treatment uptake increased over the same period ( absolute difference 13 . 4% , 95%CI 3 . 3 , 23 . 6 ) . While scabies prevalence was unchanged , the prevalence of infected scabies ( that is with superimposed pyoderma ) decreased from 3 . 7% ( 95%CI 2 . 4 , 4 . 9 ) to 1 . 5% ( 95%CI 0 . 7 , 2 . 2 ) , a relative reduction of 59% . Although pyoderma prevalence remained unacceptably high , there was a substantial reduction overall with improvements in treatment uptake a critical factor . More acceptable alternatives , such as cotrimoxazole for pyoderma and ivermectin as a community-wide scabicide , warrant further investigation in these settings . We are encouraged by progress made through this work , where local action was led by local community members and primary health care providers with external training and support . ClinicalTrials . gov NCT00884728
Pyoderma is a generic term used to describe a clinical diagnosis of superficial bacterial skin infection [1] . Also known as skin sores or impetigo , it has been estimated that there are in excess of 111 million children with pyoderma worldwide and that many of these children will also have scabies [1] . Reported pyoderma prevalence has varied , but children living in Australian Aboriginal communities and those living in the Pacific region have generally had the highest burden , often in the range of 40–90% [2] . A recent study from Fiji [3] , reported pyoderma prevalence of 25% amongst primary school children and 12% amongst infants . Like previous studies in remote Australian Aboriginal communities [4] , [5] , the Fijian study found that children with scabies were more likely to have pyoderma and that most pyoderma was due to Group A streptococcal ( GAS ) infection . Multiple GAS strains can be present at the same time in remote Australian Aboriginal communities , with household acquisition rates as high as 1 in 5 [6] . Whereas one or two sores would raise alarm in other settings , the burden amongst Aboriginal children is such that current recommendations for the Northern Territory ( NT ) specify “six infected sores” as one of the benchmarks for pyoderma treatment [7] . Controlling pyoderma and scabies could be an important primary health intervention to reduce serious bacterial infections in childhood and may have potential longer term sequelae , particularly in relation to acute rheumatic fever ( ARF ) , rheumatic heart disease ( RHD ) and chronic kidney disease . Pyoderma is a major underlying cause of serious bacterial infections ( GAS and Staphylococcus aureus ) , which are very common in Aboriginal communities and have high mortality [8] , [9] . The burden of ARF/RHD is extremely high amongst Indigenous Australians [10] . Since GAS throat infection is rare in this setting , it seems probable that pyoderma ( and scabies ) may have a key role in the ARF/RHD burden , although the causal pathway remains uncertain [9] , [11] , [12] . Over the past decade: the Indigenous incidence of end stage renal disease has been shown to be 21 times that of non-Indigenous Australians [13]; the risk of acute post streptococcal glomerulonephritis ( APSGN ) five times higher for children with pyoderma during APSGN outbreaks [14] , [15]; and APSGN in childhood has been shown to substantially increase the risk of adult renal disease [16] . Mortality from kidney failure is also much higher than non-Indigenous Australians [17] . Menzies School of Health Research and others have collaborated with NT communities and the government authorities over more than 10 years to develop strategies for tackling skin infections at a community level . The initial work , which documented the link with severe and chronic diseases , included laboratory based studies to demonstrate that dog and human scabies were not linked [18] and that the diverse range of streptococcal strains – including those associated with rheumatic fever and kidney disease – are all derived from skin infections [19] . A successful community based scabies control program , involving annual mass treatment days for scabies ( using topical 5% permethrin cream ) and active surveillance for skin infections , was conducted in one of the largest remote communities [20] . A feasibility study subsequently demonstrated widespread support for similar programs throughout NT Indigenous communities noting those communities that had successfully implemented their own programs were keen to participate in coordinated programs with related communities [21] . We report the results of a regional collaboration involving a number of these communities in the East Arnhem Region of the NT where the primary aim was to reduce the prevalence of scabies and pyoderma amongst children aged 0–14 years .
The East Arnhem Healthy Skin Project ( EAHSP ) involved five remote communities and associated homelands/outstations . The total population of each community ranged from approximately 800–2000 people , encompassing a total of approximately 2600 Indigenous children aged <15 years . Active surveillance for skin infections was conducted over a three year period from September 2004–August 2007 . Children aged <15 years were screened for skin infections at either the local health centre , home , in the community or at school using a standard data collection form . We collected additional data for a subset of children seen during school screening , which included: appearance of pyoderma ( crusted , purulent or flat/dry ) , site of pyoderma ( upper body or lower body ) , the number of sores ( <5 , 5–20 or >20 ) and , for those with scabies , whether or not the infestation had become infected ( scabies with a superficial bacterial skin infection = pyoderma [1] ) . Screening was undertaken by trained local community workers , usually accompanied by another project team member ( Aboriginal Health Worker , registered nurse , or visiting pediatrician/dermatologist ) . A formal training program was established for community workers involving both practical “on-the-job” training and formal learning packages delivered as “off-the-job” components . Eleven workers completed the training program and gained recognition against units of competency towards a Primary Health Care qualification . The community workers , all local Aboriginal people , were employed for up to 20 hours per week to undertake skin screening of children and were supported every 1–2 months by a visiting research team member . Annual healthy skin days were held at each community , commencing from September 2004 . Coordinated by the local community , these events included promotion of mass treatment with scabicide cream ( distributed in 30g tubes with 5% permethrin , Lyclear ® ) for all community members in accordance with current guidelines [22] . The cream was hand delivered to each household with adequate quantities supplied for all household members and verbal advice provided on appropriate use of the cream . Treatment application was not directly observed as this was not considered culturally appropriate . The project was approved by the Human Research Ethics Committee of the Northern Territory Department of Health Community Services and Menzies School of Health Research ( Approval number 04/11 ) with written informed consent obtained for participants . We developed two pictorial flipcharts , one to explain the Healthy Skin Story in lay terms for study participants which included family-based and environmental activities [23] , the other to assist in diagnosis focusing primarily on recognition and treatment of pyoderma , scabies and tinea [24] . Both flip charts were used by all study staff and distributed to each participating community . Skin infections were diagnosed clinically by the research team . Pyoderma lesions were not routinely swabbed nor were scabies mites routinely extracted . Children with pyoderma were referred to the clinic for treatment in accordance with the current guidelines [7] . If children screened by the research team were identified with scabies , the parent/carer was given topical 5% permethrin ( Lyclear ® ) to administer to the child and all household contacts . For infants younger than 2 months of age topical Crotamiton 10% cream was used . Given previous experience of Healthy Skin programs [4] , [5] , [20] , we expected to observe a reduction in both scabies and pyoderma prevalence immediately following the initial mass scabicide treatment . Our primary objective was to demonstrate reductions in both conditions among children aged 0–14 years in the participating communities within two years following the introduction of the program: A secondary objective was to reduce the severity of pyoderma among the target group that was classified as moderate/severe from 40% ( expected pre-program ) to <15% . We determined the prevalence of pyoderma and scabies at baseline as the proportion of all children seen in September 2004 ( immediately preceding the initial mass community-wide scabicide treatment ) . Thereafter , we determined the monthly period prevalence of both pyoderma and scabies on the same basis , calculated as a rolling average of all children seen over the preceding three-month period ( ie commencing from December 2004 through to August 2007 ) . Although some children were seen more frequently , we excluded subsequent assessments that occurred within 30 days of the initial assessment in order to ensure that only one assessment per individual per month was included . These were population level surveillance data ( non-random sample ) where the intention was to screen as many children as possible within the region given the available resources . We monitored variation in the number of children seen over the surveillance period by determining the median number , and interquartile range ( IQR ) , of children seen over each of the rolling three-month periods . In addition , we were cognizant of the potential that eagerness to demonstrate a benefit from a Healthy Skin program might preferentially encourage screening of healthier children or that awareness of the program might conversely result in an increase in care-seeking behaviour for those children with skin infections . We monitored this potential by conducting an exploratory sub-analysis of pyoderma and scabies prevalence on the first occasion that an individual subject was seen . We examined seasonal differences by comparing the average monthly prevalence during the wet season ( September–February ) and the dry season ( March–August ) . We also monitored changes over time by comparing the average monthly prevalence during the first 18 months of the study period to that for the last 18 months using the 2 sample test of proportions as an exploratory analysis . Data were analysed using Stata version 9 software .
There were 6038 skin assessments conducted on 2329 children during the three year study period , estimated to represent 78% of the target population ( Figure 1 ) . Most children ( 1245 , 53 . 5% ) were seen once or twice , 19% three times and 28% were seen from four to nine times . There were 583 children screened during the initial assessment in September 2004 with a similar number ( median 487 , IQR 405–578 ) screened over the subsequent rolling three-month periods to August 2007 ( Table 1 ) . The median number of children seen for the first time during the post baseline period was 132 ( IQR 92–211 ) , that is approximately 1 in 4 children seen during the rolling 3 month periods after September 2004 had not previously been seen . Overall , there were 2001 assessments performed by study team members at school ( 33% ) and 4037 ( 67% ) elsewhere within the community ( eg home or clinic ) . By August 2004 , there had been 2139 episodes of pyoderma diagnosed and 807 episodes of scabies . Of the 2329 children seen , almost all ( 91 . 8% , 95%CI 87 . 7 , 96 . 0 ) had pyoderma on at least one occasion , while 34 . 7% ( 95%CI 31 . 9 , 37 . 4 ) had scabies on at least one occasion . The average monthly prevalence was 35 . 5% for pyoderma and 13 . 4% for scabies ( Table 1 ) . Pyoderma prevalence was slightly higher amongst males , 38 . 8% ( 95%CI:37 . 1 , 40 . 5 ) , than females , 32 . 1% ( 95%CI:30 . 4 , 33 . 8 ) – an absolute difference of 6 . 7% ( 95%CI:4 . 3 , 9 . 1 ) . There was no evidence of a seasonal difference in pyoderma prevalence: wet seasons = 35 . 3% ( 95%CI:33 . 5 , 37 . 0 ) ; dry seasons = 35 . 8% ( 95%CI:34 . 1 , 37 . 5 ) . Pyoderma prevalence was similar between the three age groups . In contrast , scabies prevalence was highest amongst children aged <3 years and decreased with age ( Table 1 ) . Scabies prevalence amongst children aged <3 years ( 22 . 7% , 95% CI 20 . 3 , 25 . 1 ) was double that of children aged 3–14 years ( 11 . 1% , 95% CI 10 . 2 , 12 . 0 ) . Whereas pyoderma prevalence at baseline ( 46 . 7% ) was close to expectations , scabies prevalence at baseline ( 16 . 1% ) was substantially lower than we expected ( 30% ) . Thereafter , we observed a reduction in pyoderma prevalence but little if any change in scabies prevalence . During the post baseline period , the median monthly prevalence of pyoderma was 32 . 4% ( IQR 28 . 9–41 . 0% ) – an absolute reduction of 14 . 7% ( IQR 4 . 7–16 . 8% ) . A similar reduction was evident when prevalence was limited to the first time an individual child was seen – absolute reduction of 11 . 2% ( IQR 2 . 3–16 . 4% ) . In contrast , the median monthly prevalence of scabies remained similar at 13 . 1% ( IQR 11 . 7–15 . 9 ) – a marginal absolute reduction of 3 . 0% ( IQR 0 . 2–4 . 4% ) , with no discernible difference in the median monthly prevalence when limited to children seen for the first time during the post baseline period ( Table 1 ) . Both pyoderma and scabies prevalence varied over time ( Figure 2 ) but these did not closely correspond to annual mass scabicide treatment , which occurred in September 2004 , September 2005 and September–October 2006 . Whereas scabies prevalence fluctuated close to the baseline rate over the three year study period , pyoderma prevalence was notably lower over the latter 18 month period ( Figure 2 ) . For scabies , the pattern was similar when data were limited to prevalence determined amongst those children who were seen for the first time ( Figure 3 ) . In contrast , pyoderma prevalence amongst this subgroup of children actually increased towards the end of the observation period suggesting that these children were more likely to be diagnosed with pyoderma but were not more likely to be diagnosed with scabies . The reduction in pyoderma prevalence over the latter 18 month period was evident across all age groups ( Figure 4 ) , whereas there was no evidence of a comparable reduction in scabies prevalence . During the latter 18 month period , there were 18 fewer pyoderma cases for every 100 children seen ( 95%CI:−21 . 0 , −16 . 1 , p<0 . 001 ) , a 40% reduction when compared to rates during the first 18 month period . The greatest absolute reduction was amongst children aged 3–14 years: 20 fewer cases of pyoderma for every 100 children seen ( −20 . 3% , 95%CI:−23 . 0 , −17 . 6 ) . We collected additional data for children seen during school screening ( n = 2001 aged 3–14 years ) . For this subset of children , pyoderma prevalence was 40 . 2% ( 95%CI 38 . 1 , 42 . 4 ) , slightly higher than that of children in the same age group who were seen outside the school setting ( 33 . 2% , 95%CI 31 . 8 , 34 . 7 ) . For children seen during school screening , pyoderma was most commonly diagnosed on the legs ( Table 2 ) where 36 . 5% of children had lower body pyoderma compared with 17 . 6% upper body pyoderma . Multiple sores ( five or more ) were also more common on the lower body ( 14 . 2% ) than the upper body ( 4 . 5% ) . Of those school children with pyoderma , 88 . 1% ( 95%CI: 85 . 7 , 90 . 4 ) had sores on the lower body while 42 . 4% ( 95%CI:38 . 9 , 45 . 9 ) had sores on the upper body . The lower body was the region where the reduction in prevalence was most evident: 13 fewer cases of lower body pyoderma per 100 children seen and 6 fewer cases of children with multiple lower body sores ( Table 2 ) . There was also a reduction in the nature of the sores , with 10 fewer cases of crusted/purulent pyoderma per 100 children seen during the latter 18 month period . From the additional data collected , we identified 144 children who had been diagnosed with five or more pyoderma lesions . Of these , only 50 ( 34 . 7% ) were recommended for treatment – the vast bulk , 94 ( 65 . 3% ) had no treatment follow-up . We reviewed clinic records for those who had been referred for treatment . While treatment uptake remained low , there was a substantial improvement during the latter 18 month period when 17 . 2% of children with five or more sores received treatment , single dose benzathine penicillin , compared to 3 . 8% during the initial 18 month period ( Table 2 ) . For those with five or more sores , this was an absolute increase in treatment uptake equivalent to 13 children per 100 seen ( 95%CI 3 . 3 , 23 . 6 ) . By comparison , although there was an overall reduction in crusted or purulent pyoderma ( 10% ) there was not a demonstrable increase in treatment uptake for these children ( 3 . 5% , 95%CI −0 . 7 , 7 . 6 ) . While there was no reduction in scabies prevalence over the latter 18 month study period ( Figure 4 ) , there was a significant reduction in prevalence of infected scabies amongst children aged 3–14 years , falling from 3 . 7% ( 95% CI 2 . 4 , 4 . 9 ) of children seen in the first 18 months to 1 . 5% ( 95% CI 0 . 7 , 2 . 2 ) in the last 18 months ( Table 2 ) . Comparable data were not routinely collected over the entire study period for children aged less than 3 years .
Sub standard living conditions are common factors reported in many communities where skin infections are endemic [33] , [34] . Redressing inadequate living conditions , poverty and other social determinants of health are major priorities for long term health benefits for Indigenous Australians . For skin infections , a reduction in pyoderma prevalence is the primary priority and the most encouraging outcome from our study . In our view , increased treatment uptake is the most likely explanation but we believe it has been local action delivered through basic primary health services that has been the driver . Surveillance for skin infections , supported by appropriate treatment where indicated , can reduce pyoderma prevalence . Local community workers played a key role in the surveillance component but more work is needed in improving treatment uptake , both in terms of the threshold for treatment referral and perhaps also in regards to a more effective alternative to benzathine penicillin . We are currently investigating the role of cotrimoxazole in this regard . Although we found no discernible impact against scabies , we know that reducing scabies prevalence will further reduce pyoderma prevalence . The potential for oral ivermectin as a more acceptable alternative to community-wide use of scabicide cream , particularly for adult household contacts , needs to be investigated amongst Indigenous populations where scabies continues to be endemic . Like most people living in remote Indigenous communities , as well as those working in primary health care , we acknowledge that much still needs to be done . However , on balance , we believe the results presented here are a good news story for local action to address a serious public health problem . | Skin infections are endemic in many in remote Australian Aboriginal communities and have been linked to very high rates of chronic heart and kidney disease in this population . We report the results of a regional collaboration that aimed to reduce skin infections amongst children aged less than 15 years in five remote communities . The program included annual mass scabies treatment days offered to all residents and routine screening/follow-up of children . Trained community workers helped conduct over 6000 skin assessments on 2329 children over a three year period . Of every 100 children seen at the commencement of the study , 47 were found to have skin sores and many had multiple sores . We demonstrate a reduction both in the number of children with skin sores and in the severity of those sores . On average , of every 100 children seen per month , there were 14 fewer children with skin sores and seven fewer children with multiple sores . Overall improvement in treatment uptake was a critical factor . We found no discernible impact against scabies . While the burden of skin infections remains unacceptably high , we believe the results presented here are a good news story for local action to address a serious public health problem . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"dermatology/pediatric",
"skin",
"diseases,",
"including",
"genetic",
"diseases",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"dermatology/skin",
"infections",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"infectious",
"diseases/skin",
"infections",
"public",
"health",
"and",
"epidemiology/screening",
"infectious",
"diseases/bacterial",
"infections",
"pediatrics",
"and",
"child",
"health/pediatric",
"dermatology",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2009 | A Regional Initiative to Reduce Skin Infections amongst Aboriginal Children Living in Remote Communities of the Northern Territory, Australia |
Parasitic protozoa such as the apicomplexan Toxoplasma gondii progress through their life cycle in response to stimuli in the environment or host organism . Very little is known about how proliferating tachyzoites reprogram their expressed genome in response to stresses that prompt development into latent bradyzoite cysts . We have previously linked histone acetylation with the expression of stage-specific genes , but the factors involved remain to be determined . We sought to determine if GCN5 , which operates as a transcriptional co-activator by virtue of its histone acetyltransferase ( HAT ) activity , contributed to stress-induced changes in gene expression in Toxoplasma . In contrast to other lower eukaryotes , Toxoplasma has duplicated its GCN5 lysine acetyltransferase ( KAT ) . Disruption of the gene encoding for TgGCN5-A in type I RH strain did not produce a severe phenotype under normal culture conditions , but here we show that the TgGCN5-A null mutant is deficient in responding to alkaline pH , a common stress used to induce bradyzoite differentiation in vitro . We performed a genome-wide analysis of the Toxoplasma transcriptional response to alkaline pH stress , finding that parasites deleted for TgGCN5-A fail to up-regulate 74% of the stress response genes that are induced 2-fold or more in wild-type . Using chromatin immunoprecipitation , we verify an enrichment of TgGCN5-A at the upstream regions of genes activated by alkaline pH exposure . The TgGCN5-A knockout is also incapable of up-regulating key marker genes expressed during development of the latent cyst form , and is impaired in its ability to recover from alkaline stress . Complementation of the TgGCN5-A knockout restores the expression of these stress-induced genes and reverses the stress recovery defect . These results establish TgGCN5-A as a major contributor to the alkaline stress response in RH strain Toxoplasma .
Stress responses are critical to cell survival , allowing cells to adapt to changing environmental conditions . In certain pathogenic eukaryotes , such as the protozoan Toxoplasma gondii ( phylum Apicomplexa ) , the stress response takes on added significance as it triggers a developmental change into a latent cyst form . Parasitic protozoa often rely on stimuli in the environment or host organism in order to progress through the parasite life cycle . The study of stress-induced developmental changes in Toxoplasma is significant as this process underlies pathogenesis . This obligate intracellular protist develops from a rapidly growing form ( tachyzoite ) into a latent cyst form ( bradyzoite ) in response to stress [1] . In human hosts , the cyst forms can re-emerge as destructive tachyzoites if immunity wanes , causing recurring bouts of toxoplasmosis that may endanger immunocompromised individuals [2] . A major gap in our knowledge that impedes the development of novel therapeutics against Toxoplasma infection is our poor understanding of how tachyzoites reprogram their expressed genome in response to stresses that prompt cyst development . The identification of proteins that contribute to stress response and bradyzoite formation would be a significant step towards the design of new therapies to treat toxoplasmosis . How the parasite coordinates the changes in gene expression that accompany stress-induced bradyzoite development is not clear , but epigenetic mechanisms , including histone modifications , have been implicated as contributing to this process [3] , [4] . Formerly referred to as histone acetyltransferases ( HATs ) , lysine acetyltransferases ( KATs ) of the general control nonderepressible-5 ( GCN5/KAT2 ) family are well-conserved among eukaryotes [5] . While invertebrates generally possess a single GCN5 , vertebrate species harbor two: GCN5 and the highly similar homologue called PCAF ( p300/CBP-Associating Factor ) [6] . The GCN5 KAT family has been implicated in cell-cycle progression [7] , chromatin remodeling at specific promoters [8] , transcription elongation [9] , cellular differentiation [10] , and the cellular stress response [11] . Microarray analyses of knockouts made in yeast suggest that GCN5 is a gene-specific coactivator , regulating 1 . 1% of genes in Schizosaccharomyces pombe and up to 4% in Saccharomyces cerevisiae [12] , [13] . The GCN5 deletion mutant in S . cerevisiae is viable , but grows poorly on minimal media [14] . Similarly , GCN5 is not essential for growth under normal conditions in S . pombe , but is required for mounting an appropriate response to KCl and CaCl2-mediated stresses [15] , [16] . In Arabidopsis plants , GCN5 controls ∼5% of genes and is important for normal growth and development , as well as the response to cold stress [17] . GCN5 was shown to be instrumental in the control of specific morphogenetic cascades during developmental transitions in Drosophila [18] . GCN5-null mouse embryos fail to form dorsal mesoderm lineages due to a high incidence of apoptosis and die 10 . 5 days post conception , suggesting a critical role for GCN5 in mammalian development [10] , [19] , [20] . In contrast , PCAF appears to be dispensable in mice as its loss confers no distinct phenotype [10] . Collectively , these studies support the idea that GCN5 KATs modulate gene expression during stress , or exposure to other environmental stimuli , to elicit the appropriate response or developmental change . Toxoplasma is unique among fellow apicomplexan parasites and other invertebrates in possessing two GCN5 HATs , designated TgGCN5-A and –B [21] , [22] . We sought to delineate the function of TgGCN5-A by creating a genetic knockout using homologous recombination in haploid RH strain tachyzoites . As in other lower eukaryotes , Toxoplasma lacking TgGCN5-A ( ΔGCN5-A ) showed no discernible phenotype under normal culture conditions [21] , but its response to stress was not addressed . Here , we analyzed wild-type and ΔGCN5-A parasites under normal and alkaline pH growth conditions using Toxoplasma genome microarrays . The results illuminate the parasite's response to alkaline pH and demonstrate that TgGCN5-A is required for most of these changes in gene expression . We also show that ΔGCN5-A parasites exhibit greater sensitivity to alkaline pH stress - a novel role for GCN5 KATs . Moreover , Toxoplasma lacking TgGCN5-A cannot activate bradyzoite-specific genes that are normally up-regulated during alkaline pH-induced cyst development . These studies establish a novel function for GCN5 KATs in the eukaryotic response to alkaline stress and support the idea that TgGCN5-A is a key contributor to gene expression pertinent to the development of the latent cyst form of Toxoplasma .
We have previously created a null mutation of the TgGCN5-A gene by replacing the majority of the genomic locus with a hypoxanthine-xanthine-guanine phosphoribosyltransferase ( HXGPRT ) minigene in type I RH strain parasites lacking HXGPRT , designated ΔGCN5-A [21] . The loss of TgGCN5-A had no overt effect on tachyzoites grown in standard culture conditions [21] , mirroring phenotypes reported for other lower eukaryotes such as yeast [14] . In other species , GCN5 has been shown to be important for stress responses and developmental changes [23] . In Toxoplasma , stress can lead to expression of bradyzoite-specific genes and the eventual formation of tissue cysts . We predicted that if TgGCN5-A played a role in stress-induced bradyzoite gene expression , then it may be up-regulated in response to a stress agent . After 3 days in alkaline culture conditions ( pH 8 . 2 ) , message levels for TgGCN5-A increase >5-fold in wild-type RH Toxoplasma ( Fig . 1 ) . Actin was monitored as a control to show that alkaline pH stress does not globally increase gene expression ( Fig . 1 ) . To further assess if TgGCN5-A played a role in the stress response in Toxoplasma tachyzoites , we performed whole genome expression profiling to compare ΔGCN5-A and wild-type parasites grown for 3 days either in alkaline pH ( 8 . 2 ) medium or control medium . Gene profiling of certain Toxoplasma strains under stress has been reported previously , but the effects of alkaline pH stress on RH strain has not been examined [24] . Affymetrix ToxoGeneChip microarrays , which contain probe sets for ∼8000 predicted Toxoplasma genes , were used for this study . The differentially regulated genes were grouped into functional categories based on Gene Ontology ( GO ) annotations in the Toxoplasma genome database at ToxoDB . org . The entire dataset of Toxoplasma genes affected by alkaline pH stress is available as supplemental data ( Dataset S1 ) . TgGCN5-A was present in wild-type and absent in ΔGCN5-A , validating the identity of the samples and supporting the fidelity of the microarray analysis . The fidelity of select microarray results was further confirmed through independent qPCR ( Table S1 ) . Results show that ∼14% ( 1 , 114 ) of genes are differentially regulated ( p value<0 . 001 ) in intracellular wild-type parasites after 3 days of alkaline pH exposure . Similar to findings in S . cerevisiae [25] , a broad range of genes have altered expression in Toxoplasma during alkaline pH exposure , including genes involved in invasion , metabolism , protein processing , signaling and gene expression , and membrane transport . Of the 1 , 114 genes affected , 592 genes were up-regulated ( Fig . 2A ) and 522 genes were down-regulated ( Fig . 2B ) . Among the genes with largest changes in response to alkaline stress ( an arbitrary cut-off of 2-fold or more ) , 177 were up-regulated ( Table S2 ) and 84 were down-regulated ( Table S3 ) . A scatter plot of the microarray data reveals that there are virtually no differences ( R2 = 0 . 99 ) in gene expression patterns between wild-type and ΔGCN5-A parasites grown under normal culture conditions ( Fig . 3A ) , which is consistent with the indiscernible phenotype of ΔGCN5-A cultured in normal conditions . However , relative to wild-type , the ΔGCN5-A parasites differ dramatically in their ability to regulate gene expression when grown in alkaline medium ( R2 = 0 . 43 ) ( Fig . 3B ) . While 1 , 114 genes are altered in wild-type parasites exposed to alkaline stress , only 502 genes were changed in alkaline-stressed parasites lacking TgGCN5-A ( p<0 . 001 ) . Since TgGCN5-A is an activator of gene expression we focused on the genes up-regulated during alkaline stress . Further examination of up-regulated genes reveals that ΔGCN5-A parasites fail to up-regulate 439 of the 592 genes up-regulated by wild-type parasites grown in alkaline pH . In other words , TgGCN5-A is required for increased expression of 74% of the genes activated in response to alkaline pH . TgGCN5-A-dependent genes have diverse roles in signaling and gene expression ( 18% ) , protein processing and translation ( 17% ) , metabolism ( 13% ) , membrane transport ( 8% ) , and adhesion ( 7% ) ( Fig . 4 ) . TgGCN5-A also controls a substantive number of hypothetical proteins with no known function ( 37% ) . Hypothetical genes with a change of 2-fold or more are listed in Table S5 . Our microarray data ( p<0 . 05 ) identifies numerous genes related to Ca2+ signaling that were not up-regulated normally in parasites lacking TgGCN5-A . These include calcium-dependent protein kinase ( 86 . m00003 ) , calmodulin beta ( 42 . m03474 ) , calmodulin genes ( 50 . m03141 , 59 . m03587 ) , and calcium/calmodulin-dependent 3′ , 5′-cyclic nucleotide phosphodiesterases ( 583 . m05366 , 59 . m03644 ) ( Dataset S1 ) . Although the eukaryotic response to alkaline exposure is poorly understood , transient increases in intracellular calcium occur , possibly activating calcineurin and leading to a signaling cascade that results in mobilization of transcription factors [25] , [31] . Previously we have demonstrated that increased acetylation accompanies TgGCN5-A promoter occupancy [3] . To obtain in vivo confirmation that TgGCN5-A plays a direct role in the co-activation of genes shown to be up-regulated during alkaline culture , we used chromatin immunoprecipitation ( ChIP ) . RH parasites expressing FLAG-tagged TgGCN5-A ( fGCN5-A ) were employed in ChIP experiments to purify DNA in association with fGCN5-A using anti-FLAG [3] , [32] . We examined a region ∼1 . 0 kb upstream of the start ATG for phosphatidylinositol 3- and 4-kinase ( PI3-4K , 76 . m01548 ) and protein kinase ( PK , 641 . m01507 ) genes , both of which are up-regulated during pH stress ( p<0 . 05 , Dataset S1 ) . Two housekeeping genes , actin ( 25 . m00007 ) and glyceraldehyde-3-phosphate dehydrogenase ( GAPDH , 80 . m00003 ) , whose expression was not altered during pH stress , were included as controls . ChIP data show an enrichment of fGCN5-A at a region upstream of PI3K and PK during alkaline stress ( Fig . 5A ) . The levels of fGCN5-A remained unchanged at housekeeping genes , demonstrating that the increase of fGCN5-A at pH-responsive genes above is not random ( Fig . 5B ) . We performed an additional negative control for each ChIP sample using a nonspecific antibody ( anti-TgIF2K-A ) as described previously [33] , none of which produced a signal ( data not shown ) . Combined with the microarray analysis , these data support the idea that TgGCN5-A is required for proper gene activation in response to alkaline stress . The impaired response to alkaline pH stress suggests that ΔGCN5-A parasites may also be defective in bradyzoite development . The knockout was made in the hypervirulent type I RH strain , which does not fully develop into bradyzoites at high frequency . RH strain will generally not form cyst walls , but will express detectable levels of bradyzoite-specific marker genes in response to stresses that induce cyst formation [34] . In our hands , we can detect BAG1 and LDH2 mRNAs by day 4 of alkaline pH treatment . To test if TgGCN5-A plays a role in stress-induced bradyzoite gene expression , we grew wild-type or ΔGCN5-A parasites in pH 7 . 0 ( control ) or pH 8 . 2 ( stress ) media . At day 4 , intracellular parasites were harvested for quantitative real-time PCR to monitor mRNA levels for bradyzoite-specific genes BAG1 and LDH2 . While wild-type parasites up-regulate both bradyzoite marker genes in response to pH 8 . 2 , the ΔGCN5-A parasites fail to do so ( Fig . 6A and 6B ) . Actin was monitored as a control gene that does not significantly change during bradyzoite induction ( Fig . 6C ) . To ensure the defect was not an indirect effect , we complemented ΔGCN5-A parasites by stably expressing a recombinant copy of TgGCN5-A . Expression of BAG1 and LDH2 was restored in the complemented ΔGCN5-A parasites exposed to alkaline stress ( Fig . 6A and 6B ) . The mRNA levels of BAG1 and LDH2 following stress in complemented ΔGCN5-A parasites are higher than wild-type , presumably because the recombinant TgGCN5-A is being expressed above wild-type levels . Interestingly , this higher level of TgGCN5-A expression in the complemented clone does not affect levels of BAG1 or LDH2 under non-stressed conditions , implying that TgGCN5-A does not activate these bradyzoite genes until a stress signal is perceived by the parasite . We used ChIP analysis to further demonstrate the involvement of TgGCN5-A in regulating developmentally expressed genes . ChIP demonstrates that TgGCN5-A is recruited to BAG1 and LDH2 promoter regions during alkaline stress ( Fig . 6D ) . We conclude that parasites lacking TgGCN5-A are defective in up-regulating bradyzoite-specific genes in response to alkaline stress . Based on our results , it would be predicted that parasites exposed to alkaline stress would have difficulty recovering from the insult . To test this hypothesis , intracellular parasites were exposed to pH 8 . 2 for 3 or 5 days , harvested , and then inoculated into fresh host cells and cultured under normal ( pH 7 . 0 ) conditions . Parasite proliferation was monitored using the PCR-based B1 assay . Data show that intracellular parasites lacking GCN5-A exposed to pH 8 . 2 for 3 days do not recover as efficiently as wild-type or the GCN5-A complemented clone ( Fig . S1 ) . The recovery defect is even more pronounced when the intracellular ΔGCN5-A parasites are subjected to pH 8 . 2 for 5 days ( Fig . S1 ) . The preceding studies were performed on tachyzoite-infected host cells . We also examined if direct exposure to alkaline stress produced a phenotype in the parasites . In order to test if alkaline pH impacts ΔGCN5-A parasites directly , we monitored the ability of purified , extracellular ΔGCN5-A tachyzoites to recover from a short term exposure to alkaline pH . Equal numbers of extracellular ΔGCN5-A or wild-type tachyzoites were placed in media of pH 7 . 0 ( control ) or 8 . 2 for 30 min , and then allowed to infect confluent host cell monolayers under normal culture conditions . At day five , plaques in the infected monolayers were counted . The ΔGCN5-A mutant displayed increased sensitivity to alkaline pH as evidenced by its impaired ability to produce plaques following this insult ( Fig . 7A ) . Recombinant TgGCN5-A was able to restore the ability of ΔGCN5-A parasites to recover from alkaline stress ( Fig . 7A ) . We verified the results with a second , independent type of growth assay that monitors parasite proliferation through quantitative PCR of the parasite-specific B1 gene ( Fig . 7B ) . Collectively , these studies establish that TgGCN5-A is a key factor that manages the Toxoplasma response to alkaline pH stress in RH strain tachyzoites . It is curious that the complemented clone , which appears to be over-expressing TgGCN5-A , does not offer greater protection from alkaline stress despite being able to up-regulate BAG1 and LDH2 greater than wild-type ( Fig . 6 ) . GCN5 HATs function in large multi-subunit complexes , the components of which vary in different cells or under different conditions . A possible explanation could be that a “minimal” GCN5 complex can up-regulate certain stress response genes , but it is not capable of providing greater protection to the cell because other components are required that are not over produced . We examined whether ΔGCN5-A parasites were hypersensitive to other stresses , including 30 minute exposure to 0 . 6 M KCl , 5 µM arsenite , 1 µM ionophore , or 42°C heat shock . Upon being placed back in culture following the exposure to these stresses , the ΔGCN5-A parasites grew similarly to wild-type with exception of those exposed to KCl stress ( Fig . 8A and 8B ) . Remarkably , TgGCN5-A appears to have a striking specificity for managing the alkaline and possibly KCl stress responses , possibly because each disrupts ion potential . Such narrow specificity in stress response has been reported for Schizosaccharomyces pombe GCN5 [15] . Toxoplasma gondii possesses two GCN5 KATs , which is unusual as lower eukaryotes tend to have a single GCN5 . We have sought to delineate the roles of these two GCN5s by making genetic knockouts . The ΔGCN5-A mutant is viable and does not show growth defects under normal culture conditions; however , attempts to generate a knockout of TgGCN5-B have not been successful . These results support a model that TgGCN5-B is essential for housekeeping functions while TgGCN5-A is required to overcome certain stress situations . Such a role for TgGCN5-A is established by the studies described herein . It remains possible that there is some functional overlap between the two TgGCN5s , but whatever contribution is made by TgGCN5-B is not sufficient to compensate for the loss of TgGCN5-A in terms of responding normally to alkaline pH stress , including the up-regulation of bradyzoite marker genes BAG1 and LDH2 . A key finding in this study is that TgGCN5-A is required to activate developmental genes in response to pH stress . These studies were performed in type I RH strain , which is not well suited for a more thorough characterization of bradyzoite development in vitro or in vivo . Such studies would require the generation of an analogous TgGCN5-A knockout in type II strain Toxoplasma . Our initial attempts to disrupt the TgGCN5-A knockout in type II strains have not yet succeeded , probably due to technical challenges inherent in working with the slow growing type II strain , but it is possible that TgGCN5-A is essential in type II strain . It is intriguing that Toxoplasma possesses a duplicate GCN5 HAT that appears to be exquisitely tailored to respond to alkaline , and to a lesser extent , KCl stress . Adaptation to fluctuations in pH is likely to be relevant to proliferating tachyzoites , as pH stress almost certainly is encountered by Toxoplasma in vivo as it moves in and out of host cells throughout diverse regions of the body . Additionally , while not addressed for Toxoplasma infection , it has been reported that other infections elevate intracellular pH [35] . It is difficult to distinguish whether the observed changes in intracellular parasites are due to direct effects of high pH on Toxoplasma parasites themselves or indirect effects produced by the response of host cells . What is clear is that RH strain parasites lacking TgGCN5-A are defective in regulating changes in the transcriptome that accompany the response to alkaline pH . The data are significant as alkaline pH is considered a “gold standard” method for triggering bradyzoite development in vitro [1] . In summary , our studies establish that TgGCN5-A plays a major role in the normal response to alkaline pH stress , including the activation of developmentally regulated genes , in Toxoplasma RH strain . The conclusion is based on multiple lines of data from independent studies , including up-regulation of TgGCN5-A mRNA during pH stress , microarray analysis , TgGCN5-A enrichment at genes up-regulated during alkaline stress , and phenotypic analysis showing that the TgGCN5-A knockout has impaired alkaline stress recovery . Our microarray analysis also provides novel insight into the molecular basis of the alkaline stress response in intracellular Toxoplasma .
Toxoplasma tachyzoites ( wild-type ( WT ) RH , ΔGCN5-A , and TgGCN5-A complemented lines ) were cultured in primary human foreskin fibroblasts ( HFF ) using Dulbecco's Modified Eagle's Medium ( DMEM ) supplemented with 1 . 0% fetal bovine serum ( FBS , Invitrogen ) . Parasites were grown in a humidified CO2 ( 5% ) incubator at 37°C . Cultures were confirmed to be free of Mycoplasma contamination by MycoAlert Assay ( Cambrex Bio Science ) . Parasites were harvested immediately following lysis of host cell monolayers and purified by filtration through a 3 . 0 micron filter [36] . Bradyzoite growth conditions were identical except the infection medium was replaced with alkaline medium ( DMEM with 20 mM HEPBS , 2 g NaHCO3/L and 1 . 0% FBS , adjusted to pH 8 . 2 using NaOH ) and changed daily . Tachyzoites were released from HFF monolayers using a 25 gauge syringe needle , and filtered to remove the host cell debris was spun out at 500×g for 5 min . Extracellular parasites were resuspended at a concentration of 105 parasites/ml in DMEM , alkaline medium ( pH 8 . 2 , adjusted with NaOH as described above ) , or medium containing 0 . 6 M KCl , 5 µM arsenite , 1 µM ionophore . This suspension was incubated at 37°C ( or 42°C for heat shock experiment ) in 5% CO2 for 30 min , and then 103 parasites were inoculated onto HFF monolayers in 24-well plates . After 5 day incubation at 37°C , the infected monolayers were fixed with 100% methanol and stained with crystal violet to score the number of plaques [36] or processed for B1 PCR assay [37] . Confluent HFF monolayers grown in T25-cm2 flasks were infected with 106 parasites using normal parasite culture media ( above ) . After 2 hr , infection medium was replaced with normal medium ( pH 7 . 0 ) or alkaline medium ( see above ) . Flasks were maintained in humidified 37°C incubator in 5% CO2 . Medium for both normal and alkaline cultures was replaced each day for 3 days , at which point the infected monolayers were scraped with a rubber policeman . Samples were centrifuged ( 1500 rpm , 10 min ) and resuspended in sterile PBS . Intracellular parasites were released from host cells by syringe passage using a 25 gauge needle and washed in PBS . Total RNA was isolated from the purified parasites using an RNeasy Mini Kit according to the manufacturer's instructions ( Qiagen ) . To minimize genomic DNA contamination , additional treatment with DNase was performed . The cDNA and cRNA were synthesized according to the protocols recommended by Affymetrix in their GeneChip Expression Analysis Technical Manual ( Affymetrix , Santa Clara , CA ) . Briefly , cDNA was synthesized using T7 promoter-dT24 oligonucleotide as primer with the Invitrogen Life Technologies SuperScript Choice system . Biotinylated cRNA was made using the Affymetrix in vitro transcription ( IVT ) labeling kit . Fifteen µg of biotinylated cRNA was added to a total hybridization cocktail of 300 µl , and 200 µl was hybridized to an Affymetrix custom T . gondii ToxoGeneChip ( http://roos-compbio2 . bio . upenn . edu/~abahl/Array-Tutorial . html ) after adding control oligonucleotides . Hybridization was performed at 45°C for 17 h with constant rotation . The hybridization mixture was then removed and the GeneChips were washed , stained with phycoerythrin-labeled streptavidin , washed , incubated with biotinylated anti-streptavidin , and re-stained with phycoerythrin-labeled streptavidin to amplify the signals; these steps were carried out using the Affymetrix Fluidics Station . To reduce non-random error , balanced groups of samples were handled in parallel . Arrays were then scanned using the dedicated scanner , controlled by Affymetrix GeneChip Operating Software ( GCOS ) . The Affymetrix Microarray Suite version 5 ( MAS5 ) algorithm was used to analyze the hybridization intensity data from GeneChip expression probe arrays and to calculate the set of metrics that describe probe set performance . The average intensity on each array was normalized by global scaling to a target intensity of 1000 . For each of the two conditions ( normal and alkaline ) , four independent preparations of WT and ΔGCN5-A RNA were prepared; each of the 16 preparations was hybridized to its own microarray to ensure a strong statistical basis for analysis . We analyzed only those probe sets ( genes ) that were called “present” by the MAS5 detection call in at least half of the arrays for at least one of the four conditions to eliminate probe sets that are not reliably detected ( those at or near background or that may reflect cross-hybridization ) [38] . We used a Welch's t-test on log2 transformed MAS5 signals to reveal significant differences between WT and ΔGCN5-A . The resultant p-values were used to calculate false discovery rates ( FDR ) using the Benjamini and Hochberg method [39] . Fold-changes were calculated by taking the ratio of the mean of the WT and ΔGCN5-A signal values , using the larger mean as the numerator; by convention we show the result as negative if the mean of the ΔGCN5-A samples was smaller . Microarray data are available at Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/geo/ ) , accession number GSE22100 . Primers were designed using the Primer Express 2 . 0 software ( Applied Biosystems , CA ) and are listed in Table S6 . The RT reaction was performed using 1 . 0 µg total RNA isolated from designated tachyzoites , with oligo-dT primers and Omniscript reverse transcriptase according to the manufacturer's directions ( Qiagen ) . 1 . 0 µl of a 1∶10 dilution of the resulting cDNA was amplified in a 25 µl total volume containing SYBR Green PCR Master Mix ( Applied Biosystems , CA ) and 0 . 5 µM of each forward and reverse primer . All reactions were performed in triplicate using the 7500 Real-time PCR system and analyzed with relative quantification software ( 7500 software v2 . 0 . 1 ) ( Applied Biosystems , CA ) . The Toxoplasma β-tubulin gene ( Genbank accession number M20025 ) was used to normalize the qRT-PCRs . β-tubulin was determined to be equivalent between samples using actin and GAPDH as normalizing controls . ChIP was performed on tachyzoites stably expressing fGCN5-A using polyclonal anti-FLAG antibody ( Sigma F7425 ) immobilized to Dynabeads Protein A ( Invitrogen ) . Quantitative PCR was performed as described above . Immunoprecipitated DNA samples were quantified using a standard curve created with serially diluted input DNA . 0 . 1 ng of total ChIP DNA was added to each reaction and reactions were performed in triplicate . Primers used are listed in Table S6 , each pair designed to amplify ∼90 bp regions located ∼1 . 0 kb upstream of the ATG start site . To complement ΔGCN5-A parasites , the ptubXFLAG::HX Toxoplasma expression vector [32] was modified to replace its HX minigene marker with a CAT minigene to confer resistance to chloramphenicol . Recombinant , tagged full-length TgGCN5-A was cloned into the vector using the NdeI and AvrII sites , referred to as ptubMYCGCN5-AFLAG::CAT . The TgGCN5-A coding sequence was amplified from Toxoplasma cDNA using Phusion® High-Fidelity DNA Polymerase ( New England Biolabs ) and primers containing NdeI and AvrII restriction enzyme sites ( italicized below ) . The sense primer contained sequence encoding the MYC epitope tag ( underlined ) and the antisense primer lacked the TgGCN5-A stop codon to allow in-frame fusion with a FLAG tag in the vector [32]: sense , 5′-ATACCATCATATGAAAATGGCGTACCCGTACGACGTCCCGGACTACGCGGAGACTGTCGAAGTGCCTGCATTC; antisense , 5′-ATACCATCCTAGGGAAACTCCCGAGAGCCTCGACCTTGGGCC . 106 ΔGCN5-A parasites were transfected with 20 µg NotI-linearized vector and selected for resistance to 20 µM chloramphenicol before cloning by limiting dilution as previously described [36] . Multiple clones were selected and verified to express ectopic MYCGCN5-AFLAG protein by IFA using anti-FLAG ( Sigma F7425 ) . Phenotypes reported were similar for multiple independent clones . | Protozoan parasites cause significant disease in humans and livestock , and many of our current therapies have serious side effects or are being rendered useless due to the development of drug resistance . These parasites typically have complex life cycles involving multiple hosts and some , like Toxoplasma gondii , have the ability to remain in the host for life as a latent tissue cyst . Toxoplasma is one of the most successful parasites on Earth because the ability to develop into a tissue cyst greatly facilitates transmission through carnivores . Cyst formation also is responsible for recrudescent infection in immunocompromised patients . The conversion of Toxoplasma from its replicating cell to the cyst is triggered by stress , but we have little understanding of how the parasite stress response functions . In this study , we identify the genes involved in Toxoplasma's response to alkaline stress , which is a known trigger of cyst development . We also establish that a lysine acetyltransferase enzyme called TgGCN5-A is required for type I RH strain Toxoplasma to respond normally to alkaline stress . Parasites lacking TgGCN5-A are no longer capable of activating genes induced during cyst formation triggered by alkaline pH . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/epigenetics",
"molecular",
"biology/chromatin",
"structure",
"microbiology/parasitology"
] | 2010 | Toxoplasma gondii Lysine Acetyltransferase GCN5-A Functions in the Cellular Response to Alkaline Stress and Expression of Cyst Genes |
Campylobacter jejuni is a major food-borne pathogen and a common causative agent of human enterocolitis . Fluoroquinolones are a key class of antibiotics prescribed for clinical treatment of enteric infections including campylobacteriosis , but fluoroquinolone-resistant Campylobacter readily emerges under the antibiotic selection pressure . To understand the mechanisms involved in the development of fluoroquinolone-resistant Campylobacter , we compared the gene expression profiles of C . jejuni in the presence and absence of ciprofloxacin using DNA microarray . Our analysis revealed that multiple genes showed significant changes in expression in the presence of a suprainhibitory concentration of ciprofloxacin . Most importantly , ciprofloxacin induced the expression of mfd , which encodes a transcription-repair coupling factor involved in strand-specific DNA repair . Mutation of the mfd gene resulted in an approximately 100-fold reduction in the rate of spontaneous mutation to ciprofloxacin resistance , while overexpression of mfd elevated the mutation frequency . In addition , loss of mfd in C . jejuni significantly reduced the development of fluoroquinolone-resistant Campylobacter in culture media or chickens treated with fluoroquinolones . These findings indicate that Mfd is important for the development of fluoroquinolone resistance in Campylobacter , reveal a previously unrecognized function of Mfd in promoting mutation frequencies , and identify a potential molecular target for reducing the emergence of fluoroquinolone-resistant Campylobacter .
Campylobacter jejuni , a Gram-negative microaerobic bacterium , is one of the most prevalent bacterial foodborne pathogens in humans , causing more than 2 million cases of diarrhea each year in the U . S . alone [1] , [2] , [3] . As an enteric pathogen , this organism causes watery diarrhea and/or hemorrhagic colitis . Campylobacter infection is also the most common antecedent to Guillain-Barre syndrome , an acute flaccid paralysis that may lead to respiratory muscle compromise and death [4] , [5] . In developed countries , person-to-person transmission of Campylobacter is rare , and the main source of human Campylobacter infections is via food , water , or milk contaminated by Campylobacter [6] . Fluoroquinolone ( FQ ) antimicrobials are often prescribed for clinical treatment of diarrhea caused by enteric bacterial pathogens including Campylobacter [7] , [8] . However , Campylobacter is increasingly resistant to FQ antimicrobials , which has become a major concern for public health [9] , [10] , [11] . FQ-resistant ( FQR ) Campylobacter developed in food producing animals can be transmitted to humans via the food chain . Poultry are considered the major reservoir for C . jejuni and a significant source for FQR Campylobacter infections in humans , because the majority of domestically acquired cases of human campylobacteriosis result from consumption of undercooked chicken or food contaminated by raw chicken [2] , [12] , [13] . Although FQ antimicrobials have been banned since 2005 in poultry production in the U . S . , FQR Campylobacter continue to persist on poultry farms [14] , [15] , [16] . The main targets of FQs in bacteria are DNA gyrases and/or topoisomerase IV [17] , [18] . In Campylobacter , the resistance to FQ antimicrobials is mediated by point mutation in the quinolone resistance-determining region ( QRDR ) of gyrA in conjunction with the function of the multidrug efflux pump CmeABC [10] , [19] , [20] , [21] . Acquisition of high-level FQ resistance in Campylobacter does not require stepwise accumulation of point mutations in gyrA . Instead , a single point mutation in gyrA can lead to clinically relevant levels of resistance to FQ antimicrobials [19] , [20] , [22] . Specific mutations at positions Thr-86 , Asp-90 and Ala-70 in GyrA have been linked to FQ resistance in C . jejuni [10] , [19] . When enumerated by ciprofloxacin ( CIPRO ) -containing plates , spontaneous FQR Campylobacter mutants occur at a frequency as high as 10−6 [23] , suggesting that C . jejuni possess a high mutation rate to FQ resistance . CmeABC , an energy-dependent efflux system , contributes significantly to the intrinsic and acquired resistance to FQs in C . jejuni by reducing the accumulation of the antibiotics within Campylobacter cells [19] , [20] , [24] , [25] . The expression level of cmeABC also influences the frequencies of emergence of spontaneous FQR mutants [23] . One unique feature of FQ resistance development in Campylobacter is the rapid emergence of FQR mutants from a FQ-susceptible population when treated with FQ antimicrobials . This has been observed in Campylobacter-infected animals or patients treated with FQs [19] , [26] , [27] , [28] , [29] , [30] . In chickens infected with FQ-susceptible Campylobacter , treatment with enrofloxacin resulted in the emergence of FQR Campylobacter mutants that were detected in feces within 24–48 hours after the initiation of treatment , and the FQR population continued to expand during the treatment and eventually occupied the intestinal tract at a density as high as 107 CFU/g feces [19] , [29] , [30] . As shown in a comparison study , the same FQ treatment did not result in the development and enrichment of FQR E . coli in chickens [29] , suggesting that C . jejuni has a unique ability to adapt to FQ treatment . This high frequency of emergence of FQR Campylobacter mutants in response to the selection pressure may have directly contributed to the global prevalence of FQR Campylobacter . For example , multiple studies have shown the temporal link between the approval of FQ antimicrobials for use in animal production and the rapid increase of FQR Campylobacter isolates from both animals and humans [9] , [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] , [39] . In some regions of the world , the vast majority of Campylobacter isolates have become resistant to FQ antimicrobials [22] , [40] , [41] . The rapidness and magnitude of FQ resistance development in Campylobacter in response to FQ treatment suggest that both selective enrichment of pre-existing spontaneous mutants and adaptive gene expression may contribute to the emergence of FQR Campylobacter , but how Campylobacter responds to FQ treatment is unknown . Within bacterial cells , FQ antimicrobials form a stable complex with gyrases and DNA , which generates double-stranded breaks in DNA and leads to bacterial death [18] . In other bacteria , antibiotic treatments ( including FQs ) induce the SOS response , which upregulates multiple genes involved in DNA repair , recombination , and mutation as well as other functions [42] , [43] , [44] , [45] . The SOS response is controlled by LexA , a transcriptional repressor . DNA damage triggers LexA autocleavage , which derepresses the SOS genes controlled by LexA . Once activated , SOS response promotes the development of drug resistance , horizontal transfer of genetic materials , and production of virulence factors [45] , [46] , [47] . Unlike many other bacterial organisms , epsilonproteobacteria including Campylobacter and Helicobacter don't have a LexA ortholog [46] and also lack many genes involved in DNA repair , recombination , and mutagenesis , such as the mutHL genes ( methyl-directed mismatch repair ) , the umuCD genes ( UV-induced mutagenesis ) , and SOS-controlled error-prone DNA polymerases [48] , [49] , [50] . These observations suggest that Campylobacter may not have the typical SOS response system . In light of this possibility , it is intriguing to determine how Campylobacter copes with FQ treatment and what facilitates the emergence of FQR mutants in Campylobacter . In this study , we examined the gene expression profiles of C . jejuni NCTC 11168 in response to treatment with CIPRO . Consistent with the prediction , a typical SOS response was not observed in Campylobacter treated with CIPRO . However , 45 genes showed ≥1 . 5-fold ( p<0 . 05 ) changes in expression when Campylobacter was exposed to a suprainhibitory dose of CIPRO for 30 min . One of the up-regulated genes was mfd ( mutation frequency decline ) , which encodes a transcription-repair coupling factor involved in DNA repair . The mfd gene in E . coli was originally linked to the phenotype of mutation frequency decline [51] , [52] . Subsequently it was found that Mfd functions as a transcription-repair coupling factor and promotes strand-specific DNA repair [53] , [54] . DNA lesions stall RNA polymerase during transcription . Mfd displaces the stalled RNA polymerase from the DNA lesions in an ATP-dependent manner , recruits the UvrABC excinuclease complex , and enhances the repair of the DNA lesions on the transcribed strand [54] , [55] . Thus , Mfd couples transcription with DNA repair and contributes to mutation frequency decline . Recently it was reported that depending on the nature of DNA damage and the availability of NTPs , Mfd can also promote the forward translocation of arrested RNA polymerase in the absence of repair , leading to transcriptional bypass of non-repaired lesions [55] . In contrast to its previously known function in the decline of mutation frequency in other bacterial organisms [51] , [52] , Mfd in Campylobacter was found to promote the emergence of spontaneous FQR mutants and the development of FQR mutants under FQ treatments in this study . These findings define a novel function of Mfd and significantly improve our understanding of the molecular mechanisms underlying the development of FQR Campylobacter .
To understand the adaptive response of Campylobacter to FQ treatment , DNA microarray was used to analyze the transcriptional changes in C . jejuni NCTC 11168 after exposure to CIPRO . When the Campylobacter cells were treated with a subinhibitory concentration ( 0 . 06 µg/ml; 0 . 5× the MIC ) of CIPRO for 1 . 5 hours , no genes showed ≥1 . 5-fold changes in expression , suggesting that the transcriptional response to the low dose of CIPRO was very limited . When the Campylobacter cells were treated with a suprainhibitory concentration ( 1 . 25 µg/ml; 10× the MIC ) of CIPRO for 30 min , 45 genes showed ≥1 . 5-fold ( p<0 . 05 ) changes in expression ( Table 1 ) , among which 13 were up-regulated and 32 were down-regulated . The up-regulated genes are involved in cell membrane biosynthesis , cellular processes , and transcription-coupled DNA repair or have unknown functions , while the majority of the down-regulated genes are involved in energy metabolisms ( Table 1 ) . Consistent with the lack of LexA , the core genes involved in SOS responses in other bacteria , such as recA , uvrA , ruvC , ruvA , and ruvB , did not show significant changes in expression . The expression of other genes involved in DNA repair and recombination also did not change significantly . These findings indicate that C . jejuni does not mount a typical SOS response or upregulate the general DNA repair system in the early response to CIPRO treatment . Notably , cj1085c , a homolog of mfd , was upregulated in the presence of CIPRO . Two up-regulated genes , uppP and uppS , encode products involved in cell wall production [56] , [57] , while pldA encodes an outer membrane phospholipase that has been implicated in hemolysis , capsular production , and virulence [58] , [59] . According to the Q values , the identified genes would have an estimated false discovery rate ( FDR ) of 20% . However , quantitative real-time RT-RCR ( qRT-PCR ) confirmed all of the 11 genes selected from the microarray list ( Table 1 ) , suggesting that the actual FDR is lower than the estimation . Cj1085c ( 978aa ) was annotated as Mfd [48] and shows 31 . 5% amino acid identity to the E . coli Mfd protein ( 1148 aa ) . In addition , it contains the characteristic domains conserved in Mfd proteins , such as the ATP/GTP-binding site motif and the superfamily II helicase motif . Mfd in other bacteria has been shown to be involved in strand-specific DNA repair by displacing lesion-stalled RNA polymerase and recruiting enzymes involved in recombination events [54] , [60] . The mfd locus is highly conserved in Campylobacter and is present in all Campylobacter species and C . jejuni strains that have been sequenced to date . The Mfd proteins in different Campylobacter species share 57–79% identity to the Mfd in C . jejuni NCTC 11168 . Within C . jejuni , the Mfd proteins are 98–100% homologous among different strains . The mfd gene is located in the middle of a gene cluster , whose transcription is in the same direction ( partially shown in Fig . 1A ) . The downstream gene Cj1084c encodes a putative ATP/GTP binding protein , while the upstream gene Cj1086c encode a hypothetical protein [48] . It is unknown if mfd and its flanking genes form an operon , but it appeared that Cj1086c and mfd were co-transcribed because a RT-PCR product spanning both ORFs was amplified ( data not shown ) . Since mfd was the only DNA repair related gene that showed a significant change in expression in the early response of C . jejuni to CIPRO treatment ( Table 1 ) , we examined its role in the emergence of spontaneous FQR mutants in Campylobacter . Firstly , the mfd gene was inactivated by insertional mutagenesis ( Fig . 1B ) . As shown in Fig . 2 , the mfd mutant ( JH01 ) showed a approximately 100-fold reduction in the frequencies of emergence of spontaneous FQR mutants detected using plates containing three different concentrations ( 1 , 2 , and 4 µg/ml , respectively ) of CIPRO . Complementation of the mfd mutant in trans by a plasmid-carried mfd restored the frequencies of mutant emergence to the wild-type level ( JH02 in Fig . 2 ) . As determined by qRT-PCR , the expression level of mfd in the complemented construct ( JH02 ) was fully restored ( 1 . 7× the wild-type level ) . pRY112 alone ( without the cloned mfd gene ) did not complement the mfd mutant in the mutation frequency ( data not shown ) . These results indicate that Mfd contributes significantly to the rate of spontaneous mutations to FQ resistance . Secondly , we determined if the enhanced expression of mfd increases the mutation frequency . For this purpose , we constructed strain JH03 , which was a wild-type 11168 strain containing an extra copy of mfd carried on a shuttle plasmid . In JH03 , the mRNA of mfd increased 3 . 8 times compared with that in 11168 as determined by qRT-PCR . When compared with the wild-type 11168 , the frequency of emergence of FQR mutants from JH03 increased about 10-fold ( Fig . 2 ) . The increase was reproducible in multiple experiments and was statistically significant ( P<0 . 05 ) . These results indicated that overexpression of mfd increases the frequency of emergence of spontaneous FQR mutants . Given that there is only one nucleotide between the mfd gene and its downstream gene cj1084c , it was prudent to determine if the mfd mutation resulted in a polar effect on the expression of cj1084c . RT–PCR showed that cj1084c was transcribed at a comparable level in both the mfd mutant and the wild-type NCTC 11168 ( Fig . 1C ) . RT-PCR was also performed using 10-fold serial dilutions of the RNA template , which yielded comparable results between the two strains ( data not shown ) . PCR without the reverse transcriptase did not yield a product ( Fig . 1C ) , indicating that the mRNA templates had no DNA contamination . These results suggested that the insertional mutation in the mfd gene did not cause an apparent polar effect on expression of the downstream gene . This finding plus the complementation data ( Fig . 2 ) strongly indicate that loss of Mfd is responsible for the observed reduction in the mutation frequency in JH01 . To examine if the reduction in the emergence of spontaneous FQR mutants is caused by the increased susceptibility of the mfd mutant to CIPRO , we compared the MICs of several antibiotics in the mfd mutant with those in the wild type . Our results did not reveal any differences between the mutant and the wild type in their susceptibility to the tested antibiotics including erythromycin , ampicillin , streptomycin , and CIPRO ( data not shown ) . In addition , there was no apparent difference in growth kinetics between the wild-type and the mfd mutant either in MH broth ( without antibiotics ) or in MH broth supplemented with a subinhibitory concentration ( 0 . 06 µg/ml ) of CIPRO ( Fig . 3 ) . The growth rates of the mfd over-expressing strain ( JH03 ) and the complemented mutant ( JH02 ) were also similar to that of the wild type ( Fig . 3 ) . Thus , the reduced spontaneous mutation rate to FQ resistance in the mfd mutant was not attributable to decreased growth rate or increased susceptibility to antibiotics . In addition , the CIPRO-resistant colonies examined for gyrA mutations all carried the C257T mutation in gyrA and had a CIPRO MIC of >32 µg/ml regardless of the backgrounds ( 11168 or JH01 ) from which the mutants were selected . FQR Campylobacter mutants emerge rapidly from a FQ-susceptible population once treated with FQ antimicrobials [19] , [26] , [27] , [28] , [29] , [30] . To determine if Mfd influences the development of FQR Campylobacter under selection pressure , we conducted in vitro growth experiments , in which C . jejuni was treated with a suprainhibitory concentrations of CIPRO ( 4 µg/ml ) . In the first treatment experiment , 109 CFU of bacterial cells were inoculated into each flask containing 100 ml MH broth with 4 µg/ml of CIPRO , yielding an initial cell density of 107 CFU/ml . At the beginning of the treatment , 1–3 CFU/ml of FQR mutants were detected in the flasks inoculated with 11168 , while no FQR mutants were detected in the cultures inoculated with JH01 ( Fig . 4A ) . One day after the initiation of the treatment , the numbers of FQR mutants in the 11168 cultures grew to a level ranging from a few hundreds to a few thousands CFU/ml , while no mutants or about 1 CFU/ml of FQR mutants were detected in the cultures of JH01 ( Fig . 4A ) . The FQR populations expanded on day 2 in both strains , but the FQR population of JH01 was still about 1 , 000-fold less than that of 11168 . Due to the continued enrichment of the FQR mutants by CIPRO and the fact that the mutants of 11168 was entering the stationary phase , the average difference between 11168 and JH01 on day 3 decreased , but was still more than one order of magnitude ( Fig . 4A ) . In the second experiment , 2×107 CFU bacterial cells of 11168 or JH01 were inoculated into each flask containing 20 ml of MH broth with 4 µg/ml of CIPRO , yielding an initial cell density of 106 CFU/ml . At the beginning of the treatment , no FQR mutants were detected from either 11168 or JH01 ( Fig . 4B ) . On day 1 after the initiation of the treatment , FQR Campylobacter emerged from some of the cultures of 11168 and continued to expand in numbers on day 2 and day 3 . In contrary to 11168 , no FQR mutants emerged from any of the JH01 cultures during the three-day incubation ( Fig . 4B ) . In the third experiment , the inoculum was decreased to 2×104 CFU per flask ( initial cell density = 103 CFU/ml ) , and no FQR mutants were detected from either 11168 or JH01 after three day's incubation ( data not shown ) . These results indicated that emergence of FQR mutants under treatment with CIPRO was influenced by the initial bacterial cell density and facilitated by the function of Mfd . To determine if Mfd influences the emergence of FQR Campylobacter during in vivo therapeutic treatment , broiler chickens were infected with 11168 or JH01 and then treated with enrofloxacin administered in drinking water ( 50 ppm ) . The birds in both groups were quickly colonized by C . jejuni after inoculation ( Fig . 5 ) . Before the treatment with enrofloxacin , all birds were colonized by Campylobacter and the colonization levels ( CFU/g feces ) were similar in both groups ( p>0 . 05 ) . One day after initiation of the treatment , the number of colonized chickens and the levels of colonization decreased drastically in both groups , with Campylobacter detectable only in three chickens that were inoculated with the wild-type strain ( Fig . 5A ) . After that , the numbers of Campylobacter in both groups rebounded . On day 3 after the initiation of the treatment , all of the birds in the 11168 group were re-colonized by Campylobacter and remained colonized until the end of the experiment . For the group inoculated with JH01 , 6 of the11 birds became positive with Campylobacter on day 3 after initiation of the treatment ( Fig . 5A ) and 3 birds remained negative until the end of the experiment . On days 3 , 5 and 7 after initiation of the treatment , the average colonization level of the JH01-inoculated group was approximately 3 log units lower than that of the 11168-inoculated group ( Fig . 5A ) and the differences were statistically significant ( p<0 . 05 ) . The number of FQR Campylobacter in each chicken was also monitored . Prior to the treatment , no FQR C . jejuni was detected in any of the chickens ( Fig . 5B ) . On day 1 after initiation of the treatment , the three Campylobacter-positive birds of the 11168-inoculated group still carried FQ-susceptible Campylobacter . However , FQR C . jejuni appeared on day 3 in all birds of the 11168-inoculated group and in some birds of the JH01-inoculated group ( Fig . 5B ) . Comparison of the total Campylobacter counts ( Fig . 5A ) with the numbers of FQR Campylobacter ( Fig . 5B ) revealed that the birds were re-colonized by FQR mutants after initiation of the treatment . The average numbers of FQR Campylobacter in the JH01-inoculated group were approximately 3 log units lower that those of the 11168-inoculated group ( Fig . 5B ) and the differences were statistically significant ( p<0 . 05 ) . These results indicate that loss of Mfd significantly reduced the rates of emergence of FQR Campylobacter in enrofloxacin-treated chickens . Representative Campylobacter isolates obtained at different sampling times from both groups were tested for CIPRO MICs using E-test strips . The result showed that before treatment all the tested isolates from both groups were susceptible to CIPRO ( MICs = 0 . 094–0 . 125 µg/ml ) . The majority of the tested isolates from day 1 after initiation of the treatment were still susceptible to CIPRO ( MICs = 0 . 094–0 . 5 µg/ml ) . On day 3 after the initiation of treatment , 21 of the 22 tested isolates ( from both groups ) had a CIPRO MIC of >32 µg/ml and the other one had an MIC of 8 µg/ml . Similarly , the majority ( 44 out of 49 ) of the tested isolates from days 5 and 7 had a CIPRO MIC of >32 µg/ml and the rest had MICs from 1–24 µg/ml . The MIC results further confirmed the differential plating results that the chickens were re-colonized by FQR Campylobacter .
When Campylobacter cells were treated with a subinhibitory concentration ( 0 . 06 µg/ml , 0 . 5× the MIC ) of CIPRO for 1 . 5 hours , no significant changes in gene expression were detected using the cut-off criteria defined in this study . This result was somewhat similar to the study with Haemophilius influenzae [61] in that the treatment with a low concentration of CIPRO induced few changes in gene expression , but was different from that study because several genes involved in SOS response were upregulated in Haemophilius influenzae . Prolonged treatment of Campylobacter with the subinhibitory concentration of CIPRO may reveal noticeable changes in gene expression , but culturing Campylobacter with 0 . 06 µg/ml of CIPRO reduces its growth rate ( Fig . 3 ) , which will make the comparison with the non-treated control unfeasible and complicate the interpretation of the microarray results . To mimic clinical treatment , C . jejuni cells were exposed to a suprainhibitory dose ( 1 . 25 µg/ml , 10× the MIC ) of CIPRO . This dose is within the concentration range of CIPRO in gut contents during FQ treatment in chickens [62] . The reason that we treated the samples for 0 . 5 hour instead of a longer time was to detect the primary response triggered by CIPRO , instead of the secondary response caused by cell death . When Campylobacter cells were treated with this suprainhibitory dose for 0 . 5 hour , the expression of multiple genes was significantly altered ( Table 1 ) . Notably , the majority of the affected genes were downregulated and many of them are involved in cellular processes and energy metabolism ( Table 1 ) . This result is similar to the findings obtained with other bacteria [43] , [44] , [61] and supports the notion that reducing cellular metabolism is a common strategy utilized by bacteria to cope with antibiotic treatment . Within bacterial cells CIPRO interacts with gyrase and DNA , blocking DNA replication and transcription [18] . When exposed to CIPRO , the expression of gyrA and gyrB in various bacteria was either altered or unchanged [43] , [61] , [63] . In this study , we found that the expression of gyrA , gyrB , and topA was not significantly affected in Campylobacter by CIPRO . In addition , the expression of the genes encoding enzymes involved in DNA repair , recombination , or mutagenesis , such as recA , ruvABC , uvrABC , and mutS , did not change significantly . Only two genes involved in DNA metabolism ( mfd and dnaE ) were affected by CIPRO under the conditions used in this study ( Table 1 ) . Theses observations indicate that C . jejuni does not mount a typical SOS response under the treatment with FQ . These findings are also consistent with the fact that C . jejuni lacks LexA , the key regulator of bacterial SOS responses [46] . In addition to transcription-coupled DNA repair , Mfd has been associated with other functions in bacteria [64] . For example , Mfd of Bacillus subtilis is involved in homologous DNA recombination and stationary-phase mutagenesis [65] , [66] . Inactivation of the mfd gene of B . subtilis resulted in a great reduction in the number of prototrophic revertants to Met+ , His+ , and Leu+ during starvation [66] , indicating that Mfd promotes adaptive mutagenesis . This finding is in contrast to the known function of Mfd in mediating mutation frequency decline and could be explained by the role of Mfd in promoting transcriptional bypass and consequently increasing the adaptive mutagenesis rates [66] . In this study we found that Mfd increases the frequency of emergence of spontaneous FQR mutants in Campylobacter ( Fig . 2 ) . Furthermore , the mfd mutation also decreased the frequency of emergence of spontaneous streptomycin-resistant mutants in Campylobacter ( data not shown ) . Together , the results convincingly showed that Mfd is an important player in modulating the mutation rates in Campylobacter . To our knowledge , this is the first report documenting the key role of Mfd in promoting spontaneous mutation rates in a bacterial organism . How Mfd contributes to the increased mutation rates in Campylobacter is unknown , but it can be speculated that transcriptional bypass mediated by Mfd may actively occur in replicating non-stressed Campylobacter populations , resulting in an elevated level of retromutagenesis ( fixed changes in DNA sequence due to transcriptional mutation [67] ) that contributes to the size of the mutant pools . This possibility remains to be examined in future studies . Although mfd contributes significantly to the mutation rate ( Fig . 2 ) , its expression level was not precisely proportional to the mutation frequencies . For example , expression of mfd was upregulated 3 . 8-fold in JH03 , but its mutation frequency increased 10-fold . This difference is probably due to the fact that emergence of spontaneous mutants is a multi-step process and Mfd only contributes to one of the steps in the process . It is also possible that Mfd interacts with other proteins in modulating the mutation frequency . Thus , the changes in mfd expression level and the mutation frequency are not exactly at the same scale . Another interesting observation of this study is the upregulation of mfd by CIPRO . The enhanced expression may be needed for transcription repair because CIPRO treatment causes DNA damage , which stalls RNA polymerase . Alternatively , the increased production of Mfd may enhance transcriptional bypass of the non-repaired DNA lesions in order to maintain cell viability and/or promote mutations for resistance . This possibility is high given the facts that massive DNA damage incurred by a suprainhibitory dose of CIPRO may overwhelm the DNA repair system and Campylobacter must maintain certain levels of transcription to survive the treatment , that Mfd contributes significantly to the mutation rates to FQ resistance ( Fig . 2 ) , and that Campylobacter does not have the error-prone DNA polymerases , such as Pol II , Pol IV , and Pol V [48] . E . coli and other bacteria have these error-prone DNA polymerases [68] , [69] , which are repressed by LexA , but upregulated by the SOS response triggered by DNA damage . Once produced , the enzymes perform translesion DNA synthesis , allowing replication to continue without DNA repair . This special functional feature results in reduced genetic fidelity , but allows for bacterial survival under stress . The outcome of the enhanced expression of the error-prone enzymes is the increased mutation rates , which contribute to the emergence of drug resistance [70] . In the absence of a SOS response and the error-prone DNA polymerases , Campylobacter may use Mfd as an alternative pathway to increase mutation rates . Thus , enhanced expression of mfd may represent an adaptive response of Campylobacter to the stresses imposed by CIPRO treatment . How CIPRO upregulates Campylobacter Mfd is unknown and further work in this direction is warranted . FQR Campylobacter readily emerges from a FQ-susceptible population when treated with FQ antimicrobials ( Figs . 4 and 5 ) . As shown in the in vitro experiment , the development of FQR population under CIPRO treatment is influenced by the initial cell density ( Fig . 4 and the corresponding text ) as well as the functional state of Mfd . Considering the differences in spontaneous mutation rate between 11168 and JH01 ( Fig . 2 ) , it was likely that the 11168 and JH01 inocula had different numbers of pre-existing FQR mutants , which were selected by CIPRO and contributed to the differences in the FQR population detected in the cultures of the two strains . The inoculum-dependent emergence of FQR mutants in both 11168 and JH01 suggests that development of FQR Campylobacter under FQ treatment involves selection of preexisting mutants . However , the magnitude and dynamics of FQR development can not be totally explained by selection . For example , in some cultures FQR mutants were not detectible until the 2nd day of the incubation ( Fig . 4 ) . A single mutant at time zero in a culture flask would grow to a population of more than 2 , 000 cells in one day ( the generation time of C . jejuni in MH broth is about 2 hours ) , which would be readily detected by the plating method on day 1 . Thus , if selection was the only factor in the development of FQR Campylobacter , the latest time for detecting the pre-existing mutants in the mutant-positive flasks would be day 1 after initiation of the treatment . Obviously , this was not the case for all of the cultures because some of them did not show FQR mutants until day 2 ( Fig . 4 ) . In addition , some cultures were negative with FQR mutants at time zero , but showed a large number of mutants at day 1 , which could not be easily explained by sole selection of a few preexisting mutants from the inocula . Considering these unexplainable observations and the fact that a small fraction of the FQ-susceptible inoculum survived the killing effect as long as one day after the initiation of the treatment ( data not shown ) , it was possible that new FQR mutants were developed during the treatment . If this occurs , Mfd may enhance the emergence of new mutants by promoting transcriptional bypass or other mechanisms , which may partly explain the differences between 11168 and JH01 in the dynamics of emergence of FQR mutants . Thus , there is a possibility that both selection of pre-existing mutants and de nova formation of mutants are involved in the development of FQR Campylobacter during treatment with FQ antimicrobials . The role of Mfd in the development of FQR mutants was further shown by the in vivo experiment , in which Campylobacter-infected chickens were treated with enrofloxacin ( Fig . 5 ) . Previous studies have shown that therapeutic use of FQ antimicrobials in chickens promotes the emergence of FQR Campylobacter [19] , [27] , [28] , [29] , [30] , which can be potentially transmitted to humans via the food chain . In this study , we showed that inactivation of mfd significantly reduced the development of FQR Campylobacter in chickens ( Fig . 5 ) . In fact , several birds in the JH01-inoculated group became negative with Campylobacter once the treatment was initiated . Since the mfd mutant did not show a growth defect in vitro ( Fig . 3 ) and colonized chickens as efficiently as the wild-type strain ( see the colonization level before treatment in Fig . 5 ) , the observed differences in the development of FQR mutants were not due to changes in growth characteristics . These in vivo results ( Fig . 5 ) plus the in vitro findings ( Fig . 4 ) clearly showed that Mfd plays an important role in the development of FQR Campylobacter mutants under the selection pressure . To our knowledge , this is the first report that documents the role of Mfd in the development of FQ resistance in a bacterial pathogen . Since Mfd is highly conserved in bacterial organisms [64] , it would be interesting to know if this finding applies to other bacterial pathogens . In addition , inhibition of Mfd functions may represent a feasible approach to reducing the emergence of FQR Campylobacter .
C . jejuni strain NCTC 11168 ( ATCC 700819 ) was used in this study . The strain was routinely grown in Mueller-Hinton ( MH ) broth ( Difco ) or on MH agar at 42°C under microaerobic conditions ( 5% O2 , 10% CO2 , and 85% N2 ) . The media were supplemented with kanamycin ( 50 µg/ml ) or chloramphenicol ( 4 µg/ml ) as needed . Escherichia coli cells were grown at 37°C with shaking at 200 r . p . m . in Luria Bertani ( LB ) medium which was supplemented with ampicillin ( 100 µg/ml ) or kanamycin ( 30 µg/ml ) when needed . DNA microarray was used to identify genes that were differentially expressed in C . jejuni 11168 treated with CIPRO . For RNA isolation , Campylobacter cells were grown for 24 hours to the mid exponential phase ( OD600≈0 . 1∼0 . 15 ) and split into two equal portions , one of which was treated with CIPRO and the other served as a non-treated control . A subinhibitory concentration ( 0 . 06 µg/ml , 0 . 5× the MIC ) and a suprainhibitory dose ( 1 . 25 µg/ml , 10× the MIC ) of CIPRO were used in the treatments . For the treatment with 0 . 06 µg/ml of CIPRO , the treated and non-treated samples were incubated at 42°C for 1 . 5 hours under microaerobic conditions , while for the treatment with 1 . 25 µg/ml of CIPRO , the samples were incubated at 42°C for 30 min under microaerobic conditions . Immediately after the incubation , RNAprotect Bacteria Reagent ( Qiagen , Valencia , CA ) was added to the cultures to stabilize mRNA . The total RNA from each sample was extracted using the RNeasy Mini Kit ( Qiagen ) . The purified RNA samples were treated with On-Column DNase Digestion Kit ( Qiagen ) followed by further treatments with DNase to remove residual DNA contamination . RNA samples were extracted from 6 independent treatments with each concentration of CIPRO . Absence of contaminating DNA in the RNA samples was confirmed by RT-PCR . The concentration of total RNA was estimated with the NanoDrop ND-1000 spectrophotometer ( NanoDrop , Wilmington , DE , USA ) , and the integrity and size distribution of the purified RNA was determined by denaturing agarose gel electrophoresis and ethidium bromide staining . The quality of total RNA was further analyzed using the Agilent 2100 Bioanalyzer ( Agilent Technologies , Santa Clara , CA , USA ) , which showed the good quality and integrity of the RNA samples ( Data not shown ) . cDNA synthesis and labeling , microarray slide ( Ocimum Biosolutions ) hybridization , Data collection and normalization , and statistical analysis were performed as described in a previous publication [71] . For each type of treatment ( 0 . 06 µg/ml for 1 . 5 hours or 1 . 25 µg/ml for 30 min ) , six microarray slides were hybridized with RNA samples prepared from 6 independent experiments . For this study , we chose p-value<0 . 05 and the change ≥1 . 5-fold as the cutoff for significant differential expression between the treated and non-treated samples . Representative genes identified by the DNA microarray were further confirmed by qRT-PCR as described in a previous work [72] . The primers used for qRT-PCR are listed in Table 2 . An isogenic mfd ( cj1085c ) mutant of strain NCTC 11168 was constructed by insertional mutagenesis . Primers mfd-F2 ( 5′-TGTTGATGGAGAGTTAAGTGGTAT-3′ ) and mfd-R2 ( 5′-AATAGCATTCATAGCGACTTCTGTT-3′ ) were designed from the published genomic sequence of this strain [48] and used to amplify a 1 . 8-kb fragment spanning the 5′ region of mfd . Amplification was performed with Pfu Turbo DNA Polymerase ( Stratagene , La Jolla , CA , USA ) . The blunt-ended PCR product was purified using a QIAquick PCR purification kit ( Qiagen , Valencia , CA , USA ) , ligated to SmaI–digested suicide vector pUC19 , resulting in the construction of pUC-mfd , which was then transformed into E . coli DH5α . Since a unique EcoRV site ( which generates blunt ends ) occurs in the cloned mfd fragment , pUC-mfd was digested with EcoRV to interrupt the mfd gene . Primers KanNco-F ( 5′ CTT ATC AAT ATA TCC ATG GAA TGG GCA AAG CAT 3′ ) and KanNco-R ( 5′ GAT AGA ACC ATG GAT AAT GCT AAG ACA ATC ACT AAA 3′ ) were used to amplify the aphA3 gene ( encoding kanamycin resistance ) from the pMW10 vector [73] by using Pfu Turbo DNA polymerase ( Stratagene ) . The aphA3 PCR product was directly ligated to EcoRV-digested pUC-mfd to obtain construct pUC-mfd-aphA3 , in which the aphA3 gene was inserted within mfd ( the same direction as the transcription of mfd ) and the insertion was confirmed by PCR using primers mfd-F2 and Kana-intra ( 5′ GAA GAA CAG TAT GTC GAG CTA TTT TTT GAC TTA 3′ ) . The pUC-mfd-aphA3 construct , which served as a suicide vector , was electroporated into C . jejuni NCTC 11168 . Transformants were selected on MH agar containing 10 µg/ml of kanamycin . Inactivation of the mfd gene in the transformants by insertion of the ahpA3 gene was confirmed by PCR using primers mfd-F2 and mfd-R2 ( Fig . 1B ) . The mfd mutant of NCTC 11168 was named JH01 . The entire mfd gene including its putative ribosome binding site was amplified from strain 11168 by PCR using primers mfd-F5 ( 5′-CGCTTCCGCGGAACTAGTAAAATTAAAGAAGATACTATC-3′ ) and mfd-R3 ( 5′-GGCTTTAAATAATCTTTTCGAGCTCTATAAATT-3′ ) . The underlined sequences in the primers indicate the restriction sites for SacI and SacII , respectively . The PCR product was digested with SacI and SacII , and was then cloned into the plasmid construct pRY112-pABC to generate pRY112-mfd , in which the mfd gene was fused to the promoter of cmeABC . pRY112-pABC was made by cloning the promoter sequence of cmeABC [74] to shuttle plasmid pRY112 [75] . The promoter DNA of cmeABC was amplified by primers BSF ( 5′ AAAAGGATCCTAAATGGAATCAATAG 3′ ) and AR2 ( 5′ TGATCTAGATCATACCGAGA 3′ ) , digested with BamHI and XbaI , and cloned into pRY112 . There were two reasons that we used the promoter of cmeABC in the expression of mfd . First , the 5′ end of mfd overlaps with its upstream gene and the native promoter for mfd was unknown . Second , the promoter of cmeABC is moderately active in Campylobacter [74] , preventing over- or under-expression of mfd . The constructed plasmid pRY112-mfd was sequenced and confirmed that no mutations in the cloned sequence occurred . For complementation , the shuttle plasmid pRY112-mfd was transferred into JH01 by conjugation . The complemented strain was named JH02 . Limited passage of JH02 in MH broth without antibiotics indicated that the complementing plasmid was stable in the construct ( data not shown ) . The shuttle plasmid carrying the mfd gene was also transferred to wild-type 11168 to generate strain JH03 for overexpression of the mfd gene . To compare the growth kinetics of the mfd mutant with that of the wild-type , a fresh culture of each strain was inoculated into MH broth ( initial cell density of OD600 = 0 . 05 ) and the cultures were incubated at 42°C under microaerobic conditions . To determine if the mutation affects C . jejuni growth with a subinhibitory concentration of CIPRO , the various strains were grown in MH broth with 0 . 06 µg/ml of CIPRO ( 0 . 5× the MIC ) . Culture samples were collected and measured for OD600 at 0 , 3 , 6 , 12 , 24 and 48 hours post-inoculation . The minimum inhibitory concentration ( MIC ) of CIPRO was determined by using E-test strips ( AB Biodisk , Solna , Sweden ) as described in the manufacturer's instructions . The detection limit of the E-test for CIPRO was 32 µg/ml . The MICs of erythromycin , ampicillin and streptomycin for C . jejuni NCTC 11168 , JH01 , JH02 , and JH03 were determined using a standard microtiter broth dilution method described previously [24] . Each MIC test was repeated at least three times to confirm the reproducibility of the MIC patterns . The antibiotics used in this study were purchased from Sigma Chemical Co . ( erythromycin , ampicillin , streptomycin ) or ICN Biomedicals Inc . ( CIPRO ) . Wild-type 11168 , JH01 , JH02 and JH03 were compared for the spontaneous mutation rates to CIPRO resistance . In each experiment , each of the 4 strains was inoculated into three flasks , each of which contained 30 ml of antibiotic-free MH broth . The cultures were incubated to the mid logarithmic phase ( OD600≈0 . 15 ) under microaerobic conditions . The culture in each flask was collected by centrifugation and resuspended in 1 ml of MH broth . The total CFU in each culture was measured by serial dilutions and plating on MH agar plates , while the number of FQR mutants was detected using MH agar plates containing 1 , 2 or 4 µg/ml CIPRO . The frequency of emergence of FQR mutants was calculated as the ratio of the CFU on CIPRO-containing MH agar plates to the CFU on antibiotic-free MH agar plates after 2 days of incubation at 42°C under microaerobic conditions . This experiment was repeated five times . The mutation frequency data were log-transformed for statistical analysis . One-Way ANOVA followed by Tukey test was used to determine the significance of differences in the levels of spontaneous mutation rates among the strains . The data were also analyzed by the Wilcoxon rank-sum test to allow for non-normality . For the comparisons discussed in Results , the conclusion of the two tests was the same at significance level 0 . 05 . Representative FQR colonies were selected for determination of the point mutations in gyrA . The QRDR of gyrA was amplified by PCR using primer pair GyrAF1 ( 5′-CAACTGGTTCTAGCCTTTTG-3′ ) and GyrAR1 ( 5′-AATTTCACTCATAGCCTCACG-3′ ) [76] . The amplified PCR products were purified with the QIAquick PCR purification kit ( Qiagen ) prior to sequence determination . DNA sequence analysis was carried out using an automated ABI Prism 377 sequencer ( Applied Biosystems , Foster City , CA , USA ) and analyzed by the Omiga 2 . 0 ( Oxford Molecular Group ) sequencing analysis software . To determine if Mfd affects the development of FQR mutants under treatment with CIPRO , wild-type 11168 and JH01 were treated in MH broth with 4 µg/ml ( 32× the MIC ) of CIPRO . Wild-type 11168 and JH01 were grown on antibiotic-free MH agar plates under microaerobic conditions . After 20 hours of incubation , the cells were collected and resuspended in MH broth for inoculation . Three treatment experiments were conducted using three different initial cell densities . In experiment 1 , each strain was inoculated into 3 100-ml flasks with MH broth containing 4 µg/ml of CIPRO and the initial cell density was 107 CFU/ml . The cultures were incubated microaerobically at 42°C . Aliquots of the cultures were collected at different time points ( 0 , 1 , 2 , 3 days post-inoculation ) and plated onto regular MH plates for enumeration of the total bacterial number and onto MH plates containing 4 µg/ml of CIPRO for counting FQR colonies . In experiments 2 and 3 , the cultures were treated in the same way , but the initial cell densities were 106 and 103 CFU/ml , respectively . Experiment 1 was repeated three times , while experiments 2 and 3 were each repeated twice . To determine if the insertional mutation in mfd affected the expression of the downstream gene Cj1084c ( encoding a possible ATP/GTP-binding protein ) , RT-PCR was performed to assess the expression of Cj1084c . Total RNA was isolated from C . jejuni 11168 and JH01 using the RNeasy Kit ( Qiagen ) . The purified RNA samples were treated with On-Column DNase Digestion Kit ( Qiagen ) followed by further treatments with DNase to remove DNA contamination . The Cj1084c-specific primers Cj1084cF ( 5′ TTG CCT TAG CAG ATA TCA T 3′ ) and Cj1084cR ( 5′ ACC ACT TCT ACT TGC TCT TA 3′ ) were used to amplify a 430 bp region of the gene in a conventional one-step RT-PCR by using the SuperScript III One-Step RT-PCR kit ( Invitrogen ) . An RT-PCR mixture lacking the RT was included as a negative control . To examine if Mfd plays a role in the emergence of FQR Campylobacter during in vivo FQ treatment , a chicken experiment was performed using 11168 and JH01 . Day-old broiler chickens ( Ross×Cobb ) were obtained from a commercial hatchery and randomly assigned to 2 groups ( 11 birds per group ) . Each group of chickens was maintained in a sanitized wire-floored cage . Feed and water were provided ad libitum . Prior to inoculation with Campylobacter , the birds were tested negative for Campylobacter by culturing cloacal swabs . At day 3 of age , the two groups of chickens were inoculated with 11168 and JH01 , respectively , at a dose of 106 CFU/chick via oral gavage . Six days after the inoculation , the birds were treated with 50 ppm enrofloxacin . The treatment was administered in drinking water and lasted for five consecutive days . During the treatment , only medicated water was given to the birds to ensure enough consumption . Cloacal swabs were collected periodically before and after enrofloxacin treatment until the end of the experiment . Each swab was serially diluted in MH broth and plated onto two different types of MH plates: one containing Campylobacter-specific growth supplements ( SR 084E and SR117 E; Oxoid Ltd . , Basingstoke , England ) for the enumeration of total Campylobacter cells and the other containing 4 µg/ml of CIPRO in addition to the same selective agents and supplements to recover FQR Campylobacter in each chicken . At each sampling time , at least one Campylobacter colony from each chicken were selected from the regular MH agar plates ( no CIPRO ) for the determination of CIPRO MICs using the E-test ( AB Biodisk ) . The colonization data ( CFU/g feces ) were log-transformed and used for statistical analysis . The significance of differences in the level of colonization between the two groups was determined using Student's t-test , Welch's t-test to allow for non-constant variation across treatment groups , and the Wilcoxon rank-sum test to allow for non-normality . The conclusion of all three tests was the same at significance level 0 . 05 . The microarray data have been deposited in the NCBI Gene Expression Omnibus ( GEO; http://www . ncbi . nlm . nih . gov/geo/ ) database and the accession number is GSE10471 . | As a food-borne bacterial pathogen , Campylobacter jejuni is a common causative agent of gastrointestinal illnesses in humans . Development of antibiotic resistance in Campylobacter , especially to fluoroquinolone ( a broad-spectrum antimicrobial ) , compromises clinical treatments and presents a major public health threat . It is not well understood why Campylobacter is highly adaptable to fluoroquinolone treatment or how it acquires mutations associated with fluoroquinolone resistance . Understanding the molecular mechanisms involved in the resistance development will help us to reduce the emergence of fluoroquinolone-resistant Campylobacter . Using DNA microarray and other molecular methods , as well as animal studies , we uncovered the key role of Mfd in promoting spontaneous mutations and development of fluoroquinolone resistance in Campylobacter . Mfd is a transcription-repair coupling factor involved in DNA repair and was not previously known for its role in promoting mutations conferring antibiotic resistance . Our findings not only reveal a novel function of Mfd , but also provide a potential molecular target for reducing the emergence of fluoroquinolone-resistant Campylobacter . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/bacterial",
"infections",
"infectious",
"diseases/gastrointestinal",
"infections",
"infectious",
"diseases/antimicrobials",
"and",
"drug",
"resistance",
"microbiology"
] | 2008 | Key Role of Mfd in the Development of Fluoroquinolone Resistance in Campylobacter jejuni |
A proportion of all immunocompetent patients treated for visceral leishmaniasis ( VL ) are known to relapse; however , the risk factors for relapse are not well understood . With the support of the Rajendra Memorial Research Institute ( RMRI ) , Médecins Sans Frontières ( MSF ) implemented a program in Bihar , India , using intravenous liposomal amphotericin B ( Ambisome ) as a first-line treatment for VL . The aim of this study was to identify risk factors for VL relapse by examining the characteristics of immunocompetent patients who relapsed following this regimen . This is an observational retrospective cohort study of all VL patients treated by the MSF program from July 2007 to August 2012 . Intravenous Ambisome was administered to 8749 patients with VL in four doses of 5 mg/kg ( for a total dose of 20 mg/kg ) over 4–10 days , depending on the severity of disease . Out of 8588 patients not known to be HIV-positive , 8537 ( 99 . 4% ) were discharged as initial cures , 24 ( 0 . 3% ) defaulted , and 27 ( 0 . 3% ) died during or immediately after treatment . In total , 1 . 4% ( n = 119 ) of the initial cured patients re-attended the programme with parasitologically confirmed VL relapse , with a median time to relapse of 10 . 1 months . Male sex , age <5 years and ≥45 years , a decrease in spleen size at time of discharge of ≤0 . 5 cm/day , and a shorter duration of symptoms prior to seeking treatment were significantly associated with relapse . Spleen size at admission , hemoglobin level , nutritional status , and previous history of relapse were not associated with relapse . This is the largest cohort of VL patients treated with Ambisome worldwide . The risk factors for relapse included male sex , age <5 and ≥45 years , a smaller decrease in splenomegaly at discharge , and a shorter duration of symptoms prior to seeking treatment . The majority of relapses in this cohort occurred 6–12 months following treatment , suggesting that a 1-year follow-up is appropriate in future studies .
Visceral leishmaniasis ( VL ) is a neglected tropical disease that results in the loss of an estimated 1 million disability-adjusted life years annually in South East Asia [1]; it is typically fatal if untreated . VL predominantly affects the poorest strata of society and those with limited access to care [2] . The incidence is estimated to be between 146 , 700 and 282 , 800 cases per year [3] . Fifty percent of VL cases worldwide occur in India , and up to 90% of these in the state of Bihar . Although complete parasite clearance is rarely achieved , it is thought that patients with competent immune systems who are successfully treated develop an effective lifelong cellular immune response that suppresses residual parasite growth [4] . High relapse rates in HIV-positive patients have been previously described [5] , [6] . However , in nearly all studies assessing treatment effectiveness , a proportion of immunocompetent patients relapse following treatment despite negative end-of-treatment test-of-cure results . Typically , these relapses occur within 6 months of initial treatment with later recurrence considered rare [7] . Very little is known regarding the characteristics of immunocompetent patients with VL who relapse [8] , [9] , particularly in the Indian context . The aim of this observational retrospective cohort study was to identify risk factors for relapse in immunocompetent patients who had been treated with 20 mg/kg Ambisome for their primary episode of VL . With the permission of the State Health Society of Bihar and the support of the specialist VL research institute the Rajendra Memorial Research Institute ( RMRI ) , Médecins Sans Frontières ( MSF ) has been treating patients with VL in Vaishali district since 2007 , in coordination with the National Vector Borne Disease Control Programme of India . Between July 2007 and August 2012 , a total of 8749 patients diagnosed with VL were treated with intravenous 20 mg/kg liposomal amphotericin B ( Ambisome; Gilead Pharmaceuticals , Foster City , CA , USA ) . This regimen has been shown to have a 6-month cure rate of 98% in the Indian context [10] . Ambisome is a brand name for Liposomal Amphotericin B . There are a number of preparations of Liposomal amphotericin B available on the market; however due to the lack of standard and widely applicable regulations or guidance for liposomal technology , it is important that this specific preparation be named . At time of publication , none of the rival preparations have undergone peer reviewed non-inferiority studies against Ambisome nor received stringent regulatory approval for use in VL . It is for this reason that MSF and the WHO currently only use Ambisome rather than other preparations . However it is urgent that clear regulatory guidelines for endemic countries be established by a normative setting organisation like the WHO and other existing formulations be formally evaluated [11] . Using the data routinely collected from the MSF program , we determined the demographic and clinical characteristics of 119 immunocompetent patients who presented back to the program with parasitologically confirmed VL relapse . We then identified possible risk factors for relapse by comparing these patients to the 8418 patients who were discharged as cured and were not known to have relapsed .
This analysis met the Médecins Sans Frontières Institutional Ethics Review Committee's criteria for a study involving the analysis of routinely collected program data . Although a new treatment in the Indian setting , the programme utilised a recognised treatment for VL and was run in coordination with the State Health Society through a memorandum of understanding , which is the usual procedure for NGOs operating in this context . All electronic data were analysed anonymously .
Male patients had higher odds of relapsing ( unadjusted odds ratio [uOR] 1 . 8; 95% CI 1 . 2–2 . 6 ) compared with female patients . Patients aged <5 years ( uOR 3 . 6; 95% CI 1 . 8–7 . 2 ) and ≥45 years ( uOR 2 . 2; 95% CI 1 . 2–1 . 4 ) were more likely to relapse than patients aged ≥15 to <30 years . Factors not associated with relapse in the univariate analysis ( p>0 . 05 ) included: caste; living in an area where MSF was conducting information , education , and communication activities as well as supporting the primary health center; a history of previous relapse; location of treatment administration ( treatment camp or primary health center/hospital ) ; and season or year of treatment . According to the univariate analysis , patients reporting a shorter duration of symptoms prior to seeking treatment had higher odds of relapsing . Compared with the baseline group of patients who received treatment ≤4 weeks after developing symptoms ( also known as time to presentation ) , the odds of relapse progressively decreased as the duration of symptoms prior to seeking treatment increased , from 0 . 6 ( 95% CI 0 . 4–0 . 97 ) for those presenting >4 to ≤8 weeks after symptoms occurred to 0 . 4 ( 95% CI 0 . 2–0 . 8 ) for those presenting >8 weeks after symptoms occurred . Additionally , patients who exhibited a decrease in spleen size of ≤0 . 5 cm/day by the time of discharge appeared to have higher odds of relapse ( 1 . 7; 95% CI 1 . 1–2 . 5 ) compared with those who exhibited a decrease in spleen size of >0 . 5 cm/day . No other clinical factors were significantly associated with risk of relapse ( Table 2 ) . Notably , nutritional status , spleen size and Hb level upon admission , and duration of treatment were not predictive of relapse . A multivariate logistic regression model was developed for those variables that were shown to be significant by univariate analysis ( p<0 . 05 ) ( Table 3 ) . Variables that were justified a priori or were associated with relapse in other studies [8] , [9] were also included in the multivariate analysis . These additional variables , which included nutritional status and spleen size and Hb levels upon admission , were added step-wise to the model . However , the variables associated with relapse as determined by the univariate analysis remained significant in the multivariate analysis , and those that were non-significant in the univariate analysis did not attain significance in the multivariate analysis .
This cohort represents the largest number of patients with VL treated with Ambisome both worldwide and on the Indian subcontinent to date . Although based only in one district , this program has treated an estimated 5 . 8% of all reported VL cases in India between 2008 and 2011 [14] . The present study is also the only India-based study that specifically examines risk factors for and characteristics of relapse in immunocompetent patients , and describes the distribution of VL relapses >6 months after treatment . It is , therefore , of particular interest considering the move towards lower dose Ambisome as the first-line therapy for VL on the Indian subcontinent [7] . A strength of this study is the robust database that has been maintained throughout the program and has minimal missing data . A limitation is that all patients were not followed up prospectively to determine relapse status , and as such the identification of relapses depended on patients returning to the programme for assessment if their symptoms recurred . This may result in an underestimation of the number of relapses to the 20 mg/kg regimen . There are limited data available regarding VL relapses in immunocompetent patients , and risk factors for relapse appear to vary from country to country . A retrospective study of 300 VL patients treated with meglutamine antimoniate in Georgia between 2002 and 2004 identified 21 cases of relapse . Among these cases , age <1 year , time to treatment of >90 days , and Hb levels of <6 g/dL were associated with relapse [8] . However , it is unclear whether these patients were tested for HIV . No association between relapse and spleen size nor sex was observed . A more recent study examined patient characteristics and drug regimens associated with VL relapse in South Sudan between 1999 and 2007 . The treatment records for 166 patients with VL who presented with relapses were compared with the treatment records for 7924 primary VL patients who did not re-attend with relapse [9] . This study found that larger spleen size upon admission and at the time of discharge were strongly associated with relapse , as was treatment with a short-course combination treatment ( 17 days sodium stibogluconate/paromomycin vs 30 days sodium stibogluconate ) . Age , sex , nutritional status , mobility , and treatment complications were not significantly associated with relapse . The main limitation was missing data , which resulted in the inclusion of only 26 . 7% ( 166/621 ) of the relapses in the analysis . Additionally , HIV testing was not performed for the relapse patients , although the authors considered this unlikely to be a factor for relapse in this group , as the estimated prevalence of co-infection was only 0 . 5% . Our results suggest that age <5 and ≥45 , male sex , a decrease in spleen size of ≤0 . 5 cm/day at discharge , and a short duration of symptoms prior to seeking treatment are risk factors for VL relapse in immunocompetent patients in India . Younger patients may be particularly susceptible to relapse due to the lack of a mature immune system [8] . The increased number of male patients presenting with relapse may be explained by the possibility of limited access to care for females . Indeed , in an analysis of the overall field outcomes of the MSF program in Bihar , the proportion of females admitted to the program progressively decreased in older age groups [Submitted to PLOS NTD] . We are unable to explain the strong inverse correlation that the present study revealed between the duration of symptoms prior to seeking treatment ( time to presentation ) and relapse . Indeed , this correlation is contrary to a priori knowledge , which would predict an association between a longer duration of illness and a more severe clinical presentation , and therefore more serious outcomes . However , other indicators of prolonged illness , such as low Hb level , poor nutritional status and increased splenomegaly at time of admission , were also not associated with relapse . It is possible that , in the Indian context , a rapid presentation to the healthcare provider could itself be an independent indicator of more severe illness or poorer immune status . This association needs to be correlated with other programmes' outcomes and warrants further investigation . In India , the synthetic phospholipid derivative hexdecylphosphocholine ( miltefosine ) is currently recommended by the Indian National Programme as the first line treatment for VL . It is a 28-day oral treatment but its use is limited by teratogenicity , which restricts its use in pregnant and lactating women and requires 3–5 months of contraceptive cover for women of childbearing age [15] . Although miltefosine initially showed promising efficacy and tolerability , recent studies in India [16] , [17] and Nepal [13] have demonstrated relapse rates in immunocompetent patients of between 6 . 8% to 10 . 8% at 6 months respectively , and up to 20 . 0% at 12 months in Nepal . Over the past decade , several studies have examined liposomal preparations of amphotericin B . For doses of Ambisome ≥10 mg/kg , the efficacy at 6 months is >95% [11] . Following a pivotal phase III study published in 2010 [18] , the WHO expert committee on leishmaniasis adopted a single 10 mg/kg dose regimen as the recommended first-line treatment for VL in South East Asia [7] , a strategy that has yet to be introduced by the Indian National Programme . The results from the present study suggest that following treatment with 20 mg/kg Ambisome , risk factors for VL relapse include male sex , age <5 years and ≥45 years , a slower decrease in splenomegaly during treatment , and a shorter duration of symptoms prior to seeking treatment . It also indicates that when using this regimen the majority of relapses occur 6–12 months post-treatment . Recent evidence from another study in Nepal suggests that a significant number of patients relapse 6–12 months post-treatment with miltefosine [13] . Given the move towards treating VL patients with a single 10 mg/kg dose of Ambisome or short-course combination therapies , and the aim of elimination in the Indian subcontinent , we suggest that a 1-year follow-up is essential and should be recommended for all VL treatments . | Visceral leishmaniasis ( VL ) , also known as Kala azar , accounts for the second-highest burden of parasitic disease worldwide . Fifty percent of VL cases worldwide occur in India , and up to 90% of these in the state of Bihar . Up to 10% of VL patients relapse within 6 months of treatment , particularly if co-infected with HIV . Limited data are available regarding relapse in patients with intact immune systems . Between 2007–2012 , with support of the Rajendra Memorial Research Institute ( RMRI ) , Médecins Sans Frontières ( MSF ) treated VL patients in Bihar with 20 mg/kg liposomal amphotericin B ( Ambisome ) . Here we identify risk factors for relapse in immunocompetent patients by comparing the demographics and clinical characteristics of 119 patients testing negative for HIV who experienced parasitologically-confirmed VL relapse against those of the remaining 8418 patients not known to have relapsed . Male sex , age <5 years and ≥45 years , shorter duration of symptoms prior to seeking treatment , and a smaller reduction in spleen size by time of discharge were all risk factors for relapse . The majority of relapses occurred 6–12 months post-treatment with Ambisome . The conventional period of follow-up is 6 months . Considering the aim of elimination in the Indian subcontinent , this data suggests that 1-year follow-up is necessary . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2014 | Risk Factors for Visceral Leishmaniasis Relapse in Immunocompetent Patients following Treatment with 20 mg/kg Liposomal Amphotericin B (Ambisome) in Bihar, India |
5-Aza-2′-deoxycytidine , approved by the FDA for the treatment of myelodysplastic syndrome ( MDS ) , is incorporated into the DNA of dividing cells where it specifically inhibits DNA methylation by forming covalent complexes with the DNA methyltransferases ( DNMTs ) . In an effort to study the correlations between DNA methylation , nucleosome remodeling , and gene reactivation , we investigate the integrated epigenetic events that worked coordinately to reprogram the methylated and closed promoters back to permissive chromatin configurations after 5-Aza-2′-deoxycytidine treatment . The ChIP results indicate that H2A . Z is deposited at promoter regions by the Snf2-related CBP activator protein ( SRCAP ) complex following DNA demethylation . According to our genome-wide expression and DNA methylation profiles , we find that the complete re-activation of silenced genes requires the insertion of the histone variant H2A . Z , which facilitates the acquisition of regions fully depleted of nucleosome as demonstrated by NOMe–seq ( Nucleosome Occupancy Methylome–sequencing ) assay . In contrast , SRCAP–mediated H2A . Z deposition is not required for maintaining the active status of constitutively expressed genes . By combining Hpa II digestion with NOMe–seq assay , we show that hemimethylated DNA , which is generated following drug incorporation , remains occupied by nucleosomes . Our data highlight H2A . Z as a novel and essential factor involved in 5-Aza-2′-deoxycytidine–induced gene reactivation . Furthermore , we elucidate that chromatin remodeling translates the demethylation ability of DNMT inhibitors to their downstream efficacies , suggesting future therapeutic implications for chromatin remodelers .
The eukaryotic genome is compacted into chromatin and associated proteins . The fundamental repeating unit of chromatin is the nucleosome , which contains ∼147 bp of DNA wrapped around a histone protein octamer [1] . However , chromatin conformations change during various cellular processes , such as the cell cycle , transcription or DNA damage [2] . During gene activation , transcription factors compete with chromatin packaging proteins in order to gain access to the DNA sequence and read the genetic information accurately . Accumulated evidence shows that the chromatin architecture of gene promoter regions strongly regulates gene transcription [3] . This chromatin environment might be altered by DNA methylation , post-translational modifications of histone proteins , histone variants and nucleosome positioning [4] . In mammalian cells , ∼60% of gene promoters are located within CpG islands , where cytosine methylation of CpG dinucleotides impairs gene expression . Histone modifications and histone variants are also strongly correlated with transcriptional status [3] . Nucleosome positioning plays an essential role in gene transcriptional regulation according to recent genome wide studies , which show that the majority of active or poised promoters have decreased nucleosome density [5] . Furthermore , the histone variants H2A . Z and H3 . 3 , which are located at specific genome regions such as promoters , enhancers and insulators , work coordinately to destabilize nucleosomes [6]–[8] . The ATP dependent nucleosome remodelers catalyzing H2A . Z incorporation , namely SRCAP and p400 complexes in mammalian cells , have been suggested to be involved in transcriptional regulation , however , the role of H2A . Z remains controversial [9]–[15] . Abnormalities in epigenetic modifications play an essential role in tumorigenesis [16] , and the reversal of them is the basic concept of epigenetic therapy for cancer . DNA methyltransferases ( DNMT inhibitors ) , such as 5-azacytidine ( 5-Aza-CR ) and 5-Aza-2′-deoxycytidine ( 5-Aza-CdR ) , are approved by the FDA for the treatment of MDS [17]–[18] . Although CpG demethylation is the direct and immediate consequence of treatment with DNMT inhibitors ( 5-Aza-CR and 5-Aza-CdR ) [19] , the level of demethylation in tumor suppressor genes does not predict clinical outcome , which suggests that unknown biological processes link the demethylation effects of DNMT inhibitors to their clinical benefits [20] . Several reports have already shown that , in addition to CpG demethylation , DNMT inhibitors indirectly reduce some repressive histone marks , increase acetylation of histone H3 and promote nucleosome depletion upstream of the transcription start sites ( TSS ) [21]–[24] . Here , taking advantage of a high resolution nucleosome positioning assay developed by our laboratory , we further study the integrated epigenetic changes following 5-Aza-CdR induced demethylation . In addition to the rapid enrichment of H3K4me3 at promoter regions , we find that H2A . Z incorporation increases in response to demethylation . Notably , CpG demethylation induced enrichment of H2A . Z and H3K4me3 , as well as nucleosome depletion coordinately constitute a “permissive” chromatin architecture independently of histone acetylation levels . Inhibiting H2A . Z deposition by SRCAP knockdown lessens the establishment of “permissive” promoter environments and ultimately reduces the levels of gene reactivation after 5-Aza-CdR treatment . Genome-wide gene expression and DNA methylation studies further confirm that SRCAP-mediated H2A . Z insertion promotes DNA demethylation induced gene re-expression but has minimal effects on constitutively active genes . Our study reveals an important function of SRCAP/H2A . Z in promoting the reactivation process induced by 5-Aza-CdR but not in maintaining the expression of constitutively active genes and provides an insight to the chromatin structure of hemimethylated DNA .
To investigate the effects of DNA demethylation on chromatin architecture and gene expression , we treated the RKO colon cancer cell line , with 1 uM 5-Aza-CdR for 24 hours and followed the sequential changes in mRNA expression , DNA methylation and histone marks at the promoters of the MLH1 , CDKN2A and MYOD1 , which are methylated and silenced in RKO cells ( Figure 1 ) . MLH1 expression began to rise at D2 after 5-Aza-CdR treatment , reaching a maximal level at D3 and remained constant for 4 days ( Figure 1A ) . We performed a quantitative Methylation-sensitive Single-Nuleotide Primer Extension ( Ms-SNuPE ) assay to detect the DNA methylation changes at the indicated days ( Figure 1B ) . A striking decrease in DNA methylation was observed at D2 ( ∼40% ) . The methylation level at the MLH1 promoter remained nearly constant from D2 to D9 . We then used ChIP to monitor the changes of histone marks after 5-Aza-CdR treatment ( Figure 1C ) . The enrichments of H2A . Z , H3K4me3 and H3K9/14 acetylation ( acH3K9/14 ) were normalized to histone H3 levels to eliminate the potential influence of nucleosome depletion after drug treatment , and the ChIP primers were designed to amplify stable nucleosome regions which located just downstream of TSSs [5] . Interestingly , our results showed that the H2A . Z enrichment significantly changed after 5-Aza-CdR treatment ( p<0 . 001 ) , and could be observed as early as D2 when DNA methylation was substantially decreased ( Figure 1C ) . H3K4me3 increased immediately after treatment and the enrichment of acH3K9/14 modestly increased at D2 and peaked by D3 displaying a similar pattern to the levels of MLH1 expression ( Figure 1A ) . The mRNA levels of CDKN2A rose steadily from D2 after 5-Aza-CdR treatment and peaked at D7 but abruptly dropped at D9 ( Figure 1D ) . Although the active histone marks increased at the promoter of CDKN2A in a manner similar to MLH1 , the H2A . Z level ( p<0 . 05 ) diminished nearly to the baseline at D9 along with a rapid decline of acH3K9/14 from D5 to D9 ( Figure 1F ) . The mRNA level of MYOD1 , a self-regulated gene expressed exclusively in muscle cells , remained undetectable and showed no RNA polymerase II ( pol II ) enrichment after 5-Aza-CdR treatment even though it showed a demethylation pattern similar to that of MLH1 ( Figure 1G , 1H; Figure S1A ) . Most interestingly , we observed modest changes in H2A . Z and H3K4me3 at the MYOD1 promoter , whereas acH3K9/14 remained extremely low , in agreement with its low expression level ( Figure 1I ) . Therefore the MYOD1 promoter acquired a “permissive” state for expression after 5-Aza-CdR treatment but the gene was not expressed . In addition to analyzing DNA methylation changes at the single strand level ( Figure 1B , 1E , 1H ) , we detected and quantified asymmetrically methylated DNA after 5-Aza-CdR treatment . We performed a hemimethylation Ms-SNuPE assay based on Hpa II digestion as we have developed previously and used the CDKN2A gene as a model ( Figure S1B-S1D ) [25] . The majority of DNA molecules were hemimethylated DNA duplexes at D2 ( ∼60% ) . Even at D5 , ∼5% of the DNA duplexes were composed of hemimethylated DNA molecules . There was a small portion of fully demethylated DNA duplexes at D2 , however , the maximal levels of double strand demethylation ( ∼40% ) was detected at D3–5 when hemimethylated DNA levels dropped . Collectively , our results demonstrate that 5-Aza-CdR treatment eventually produces two types of demethylated DNA duplexes , hemimethylated and fully demethylated DNA . After demethylation H2A . Z and H3K4me3 are deposited in all three promoters . Interestingly , only acH3K9/14 shows a correlation between its enrichment and mRNA expression . We next investigated the effects of 5-Aza-CdR on nucleosome occupancy ( Figure 2 , Figure 3 ) . We previously examined the nucleosome occupancy status of the MLH1 promoter in RKO and LD419 cells using MNase-ChIP assay and Methylase-based Single-Promoter Analysis assay ( MSPA ) [22] . We confirmed our previous findings using a recently developed high resolution assay , which uses the GpC methyltransferase ( M . CviPI ) instead of CpG methyltransferase ( M . SssI ) to methylate GpC sites that are not occupied by nucleosomes or tightly bound transcription factors [26] . By analyzing the methylation status of GpC sites , this NOMe-seq ( Nucleosome Occupancy Methylome-sequencing ) assay provides a digital footprint of nucleosome occupancy and allows the study of nucleosome positioning in both CpG islands and CpG poor regions regardless of their CpG methylation status . Nucleosome occupancy at the MLH1 and CDKN2A promoters in LD419 and RKO cells were analyzed by NOMe-seq using PCR primers lacking CpG or GpC sites to avoid complications due to cytosine methylation in the parental molecules ( Figure 2 ) . Both promoters were unmethylated in LD419 cells and had clear nucleosome depleted regions ( NDRs ) which were accessible to the exogenous M . CviPI . In contrast , the promoters of MLH1 and CDKN2A were methylated in RKO cells and were inaccessible to the GpC methyltransferase , indicating that hypermethylated promoters were fully occupied by nucleosomes . To investigate the nucleosome occupancy changes accompanying drug induced DNA demethylation , we used primers which were specifically designed to amplify the DNA strands which had become demethylated at CpG sites and studied the accessibility of these demethylated molecules to M . CviPI ( Figure 3; Figure S2 ) . The maximum demethylation of CpG sites was observed at D2 ( Figure 1B ) , but only one out of twenty-five demethylated DNA strands ( 4% ) had a nucleosome depleted region larger than 146 bp around the TSS ( green bar on the graph ) ( Figure 3A ) . At D3 , a proportion of unmethylated DNA strands at the gene promoters ( 24% ) showed a nucleosome depleted area large enough to accommodate at least one nucleosome , consistent with the presence of active histone marks and increased gene expression . Extensive nucleosome depletion ( 40% ) , the H2A . Z enrichment and gene reactivation reached a maximal level at D5 ( Figure 3A ) . NOMe-seq analysis of the CDKN2A promoter yielded similar results and showed depletion of the −1 nucleosome at D5 ( 12% ) ( Figure 3B ) . Interestingly , the MYOD1 promoter showed drug-induced enrichment of H2A . Z and H3K4me3 as well as nucleosome depletion around the TSS ( 20% ) without MYOD1 expression ( Figure 3C ) . Therefore , changes in histone modifications and nucleosome depletion were the direct consequences of DNA demethylation and did not require transcriptional activation for some genes , such as MYOD1 . Although the nucleosome occupancy at the MLH1 promoter was dramatically decreased on the demethylated DNA strands at D5 after 5-Aza-CdR treatment , a portion of demethylated DNA strands remained inaccessible to M . CviPI ( Figure 3A ) . As shown in Figure S1C , the majority of demethylated DNA strands were associated with hemimethylated DNA duplexes at D2 . And the demethylated DNA strands at the D2 were highly occupied by nucleosomes . To further investigate the nucleosome occupancy on hemimethylated DNA , we pre-digested the NOMe-seq DNA samples before bisulfite treatment with an excess of Hpa II . Demethylated DNA strands associated with symmetrically demethylated DNA duplexes are destroyed by Hpa II digestion , whereas the demethylated strands in hemimethylated DNA duplexes are resistant to digestion ( Figure S1B ) . Next , we used the same PCR primers as shown previously to amplify the remaining demethylated DNA strands . The NOMe-seq results from Hpa II digested DNA clearly showed that the promoters of MLH1 and CDKN2A were occupied by nucleosomes when the underlying DNA was hemimethylated ( Figure 3D , 3E ) . Our results show that DNA demethylation at promoter regions induces substantial changes in nucleosome occupancy which only occurs on symmetrically demethylated but not hemimethylated DNA . The enrichment of H3K4me3 after demethylation has been well studied [27] , however the role of H2A . Z insertion in gene reactivation is unclear . Thus , we explored the potential role of H2A . Z in 5-Aza-CdR induced gene reactivation by knocking down SRCAP ( Figure S3A ) , which catalyzed H2A . Z deposition in a cell cycle independent manner [28] , and subsequently treating the cells with 5-Aza-CdR ( Figure 4A ) . The expression of MLH1 and CDKN2A was strongly attenuated by SRCAP knockdown , concomitant with a reduction of H2A . Z levels at the promoters ( Figure 4A ) . The enrichment of acH3K9/14 at the reactivated promoters was reduced after SRCAP knockdown as well . Interestingly , the 5-Aza-CdR induced H2A . Z deposition was also inhibited by SRCAP knockdown at the MYOD1 promoter , but the gene expression and acH3K9/14 levels remained undetectable as expected . Of note , SRCAP knockdown showed minimal effects on the H3K4me3 levels at all three promoters , suggesting that the H3K4me3 mark was independent of H2A . Z levels . In addition , knockdown of SRCAP did not affect DNA methylation levels at promoters of these examined genes ( Figure S3B ) . In contrast , the mRNA level of GRP78 , which is usually over-expressed in cancer cells and has enriched H2A . Z around its promoter [2] , was not reduced and even modestly increased after SRCAP knockdown . The H2A . Z level at the GRP78 promoter dramatically dropped by nearly 90% compared with non-target ( NC ) siRNA treated cells . Meanwhile , the levels of H3K4me3 and acH3K9/14 remained high showing that these marks were independent of H2A . Z levels . In addition to GRP78 , we analyzed two more genes , LAMB3 and G3BP , both of which were unmethylated and expressed in RKO cells ( Figure S3D , S3E ) . The enrichment of H2A . Z , which has been identified at both promoters [10] , [29] , was reduced by SRCAP knockdown , but neither the mRNA expression nor the histone marks were significantly affected . Remarkably , knockdown of SRCAP in LD419 cells did not inhibit the expression of MLH1 and CDKN2A , though the H2A . Z enrichment at the promoter regions had been reduced . Again , no difference in H3K4me3 and acH3K9/14 levels was detected after SRCAP siRNA treatment ( Figure S4A ) . We next investigated the function of SRCAP-mediated H2A . Z deposition on 5-Aza-CdR induced nucleosome occupancy changes . Substantial nucleosome depletion at the MLH1 promoter was observed on the demethylated DNA strands in the NC siRNA treated cells as previously shown ( Figure 4B ) . When SRCAP-mediated H2A . Z incorporation was inhibited , nucleosome depletion was much curtailed at the promoter ( 32% to 20% ) . Similarly , we observed that inhibiting SRCAP-mediated H2A . Z deposition prevented the depletion of nucleosome induced by the 5-Aza-CdR in the vicinity of the CDKN2A ( 20% to 4% ) and MYOD1 ( 16% to 8% ) TSSs . In contrast , the NDRs at the GRP78 , LAMB3 and G3BP promoters were not reduced and even showed modest increases ( Figure S3C , S3F , and S3G ) . Similarly , the NDRs upstream of the TSS of MLH1 did not change after SRCAP knockdown in LD419 cells ( Figure S4B ) . Taken together , these results demonstrate that SRCAP-mediated H2A . Z deposition and associated nucleosome depletion play a key role in re-constructing a poised chromatin architecture around demethylated promoters . In contrast , continued H2A . Z presence is not critical in maintaining an open chromatin environment of actively transcribed genes . To elucidate the importance of H2A . Z deposition for gene reactivation following 5-Aza-CdR induced demethylation , we conducted genome-wide studies to assay global DNA methylation and gene expression changes after drug treatment . We interrogated global promoter DNA methylation patterns using the Infinium HumanMethylation27 platform , which includes 27 , 578 CpG dinucleotides spanning 14 , 495 well-annotated , unique gene promoter and/or 5′ gene regions ( from −1 , 500 to +1 , 500 from the TSS ) . The DNA methylation level for each interrogated CpG site is reported as a beta value , ranging from zero ( low DNA methylation ) to one ( high DNA methylation ) . In NC siRNA treated cells , the CpG sites could be roughly separated into two groups based on the bimodal distribution of the beta values: a hypomethylated group ( beta value<0 . 2 ) and a hypermethylated group ( beta value>0 . 8 ) [30] ( Figure 5A ) . Consistent with the Ms-SNuPE results ( Figure 1B , 1E , 1H ) , the CpG sites within the promoters of MLH1 , CDKN2A and MYOD1 had beta values higher than 0 . 8 . The peak representing hypermethylated probes was notably shifted towards the left after 5-Aza-CdR treatment . To further expand this observation , CpG probes were plotted between 5-Aza-CdR and PBS treatment ( control ) in NC siRNA treated cells ( Figure 5B ) . Using a beta value difference of 0 . 25 as a threshold for differential DNA methylation and separating CpG probes based on beta values>0 . 8 associated with control treatment , 2 , 638 CpG probes ( 1278 genes ) were identified to be demethylated by 5-Aza-CdR in NC siRNA treated cells and knockdown of SRCAP only showed little effect on DNA methylation patterns , as 2 , 515 CpG probes ( 1208 genes ) were found to be demethylated in SRCAP siRNA treated cells ( Figure 5C; Figure S5A , S5B ) . To evaluate gene expression changes , we performed a permutation analysis ( 1 , 000 permutations ) using Significance Analysis of Microarrays ( SAM ) in NC or SRCAP siRNA transfected cells after 5-Aza-CdR treatment [31] . The identification of differentially expressed genes was performed among the indicated groups ( Figure 5D–5F; Figure S5C , S5D ) . We found that SRCAP knockdown had minimal impact on global gene expression ( Figure 5E ) , while 5-Aza-CdR treatment significantly up-regulated 97 genes ( representing 130 different transcripts ) in NC treated cells and 86 genes ( representing 105 different transcripts ) in SRCAP siRNA treated cells , with an 81% overlap between two groups ( Figure 5D; Figure S5E , S5F ) . We did not observe any gene significantly down-regulated by 5-Aza-CdR treatment in either NC or SRCAP siRNA treated cells . To visualize the global gene expression difference between SRCAP and NC siRNA , we plotted the observed log2 fold change for all the interrogated transcripts on the platform ( Figure 5F ) . After extracting the promoter DNA methylation beta values of 97 genes , which were reactivated by 5-Aza-CdR in NC siRNA treated cells from the Infinium array , we found that 44 genes ( representing 92 different CpG loci ) , including MLH1 and CDKN2A , had beta value differences greater than 0 . 2 as shown in the heatmap and box plot ( Figure 5G , 5H ) . We next concentrated on these 44 genes which were demethylated and subsequently reactivated . Within this group of genes , knockdown of SRCAP significantly inhibited the reactivation of some transcripts such as EPM2AIP1 from up-regulated to non-responsive . Although the majority of genes that were up-regulated by 5-Aza-CdR in NC siRNA treated cells were still induced in SRCAP siRNA treated cells , the reactivation levels were strikingly decreased ( red circles ) . The fold changes of the 44 genes induced by demethylation were calculated in the inserted box plot , further showing the significant effects of SRCAP knockdown ( p<2×10−16 ) . To validate this genome-wide analysis , we randomly selected four candidate genes ( CHFR , CTCFL , SYCP3 and EPM2AIP1 ) from the pool of 44 genes and analyzed the expression changes ( Figure S6A–S6C ) . The reactivation levels of four methylated genes were significantly suppressed by SRCAP knockdown . We then validated the histone marks changes on the promoters of CHFR and SYCP3 . We found that SRCAP knockdown prevented H2A . Z deposition as well as diminished gene reactivation , which was consistent with our results from MLH1 and CDKN2A . To confirm that the H2A . Z insertion was causing the observed effects , we knocked down p400 , which is a homolog of SWR1 and has been identified as another key player in H2A . Z deposition [13] . We found that inhibiting p400 could also reduce the reactivation of MLH1 and CDKN2A in RKO cells ( Figure S6D ) . In addition to knocking down the two catalytic subunits of SRCAP and p400 complexes , inhibition of YL-1 , the binding partner of H2A . Z in the SRCAP complex [32]–[33] , also suppressed 5-Aza-CdR induced gene reactivation ( 6E ) and depletion of the −1 nucleosome ( Figure S6F ) . Our integrated study reveals that 5-Aza-CdR robustly reduces global promoter DNA methylation levels , and subsequently reactivates gene expression . Decreasing SRCAP expression inhibits global gene reactivation but has no effect on DNA methylation at promoter . However the maintenance of active gene expression might not require highly enriched H2A . Z .
Although recent studies have begun to explore the epigenetic factors involved in the 5-Aza-CdR mediated demethylation process [21] , [24] , our study focuses on the dynamic changes in chromatin architecture immediately after 5-Aza-CdR treatment ( Figure 6 ) . We demonstrate that removing DNA methylation rapidly induces H2A . Z incorporation , which confirms the antagonistic relationship between H2A . Z and DNA methylation observed in genome-scale studies of arabidopsis , human breast tissue and tumorigenesis of a B-cell lymphoma model in mouse [34]–[36] . Although some reports show that loss of pie1 in Arabidopsis or H2A . Z in mammals increases DNA methylation levels at gene body regions [34] , [37] , our data demonstrate that DNA methylation levels at promoter regions are not affected by transiently inhibiting H2A . Z insertion . Previous reports have showed positive correlations between H2A . Z insertion and the expression of CDKN1A , estrogen receptor target genes , muscle differentiation-specific genes and PcG protein target genes in ES cells [11] , [13] , [33] . Our genome-wide expression results demonstrate that SRCAP-mediated H2A . Z deposition at promoter regions is necessary for complete gene reactivation induced by DNA demethylation . We show that inhibition of SRCAP-mediated H2A . Z insertion prevents nucleosome depletion at the promoters of MLH1 and CDKN2A after 5-Aza-CdR treatment . In addition , knockdown of YL-1 , the binding partner of H2A . Z in the SRCAP complex , also reduces the CDKN2A gene reactivation and nucleosome depletion around the TSS region that are induced by 5-Aza-CdR treatment . Collectively , our data provides evidence for the hypothesis that SRCAP/H2A . Z directly promotes transcription by reducing nucleosome occupancy at promoter regions [38] . Nevertheless , H2A . Z enrichment is necessary but not sufficient for gene reactivation according to our data . As shown at the MYOD1 promoter , the modest enrichment of H2A . Z contributes to the establishment of a “permissive” environment regardless of the subsequent gene reactivation status . The reduction of M . CviPI accessibility at the MYOD1 promoter after SRCAP knockdown suggests that H2A . Z mediated nucleosome depletion is not the consequence of gene expression and might be an early event in transcription initiation . In addition , Hardy et al [15] showed that H2A . Z was recruited to the promoter regions prior to pol II binding . We observed pol II enrichment at the MLH1 promoter but not at the MYOD1 promoter , though both promoters have been remodeled structurally , which suggested the formation of “permissive” promoter regions might not require the presence of pol II at an early stage . Unlike H2A . Z , histone H3 acetylation occurs concomitantly with gene expression , and is not required for establishing this early “permissive” promoter . Furthermore , the data from the SRCAP knockdown experiments strongly indicate that H2A . Z incorporation , especially SRCAP mediated deposition , is independent of H3K9/K14 acetylation . Functional studies of the Tip48/49 complex , which shares some components of the SRCAP complex , showed that acetylation of H2A enhanced H2A . Z insertion [39] . However , a recently published report demonstrated that inhibiting NF-Y , one of the proteins with highly similarity to core histones , prevented H2A . Z deposition at promoter regions but had no observable effect on histone H3K9/14 acetylation , indirectly supporting our conclusions [40] . Similarly to the behavior of H2A . Z , H3K4me3 enriches at promoters following demethylation , in agreement with the reported mutually exclusive relationship between H3K4me3 and DNA methylation [41] . Although many reports show that H3K4me3 levels are correlated with gene expression status , Thomson et al [42] demonstrated that artificially inserted promoter-less DNA sequences containing unmethylated CpG sites were sufficient to acquire H3K4me3 . In our study , DNA demethylation associated H3K4me3 enrichment creates a “permissive” promoter configuration; however , such a configuration is not sufficient for gene activation . The presence of key transcription factors is also necessary [29] . In yeast , Set1 , an H3K4 methyltransferase , and H2A . Z have redundant functions in preventing the spread of Sir-mediated silencing , indicating that the presence of H3K4 methylation marks and H2A . Z are not dependent on each other [43] . Interestingly , our results show that the enrichment of H3K4me3 is not affected after inhibiting SRCAP-mediated H2A . Z deposition and suggest that in mammalian cells the regulation of these histone marks might not depend on each other . Therefore , it would be interesting to study the specific effects of H3K4me3 on chromatin remodeling after DNA demethylation in the future . It has been reported that certain DNA sequences and the binding of pol II or transcription factors influence nucleosome occupancy by different mechanisms [44]–[45] . Recent reports have shown that methylated DNA facilitates nucleosome assembly in vitro , and reciprocally , stable nucleosomes contribute to the establishment and maintenance of DNA methylation [46]–[48] . Here , we apply the now well-established NOMe-seq assay to investigate the correlation between nucleosome positioning and DNA methylation after drug treatment [45] , [49] . Our data demonstrate that promoter regions are highly occupied by nucleosomes when DNA duplexes are either symmetrically methylated or hemimethylated in living cells , suggesting the dominant role of DNA methylation in maintaining stable nucleosomes . Using reconstituted histone octamers and single-stranded M13 constructs , Deobagkar et al show that hemimethylated DNA prevents chromatin expression [50] . Thus complete nucleosome depletion takes place only on symmetrically demethylated DNA after 5-Aza-CdR treatment , which is probably required for full gene activation . Furthermore , our data have confirmed the feasibility of utilizing NOMe-seq in future to investigate the drug induced nucleosome remodeling globally . According to our genome-wide analysis , 5-Aza-CdR induces global demethylation yet only a limited number of genes are significantly reactivated , indicating that CpG demethylation and the subsequent establishment of open chromatin architectures are essential but not sufficient to induce gene reactivation . The “permissive” promoters created by 5-Aza-CdR treatment , such as the MYOD1 promoter , have active histone marks and NDRs , which are needed for assembling of the transcriptional machinery , but other regulatory factors are required to fully reactivate these genes . Therefore , epigenetic modulators which regulate histone marks and nucleosome positioning have strong abilities to promote or impede the pharmacological functions of 5-Aza-CdR . Our results provide a rationale to design clinical trials combining DNMT inhibitors with other anticancer drugs , especially histone deacetylase inhibitors , which facilitate histone acetylation . A comprehensive understanding of the coordinated interplay between epigenetic regulators and 5-Aza-CdR will help explain the drug's clinical outcomes as well as promote the discovery of novel therapeutic targets .
RKO , a colon cancer cell line , was purchased from ATCC and was maintained in MEM medium with 10% FBS . LD419 , a normal human bladder fibroblasts was generated by Dr Louis Dubeau and was maintained in McCoy's 5A supplemented with 20% FBS . RKO cells were plated at 2×106 cells/100-mm dish and treated with 1 µM of 5-Aza-CdR ( Sigma Chemical Co . , St . Louis , MO ) for 24 hours . The NC pool siRNA ( D-001810-10-05 ) and the ON-TARGET plus siRNA targeting SRCAP ( L-004830-00-0005 ) , p400 ( L-021272-01-0005 ) and YL-1 ( VPS72 L-020097-00-0005 ) ( Thermo Fisher Scientific Inc . ) were transfected into RKO cells 24 hours before 5-Aza-CdR treatment using DharmaFECT™ siRNA transfection reagents ( Thermo Fisher Scientific Inc . ) . Total RNA was extracted using RNeasy kit ( Qiagen ) and was converted to cDNA by M-MLV Reverse Transcriptase ( Promega ) using random primer ( Promega ) . The sequences of gene specific primers and taqman-probs are available upon request . With each set of PCR primers , titrations of known amounts of DNA were included as a standard for quantization . DNA methylation level was determined by Ms-SNuPE as described previously [51] . Briefly , CpG sites were interrogated for each promoter . The methylation level of each gene is the average of the three CpG sites examined by Ms-SNuPE . ChIP was performed as described previously [52] . Ten µg of the following antibodies were used: anti-Histone H3 ( Abcam ) , anti-acetylated Histone H3K9/14 ( Milipore ) , anti-H2A . Z ( Abcam ) and anti-RNA polymerase II ( Abcam ) . Five µl of anti-H3K4me3 ( Active Motif ) antibody was used . Ten µg of Rabbit IgG ( Millipore ) was used as a non-specific antibody control . PCR primers are available upon request . Nuclei preparation and GpC Methyltransferase treatment were performed as described previously [26] . Briefly , freshly extracted nuclei were treated with 200 U of GpC methyltransferase for 15 min at 37°C . An equal volume of stopping solution ( 20 nM Tris-HCl , 600 mM NaCl , 1% SDS , 10 mM EDTA ) was added to stop the reaction . The final mixture was incubated at 55°C overnight with 400 µg/ml proteinase K . DNA was isolated and bisulfite converted . The regions of interest were amplified and cloned into pCR 2 . 1-TOPO vector ( Invitrogen ) for DNA sequencing . Hemimethylation analysis was performed as described previously . Undigested or Hpa II-digested DNA from RKO cells before and after treatment was subjected to bisulfite modification . Hpa II digests unmethylated DNA but does not cut a fully or hemimethylated configuration of its CCGG target sequence . Bisulfite-treated DNA was then amplified by PCR using Ms-SNuPE primers that flanked one Hpa II site at the CDKN2A promoter . The equations used to determine hemimethylation levels used were as described previously [25] . The Illumina Infinium DNA methylation assay technology has been described previously [30] . The Infinium DNA methylation assay was performed at the USC Epigenome Center according to the manufacturer's specifications ( Illumina , San Diego , CA ) . The Illumina Infinium DNA methylation assay ( HumanMethylation27_270596_v . 1 . 2 ) examines DNA methylation status of 27 , 578 CpG sites located at promoter regions of 14 , 495 protein-coding genes and 110 microRNAs . Downstream processing and beta value calculations were done as previously described [53] . Expression analysis was performed using the Illumina whole-genome expression BeadChip ( HumanWG-6 v3 . 0 , 48 , 803 transcripts ) ( Illumina ) . The hybridized chips were stained and scanned using the Illumina HD BeadArray scanner ( Illumina ) . Scanned image and bead-level data processing were performed using the BeadStudio 3 . 0 . 1 software ( Illumina ) . The summarized data for each bead type were then processed using the lumi package in Bioconductor [54] . The data were log2 transformed and normalized using Robust Spline Normalization ( RSN ) as implemented in the lumi package . All statistical tests were done using R software ( R version 2 . 12 . 1 , 2010-12-16 , R Development Core Team , 2009 ) . ‘lumi’ package was used to normalize and process gene expression data . ‘samr’ ( version 1 . 28 ) package was used for all permutation tests to access significance of gene expression changes . Differential gene expression ( significance ) change was established for each application by setting the cut-off on a FDR of q = 0 . 05 after applying 1000 permutation . The following CRAN packages were used to generate plots: ‘ggplot2’ and ‘LSD’ ( version 1 . 0 ) . The H2A . Z ChIP results from three biological experiments were analyzed by one way ANOVA using Prism 3 ( GraphPad ) . All summarized probe profile data and processed expression data and DNA methylation data which are used in this study have been deposited to Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/projects/geo/ ) under accession Number GSE26685 . | Epigenetic changes , which include chemical modifications to the DNA and changes in the proteins that package DNA to fit into a cell , play an important role in gene expression regulation . The fact that a number of abnormal epigenetic changes that lead to the silencing of genes occur during tumorigenesis has prompted the design of epigenetic therapies . The ultimate goal of these therapies is to reverse the aberrant epigenetic modifications observed in cancer cells , thereby restoring cells to a “normal” state . 5-Aza-CdDR , a FDA approved drug for MDS treatment , reverses a chemical modification of the DNA resulting in gene reactivation . The data presented here show the importance of H2A . Z , a special DNA packaging protein variant , in the gene reactivation process induced by 5-Aza-CdR . The presence of H2A . Z facilitates the access of proteins at gene regulatory regions , which is a necessary step for gene re-expression . A better understanding of the events that follow 5-Aza-CdR treatment is a necessary step towards the design of combination and/or personalized epigenetic therapies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"oncology",
"medicine",
"gene",
"expression",
"genetics",
"cancer",
"treatment",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | Gene Reactivation by 5-Aza-2′-Deoxycytidine–Induced Demethylation Requires SRCAP–Mediated H2A.Z Insertion to Establish Nucleosome Depleted Regions |
Fascioliasis is an emerging zoonotic disease of considerable veterinary and public health importance . Triclabendazole is the only available drug for treatment . Laboratory studies have documented promising fasciocidal properties of the artemisinins ( e . g . , artemether ) . We carried out two exploratory phase-2 trials to assess the efficacy and safety of oral artemether administered at ( i ) 6×80 mg over 3 consecutive days , and ( ii ) 3×200 mg within 24 h in 36 Fasciola-infected individuals in Egypt . Efficacy was determined by cure rate ( CR ) and egg reduction rate ( ERR ) based on multiple Kato-Katz thick smears before and after drug administration . Patients who remained Fasciola-positive following artemether dosing were treated with single 10 mg/kg oral triclabendazole . In case of treatment failure , triclabendazole was re-administered at 20 mg/kg in two divided doses . CRs achieved with 6×80 mg and 3×200 mg artemether were 35% and 6% , respectively . The corresponding ERRs were 63% and nil , respectively . Artemether was well tolerated . A high efficacy was observed with triclabendazole administered at 10 mg/kg ( 16 patients; CR: 67% , ERR: 94% ) and 20 mg/kg ( 4 patients; CR: 75% , ERR: 96% ) . Artemether , administered at malaria treatment regimens , shows no or only little effect against fascioliasis , and hence does not represent an alternative to triclabendazole . The role of artemether and other artemisinin derivatives as partner drug in combination chemotherapy remains to be elucidated .
Fascioliasis , a zoonotic disease caused by a liver fluke infection of the species Fasciola hepatica and F . gigantica , is of considerable veterinary and public health importance [1] , [2] . Owing to global changes , infections with Fasciola spp . appear to be emerging or re-emerging in several parts of the world [1] . An estimated 91 million people are at risk of fascioliasis , whereas the estimated number of infections shows a large range from 2 . 4 to 17 million [3] . Severe clinical complications in the chronic phase of a Fasciola infection include cholangitis , cholecystitis , jaundice , and biliary colic [1] , [4] . In Egypt , fascioliasis is an important clinical problem , particularly among school-aged children living in rural areas of the Nile Delta [5] , [6] . Prevalence rates of Fasciola infections have been reduced in recent years , explained by control measures put forth by the Egyptian governorates , including triclabendazole administration [6] . Indeed , chemotherapy with triclabendazole , a member of the benzimidazole family of anthelmintics , is the current mainstay for morbidity control of fascioliasis [7] . It should be noted , however that triclabendazole is often difficult to obtain , since it is currently registered in only four countries for human treatment [7] . In addition , resistant fluke populations have been reported from several countries [7]–[9] . Unfortunately , no vaccine is currently available for prevention of fascioliasis [10] . There is a need to develop new fasciocidal drugs . Several studies have documented that the artemisinins ( e . g . , artemether and artesunate ) , which have become the most important antimalarial drugs , particularly when deployed as artemisinin-based combination therapy ( ACT ) [11] , also possess schistosomicidal [12] and fasciocidal activities [13] . Regarding fascioliasis , complete elimination of worms was achieved in rats experimentally infected with adult F . hepatica when artesunate and artemether were administered at single oral doses ( 400 and 200 mg/kg , respectively ) 8 weeks postinfection [14] . Severe tegumental changes and death of flukes occurred when Fasciola spp . were incubated with an artemisinin derivative ( 50–100 µg/ml ) in vitro [14]–[17] . Artesunate and artemether , given by the intramuscular route , yielded high egg and worm burden reductions in natural F . hepatica infections in sheep [18] , [19] . Finally , a study in 100 Vietnamese patients has shown that artesunate might also play a role in the treatment of acute fascioliasis , as patients treated with artesunate were significantly more likely to be free of abdominal pain when compared to triclabendazole-treated patients [20] . The aim of the present study was to assess the efficacy and safety of oral artemether , adhering to two different malaria treatment regimens [21] , [22] , in patients with a chronic Fasciola spp . infection . The study was carried out in a Fasciola-endemic area of Egypt , where Schistosoma mansoni co-exists , but malaria is absent .
Ethical clearance was obtained from the Theodor Bilharz Research Institute ( Giza , Egypt ) , the Ministry of Health and Population ( Cairo , Egypt ) , and the Ethics Committee of Basel , Switzerland ( EKBB , reference no . 54/07 ) . The trial is registered with Current Controlled Trials ( reference no . ISRCTN10372301 ) . Written informed consent was obtained from eligible study participants or parents/legal guardians from individuals aged below 16 years . The study was designed as an interventional , open-label , non-randomized , proof-of-concept trial , consisting of two separate single-arm studies , to evaluate the efficacy and safety of two artemether regimens in the treatment of asymptomatic Fasciola-infected patients . Twenty individuals were assigned to each study , following recommendations for pilot studies of at least 12 patients per treatment [23] and sufficient number of patients who might not comply to follow-up . The primary end points were cure rate ( CR , defined as percentage of patients who became Fasciola egg-negative after treatment , who were egg-positive at study enrollment ) and egg reduction rate ( ERR , defined as reduction of geometric mean ( GM ) egg output after treatment divided by the GM of the same individuals before treatment , multiplied by a factor 100 ) of Fasciola infection , 28 days after the final dosing . Incidence of adverse events , monitored up to 2 days after the final dosing , was used as secondary outcome measure . Paticipants who remained Fasciola positive following artemether treatment were orally treated with a single 10 mg/kg dose of triclabendazole . Efficacy of triclabendazole was determined in the frame of the second intervention study . Patients who were still found with Fasciola eggs in their stool following 10 mg/kg triclabendazole were treated with 20 mg/kg triclabendazole in two divided doses . Study 1 was carried out between April and July 2007 in El-Haddad El-Bahary village , Behera governorate , north-east of Delta . El-Haddad El-Bahary village is s a typical rural setting , with canals fed from the Nile River and no access to the Mediterranean . The total population in the village is 8144 . Study 2 was conducted between August 2008 and May 2010 in Abis village , located south-west of Alexandria . It comprises 10 sub-villages , with an estimated total population of 35 , 000 . Abis village is fed by water canals drawn from the Nile River , with no access to the Mediterranean . Artemether , formulated as 40 mg capsules ( study 1 ) and 50 mg tablets ( study 2 ) was purchased from Kunming Pharmaceutical Cooperation ( Artemidine®; Kunming , People's Republic of China ) . The following two treatment schemes were investigated: ( i ) 6×80 mg over 3 consecutive days ( study 1 ) and ( ii ) 3×200 mg within 24 h ( study 2 ) . Treatment was supervised by a physician with date and precise time of drug administration recorded . Patients were observed for 1 h to ensure retention of medication . In case of vomiting or any treatment-related adverse events , a second dose of artemether was administered . Triclabendazole ( Egaten® 250 mg tablets , scored tablets ) was the product of Novartis ( Basel , Switzerland ) . Patients who failed to become Fasciola egg-negative following artemether administration received 10 mg/kg triclabendazole . The triclabendazole dosage , according to the patients' weight , was calculated in half-tablet increments with a maximum of 2 . 5 tablets ( 625 mg ) . In case of triclabendazole treatment failures ( assessed in study 2 ) , patients were provided two doses of 10 mg/kg of triclabendazole given on subsequent days according to manufacturer's instructions . Several weeks before conducting a parasitological baseline survey , the health directorate of Beheira ( study 1 ) and Alexandria governorate ( study 2 ) were informed about the objectives , procedures , and potential risks and benefits . After written informed consent was obtained , participants were asked to provide a stool sample in order to screen for the presence of F . hepatica and/or F . gigantica eggs . Stool collection containers were labeled with patient's name and a unique identifier ( ID ) . Filled containters were transfered to a laboratory for diagnostic work-up . Two additional stool samples were collected on consecutive days among participants who were found with Fasciola eggs in their feces . In addition , a blood sample was collected before drug administration to examine hematologic parameters , liver , and kidney functions . At enrollment a full clinical examination was carried out to assess participants' general health status . Exclusion criteria were: ( i ) age below 5 years , ( ii ) pregnancy , ( iii ) major systemic illnesses ( e . g . , history of chronic illness such as cancer , diabetes , hypertension , chronic heart , liver or renal disease , severe liver disease of other etiology ) , and ( iv ) recent history of anthelmintic treatment ( e . g . , albendazole , bithionol , dehydroemetine , mebendazole , praziquantel , and triclabendazole taken within the past 4 weeks ) . Patients meeting our inclusion criteria were treated with artemether , which was administered over 3 consecutive days ( study 1 ) or within 24 h ( study 2 ) . Adverse events were monitored on each treatment day and for 24–48 h following the final dosing . Participants were asked to report any potential drug-related signs and symptoms using a standardized questionnaire . Full clinical examinations were performed on all participants . Adverse events were graded ( i . e . , mild , moderate , severe , and serious ) and recorded . Therapy was offered to patients presenting with adverse events , as judged by the study physician . Five and 28 days posttreatment , blood samples were collected for clinical chemistry analyses . The final parasitological assessment started on day 28 posttreatment: stool samples were obtained from all study participants over 5 consecutive days . Patients found with Fasciola eggs in their stool following artemether administration were treated with 10 mg/kg triclabendazole . In study 2 , stool samples were collected from triclabendazole-treated patients 28 days posttreatment over 3 consecutive days and CRs and ERRs were determined . Those patients who remained Fasciola positive were retreated with a double dose of triclabendazole ( 20 mg/kg given 24 h apart ) [24] and efficacy ( CRs and ERRs ) was assessed 28 days posttreatment , on the basis of three stool samples . In both groups of triclabendazole-treated patients , liver and renal function and hematological parameters were determined pre- and posttreatment ( 5 and 28 days after drug administration ) . For detection and quantification of Fasciola eggs , all stool samples were processed shortly after collection using the Kato-Katz technique [25] . From each stool sample , 3–6 thick smears were prepared on microscope slides . The slides were transported in enumerated boxes to the Theodor Bilharz Research Institute and examined within a maximum of 48 h . The presence of S . mansoni and soil-transmitted helminths ( i . e . , Ascaris lumbricoides and Trichuris trichiura ) was also determined and recorded for each participant individually . Each slide was examined independently in a blind manner by two microscopists . For quality control , several slides were re-examined by a senior staff . For confirmation of Fasciola and other helminth eggs , at baseline the merthiolate-iodine formaldehyde ( MIF ) concentration technique [26] was employed for one stool sample per participant . Briefly , 2 . 35 ml of stock MIF solution was added to at least about 0 . 5 g of each stool sample in a 15 ml centrifuge tube , closed with a rubber stopper , and placed in a refrigerator for subsequent examination . On the next morning 0 . 15 ml of Lugol's iodine solution was added to each tube . After centrifugation , the upper layers of sedimented feces containing parasite material were examined under a microscope . Laboratory investigations of blood included total leukocyte count , hemoglobin , eosinophilic count , alanine transpeptidase ( ALT ) , aspartate transpeptidase ( AST ) , alkaline phosphatase ( ALP ) , gamma glutamyl transpeptidase ( GGT ) , total serum bilirubin , blood urea , and serum creatinine . The blood specimens were collected into gel serum tubes ( for clinical chemistry variables ) and EDTA tubes ( for hematology variables ) . Blood specimens collected into gel tubes were centrifuged at 1800–2000 g for 10–15 min . All blood specimens were analyzed on the day of collection . Data were entered using EpiData version 6 . 04 ( Epidata Association; Odense , Denmark ) . CR was calculated as proportion of individuals excreting Fasciola eggs before treatment and absence of eggs at study end . To determine infection intensity , the number of Fasciola eggs per Kato-Katz thick smear ( 41 . 7 mg of stool ) was multiplied by a factor 24 to obtain eggs per gram of stool ( EPG ) . Fecal egg counts ( FECs ) of multiple slides per individual were averaged , using the arithmetric mean . To calculate the reduction in infection intensity , individual egg counts were logarithmically transformed ( log ( count + 1 ) and the GM expressed as the antilogarithm of the mean . The ERR was calculated as [1 - GM FEC after treatment divided by GM FEC at admission multiplied by a factor 100] . Although infection intensity thresholds are currently lacking for infections with Fasciola [27] , we classified infections into two groups: ( i ) light ( 1–99 EPG ) and ( ii ) moderate/heavy ( ≥100 EPG ) . Of note , a threshold of 100 EPG is also used to distringuish between light and moderate ( 100–399 EPG ) and heavy ( ≥400 EPG ) S . mansoni infection [27] . Fisher's exact test , including 95% confidence intervals ( CI ) , and Mann-Whitney U test were used to compare the outcome of both studies ( 2-sided P values ) assuming no difference in population or sensitivity of the parasite strain . The 2-tailed paired t-test and the Kruskal-Wallis tests were employed to compare the clinical parameters before and after treatment .
Of 584 villagers and 51 school-aged children screened in El-Haddad El-Bahary village ( study 1 ) , 22 individuals were found Fasciola-positive . Two patients were excluded ( pregnancy , n = 1; age below 5 years , n = 1 ) . Twenty patients ( 10 females , 10 males; aged 5–70 years with a mean of 24 years ) were included in study 1 ( Table 1 ) . In the second study , 631 individuals were examined and 19 Fasciola-positive subjects were identified . Of these , 17 patients ( 10 females , 7 males; aged 5–26 years , with a mean of 14 years ) ( Table 1 ) were included in the study . However , two of the positive cases were excluded because the initial diagnosis by the Ministry of Health and Population could not been confirmed . The baseline GM Fasciola FECs in the two studies were 28 . 3 EPG and 29 . 1 EPG ( Table 2 ) . Twenty-six individuals were classified as lightly infected ( 1–99 EPG ) , whereas 10 individuals had a moderate/heavy infection ( ≥100 EPG ) . Ten participants were concurrently infected with Fasciola spp . and S . mansoni , and one patient was identified with a double infection of Fasciola spp . and Hymenolepis nana . Data from all patients were included in the analysis , as no patient was lost to follow-up ( per-protocol analysis ) . CRs achieved with 6×80 mg and 3×200 mg artemether were 35% and 6% , respectively ( Table 2 ) . Fisher's exact test showed a statistical difference between the CRs obtained with the different treatment schedules ( P = 0 . 048; 95% CI: 0 . 002–1 . 15 ) . None of the patients characterized by an infection intensity of 100 EPG and above was cured after artemether administration regardless of the treatment regimen , while CRs documented in patients with a light Fasciola infection were 54% ( 6×80 mg artemether ) and 8% ( 3×200 mg artemether ) ( CRs of light infections were significantly higher in study 1 compared to study 2; P = 0 . 013; 95% CI: 0 . 001–0 . 77 ) . Treatment with artemether over 3 consecutive days resulted in ERRs of 63% ( 67% for light infections and 55% for infections ≥100 EPG ) . The individual pretreatment and posttreatment FECs are presented in Figure 1 . No effect on FECs were observed when artemether was administered on a single treatment day with the exception of a very low ERR of 6% among patients with an infection intensity ≥100 EPG . The overall ERR between the two studies differed significantly ( P<0 . 001 ) . In each of the two studies , five patients were co-infected with S . mansoni . At treatment follow-up , three out of the five patients in each study were recorded egg-free ( CR: 60% ) . Sixteen patients who were still found Fasciola-positive after treatment with 3×200 mg artemether were administered a single 10 mg/kg oral dose of triclabendazole . CR and ERR were 69% and 94% , respectively; significantly higher than CR ( P<0 . 001; 95% CI: 3 . 19–1605 . 7 ) and ERR ( P<0 . 001 ) observed following treatment with 3×200 mg artemether . The infection intensity did not influence the treatment outcome ( data not shown ) . Four out of five patients who were still passing Fasciola eggs following a single triclabendazole dose were provided a double dose of triclabendazole and the respective CR and ERR were 75% and 96% .
While the veterinary importance of fascioliasis cannot be overemphasized , this zoonotic disease is also of considerable and growing public health importance , yet it often remains neglected . A major challenge is that treatment is restricted to a single drug , i . e . , triclabendazole , which is registered for human use only in Ecuador , Egypt , France , and Venezuela [7] . Results from a study carried out in Vietnam raised some hope for an alternative; artesunate administered to patients with symptomatic fascioliasis pointed to a potential role of the artemisinins against fascioliasis . Indeed , the authors concluded that it is worthwhile to investigate this drug class in more detail , including additional clinical trials [20] . We now present the first results with artemether in the treatment of chronic fascioliasis in two epidemiological settings of Egypt . Artemether ( monotherapy ) was administered following the dosing regimen of a commonly used ACT , the 6-dose regimen of artemether-lumefantrine [21] , and a previously employed 3-dose malaria treatment schedule administered on a single day [22] . Egypt was selected because of the known fascioliasis endemicity , particularly in the Nile Delta , and the absence of malaria [28] , [29] . The prevalence of Fasciola spp . observed in the two study sites ( i . e . , Behera and Alexandria; prevalence 3–4% ) was similar to previous studies in these areas [5] , [28] , [30] , despite frequent community treatment programs with triclabendazole . Our study failed to extend promising findings obtained with the artemisinins in rats experimentally , and sheep naturally , infected with F . hepatica [31] . Indeed , we found low CRs ( 6–35% ) when artemether was given at two different malaria treatment schedules . Nonetheless , a moderate ERR of 63% was observed following the 6-dose course of artemether . The difference in the ERR between the two artemether treatment schedules ( nil vs . 63% ) is striking , yet difficult to explain . Since the half life of artemether is very short ( <1 h ) [32] , parasite exposure to the drug might have been insufficient if the drug is given on a single treatment day . However , detailed in vitro drug sensitivity and pharmacokinetic studies are required to further elucidate this issue . It is interesting to note that the CRs ( nil vs . 54% ) and ERRs ( 55% vs . 67% ) were higher in patients classified as lightly infected compared to moderate/heavy infections in the 6-dose regimen . A similar trend was observed in a recent study , which assessed the efficacy of an artesunate-sulfalene plus pyrimethamine combination in S . mansoni-infected school-aged children in Kenya: significantly higher CRs were observed in children harboring a light S . mansoni infection compared to moderate and heavy infections [33] . In the present study , five participants in each of the two studies were co-infected with Fasciola spp . and S . mansoni . Moderate CRs were observed against S . mansoni ( 60% ) regardless of the selected treatment regimen . This finding is in line with previous studies , which documented low-to-moderate efficacies of an artemisinin monotherapy in the treatment of chronic infections with Schistosoma spp . [34]–[36] . An opposite trend , a CR of 70% and an ERR of 86% was reported following treatment of Nigerian children using two doses of artesunate at 6 mg/kg given 2 weeks apart [37] . In recent years also the effect of ACTs on schistosomiasis has been studied ( for a summary of studies , see Utzinger et al . ( 2010 ) [38] ) Overall , a moderate efficacy was observed using ACTs against the two major schistosome species , S . mansoni and S . haematobium . Although promising results were obtained in small exploratory trials with the artemisinins against schistosomiasis , larger clinical trials could not confirm these findings , and hence praziquantel remains the drug of choice [33] , [38] , [39] . CRs of 69% and 75% were observed in patients treated with triclabendazole at 10 mg/kg and 20 mg/kg , respectively . The observed efficacy is slightly lower than a calculated overall CR of 83% following 10 mg/kg and reported CRs ranging from 93 to 100% following a double dose of triclabendazole [40] . Additionally , a recent study with 10 mg/kg triclabendazole in Egypt reported a complete cure following triclabendazole ( 10 mg/kg ) [41] . However , care is indicated in these comparisons because of the small sample sizes in the current study , although strain differences in the susceptibility of Fasciola spp . to triclabendazole might play a role in the somewhat lower efficacies observed here compared to previous studies . Participants treated with triclabendazole showed a higher incidence of abdominal pain compared to those treated with artemether , which might be related to the higher efficacy of triclabendazole ( dying worms ) . In conclusion , significantly higher CRs and ERRs were observed with triclabendazole when compared to artemether , the latter administered following two malaria treatment schedules . Hence , triclabendazole remains the drug of choice against fascioliasis . In view of threatening triclabendazole resistance development , concerted efforts are required , including structure-activity relationships with the synthetic peroxides in F . hepatica-infected rats [42] . Combination chemotherapy is also recognized as a potential strategy for reducing the emergence of drug resistance [43] , [44] . Since we have observed synergistic interactions of combinations of triclabendazole ( 2 . 5 mg/kg ) plus artemether ( 6 . 25–100 mg/kg ) on adult worm burden in F . hepatica-infected rats [15] further preclinical studies to investigate the efficacy and safety of an artemether-triclabendazole combination are warranted . Combination chemotherapy with artemether and triclabendazole might offer an advantage over triclabendazole monotherapy , in particular in the case of possible future treatment failures with triclabendazole alone . | Fasciola hepatica and F . gigantica are two liver flukes that parasitize herbivorous large size mammals ( e . g . , sheep and cattle ) , as well as humans . A single drug is available to treat infections with Fasciola flukes , namely , triclabendazole . Recently , laboratory studies and clinical trials in sheep and humans suffering from acute fascioliasis have shown that artesunate and artemether ( drugs that are widely used against malaria ) also show activity against fascioliasis . Hence , we were motivated to assess the efficacy and safety of oral artemether in patients with chronic Fasciola infections . The study was carried out in Egypt and artemether administered according to two different malaria treatment regimens . Cure rates observed with 6×80 mg and 3×200 mg artemether were 35% and 6% , respectively . In addition , high efficacy was observed when triclabendazole , the current drug of choice against human fascioliasis , was administered to patients remaining Fasciola positive following artemether treatment . Concluding , monotherapy with artemether does not represent an alternative to triclabendazole against fascioliasis , but its role in combination chemotherapy regimen remains to be investigated . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"drugs",
"and",
"devices",
"neglected",
"tropical",
"diseases",
"veterinary",
"science",
"veterinary",
"medicine"
] | 2011 | Efficacy and Safety of Artemether in the Treatment of Chronic Fascioliasis in Egypt: Exploratory Phase-2 Trials |
The synaptonemal complex ( SC ) is a proteinaceous , meiosis-specific structure that is highly conserved in evolution . During meiosis , the SC mediates synapsis of homologous chromosomes . It is essential for proper recombination and segregation of homologous chromosomes , and therefore for genome haploidization . Mutations in human SC genes can cause infertility . In order to gain a better understanding of the process of SC assembly in a model system that would be relevant for humans , we are investigating meiosis in mice . Here , we report on a newly identified component of the murine SC , which we named SYCE3 . SYCE3 is strongly conserved among mammals and localizes to the central element ( CE ) of the SC . By generating a Syce3 knockout mouse , we found that SYCE3 is required for fertility in both sexes . Loss of SYCE3 blocks synapsis initiation and results in meiotic arrest . In the absence of SYCE3 , initiation of meiotic recombination appears to be normal , but its progression is severely impaired resulting in complete absence of MLH1 foci , which are presumed markers of crossovers in wild-type meiocytes . In the process of SC assembly , SYCE3 is required downstream of transverse filament protein SYCP1 , but upstream of the other previously described CE–specific proteins . We conclude that SYCE3 enables chromosome loading of the other CE–specific proteins , which in turn would promote synapsis between homologous chromosomes .
Meiosis is a special type of cell division which gives rise to haploid , genetically diverse gametes . For organisms that reproduce sexually the correct haploidization of paternal and maternal genomes is of utmost importance . To ensure the correct separation of homologous chromosomes during anaphase I , homologs first have to find each other before coming into close physical proximity . This process takes place during prophase I of meiosis , which is highly regulated and can be subdivided into five stages: leptotene ( chromosome condensation ) , zygotene ( initiation of synapsis ) , pachytene ( full synapsis ) , diplotene ( desynapsis ) , and diakinesis ( visible chiasmata ) [1] . One key component that enables synapsis and crossover formation is the synaptonemal complex ( SC ) , a largely proteinaceous , meiosis-specific nuclear structure . The SC consists of three components: two lateral elements ( LEs ) , each of which are associated with a pair of sister chromatids , and a central region between the two LEs , that is composed of numerous transverse filaments and the central element ( CE ) . The central region physically links homologous chromosomes in a zipper-like manner and thus mediates synapsis ( reviewed in [2] , [3] ) . The tripartite structure of the SC is strikingly conserved from budding yeast to humans emphasizing its prominent function during meiosis . Much of our current understanding in the field was obtained from organisms such as Saccharomyces cerevisiae , Caenorhabditis elegans , and Drosophila melanogaster [4]–[10] . However , the mouse also has advanced to a widespread model system for the analysis of SC composition and regulation [11] , [12] . Investigations in the mouse are expected to provide a better insight into causes of infertility in humans [13]–[15] . Several mouse models have been generated over recent years in order to illuminate the process of mammalian SC assembly . With the aid of these a first molecular model of SC assembly , synapsis initiation and propagation was proposed [11] . SC assembly in mice is initiated during leptotene with the formation of the axial elements ( AEs; i . e . the precursor structures of LEs ) by the AE proteins SYCP2 and SYCP3 [2] , [16] . The AEs colocalize with cohesin cores consisting of cohesin complex proteins [17]–[19] . Mice deficient for SYCP3 lack AEs and fail to form continuous cohesin cores [20]–[22] . Interestingly , knockout of the Sycp3 gene results in a sexually dimorphic phenotype: SYCP3-deficient males are sterile due to massive apoptotic cell death during zygotene [20] , whereas females are fertile , but exhibit a sharp reduction in litter size caused by aneuploidy resulting in embryo death in utero [21] . AEs become linked by numerous transverse filaments ( hence , AEs are referred to as LEs ) when synapsis is initiated in zygotene of wild-type mice . Transverse filaments ( TFs ) are mainly composed of SYCP1 , a fibrillar protein with a central coiled-coil domain flanked by two globular N- and C-terminal domains [23] . SYCP1 molecules most likely form homodimers with a parallel orientation with both C-termini anchored in an LE and the N-termini interacting head-to-head in the CE with a SYCP1 dimer of the opposite LE [24]– . Two facts indicate the prominent function of SYCP1 during SC assembly: ( 1 ) disruption of the mouse Sycp1 gene leads to sterility in both sexes , which is caused by massive apoptotic events during spermatogenesis and oogenesis . Detailed analysis of Sycp1 null mice revealed a complete breakdown of synapsis of homologous chromosomes , although the mice show normal AE formation and chromosome alignment . Furthermore , SYCP1 is required for crossover formation and the repair of DNA double-strand breaks ( DSBs ) [27] . ( 2 ) When expressed in a heterologous system SYCP1 molecules have the capability to self-assemble and form structures that closely resemble SCs ( i . e . polycomplexes; [26] ) . Therefore , SYCP1 may function as a molecular framework to which other proteins attach to accomplish SC assembly and progression of recombination events ( [26] , [27]; and below ) . Because of the prominent function of SYCP1 in meiosis and its relevance for fertility , the identification and characterization of its interaction partners has received intense interest over the past years . Of particular interest are three proteins , SYCE1 , SYCE2 and Tex12 , which exclusively localize to the CE and , interestingly , contain a coiled-coil domain , which constitutes a common protein interaction motif [28] , [29] . These proteins , together with SYCP1 , form a complex in the CE . SYCE1 and SYCE2 were found to bind to each other and to SYCP1 , whereas Tex12 forms a complex with SYCE2 and , therefore , indirectly interacts with SYCP1 . The hypothesis that SYCP1 serves as a molecular framework is supported by the fact that disruption of the central region by eliminating Sycp1 results in a mislocalization of all three CE-specific proteins [29] . In addition , knockout models of each of the three known CE proteins revealed that loss of either of these alters higher order polymerization properties and localization of SYCP1 [30]–[32] . This indicates that the in vivo relationship between SYCP1 and all currently known CE proteins is reciprocal . Correct SC assembly is required for proper meiotic recombination and conversely , recombination is essential for correct SC assembly . This is particularly well documented in mice deficient for Spo11 , which are characterized by the lack of Spo11-dependent DSBs . In these mice SC formation is much reduced or SCs form between non-homologous chromosomes [33] , [34] . Thus SC assembly and meiotic recombination are mutually dependent on each other . Here we report on the characterization of SYCE3 , a novel protein specifically located in the CE of the mammalian SC . SYCE3 is exclusively expressed during male and female meiosis and colocalizes with SYCP1 and SYCE1 . We show that SYCE3 is part of the previously described CE complex [28] , [29] where it interacts with SYCE1 and SYCE2 . To gain more insights into the function of SYCE3 we generated a Syce3 knockout mouse . These mice are characterized by infertility in both sexes as well as complete disruption of synapsis and mislocalization of previously described CE proteins , indicating that SYCE3 is required for synapsis initiation and chromosomal loading of the other CE proteins ( i . e . SYCE1 , SYCE2 and Tex12 ) . Furthermore , we demonstrate that loss of SYCE3 has no influence on the initiation of meiotic recombination , but is required for its progression .
We selected a set of genes identified by means of ( 1 ) predominant expression in testis and ( 2 ) a predicted nuclear localization , as well as a coiled-coil domain , within the encoded protein sequence from a gene expression profile initially used to elucidate the impact of Dazl knockout on gene expression in the developing gonads [35] . Using RT-PCR we demonstrated the selective expression of one of these genes -1700007E06Rik [35] - in adult testes and embryonic ovaries , and its absence in somatic tissues ( Figure 1B ) . 1700007E06Rik encodes a protein ( which we have named SYCE3 ) consisting of 88 amino acids . SYCE3 can be found in all vertebrate classes from fish to human ( Figure 1A ) . It is highly conserved among mammals with an identity of 90% ( 96% similarity ) at the amino acid level between mouse and human . PSORTII and SMART prediction of protein structural motifs revealed that SYCE3 contains a short coiled-coil motif ( amino acids 6–39 ) which is conserved in all analyzed vertebrate sequences except for fish . In order to determine the exact temporal expression pattern of SYCE3 mRNA during spermatogenesis , we performed additional RT-PCR experiments with whole mRNA fractions of pubertal mice testes of different ages . Analysis of the first wave of spermatogenesis revealed that SYCE3 mRNA is first detectable on day 12 ( i . e . the onset of the prophase I of meiosis ) and persists in older animals . Comparison of the temporal expression profile of SYCE3 mRNA with that of SYCP3 revealed a striking similarity , suggesting that SYCE3 also functions during the prophase I stage of meiosis ( Figure 1C ) . To analyze SYCE3 expression at the protein level , we generated two antibodies against the full-length protein ( see materials and methods ) . Western blot analysis of total protein fractions of pubertal mouse testes of different ages revealed a single band with the expected mass ( 12 kDa ) from day 12 onwards , validating the fact that SYCE3 is first expressed with the onset of meiosis ( Figure 1C ) . In addition , SYCE3 restriction to meiotic cells was confirmed by immunofluorescence analysis on frozen mouse testis sections: SYCE3 expression is confined to spermatocytes and completely absent in spermatogonia , spermatids and spermatozoa ( Figure S1B ) . Using immunocytochemistry on spread mouse spermatocytes , we investigated the subcellular localization of SYCE3 in meiotic cells . For proper staging of meiotic cells we used SYCP3 as a marker for AEs/LEs ( Figure 2A ) . Fluorescence staining of SYCE3 showed that it specifically localizes to the synapsed regions of homologous chromosomes , whereas SYCP3 was present in AEs and LEs . SYCE3 was first detectable during zygotene when synapsis is initiated and staining persisted on synapsed regions of homologous chromosomes until diplotene . The localization to synapsed regions was verified by double-labeling experiments performed with SYCE3 and TF-protein SYCP1 ( a marker for synapsed regions ) in which a virtually identical staining pattern was observed ( Figure 2B ) . In pachytene oocytes SYCE3 also localized to synapsed chromosomes ( Figure 2C ) . For a detailed analysis of SYCE3 localization within the SC we performed immuno-gold electron microscopy on testis sections using an affinity-purified SYCE3 antibody and an antibody against the coiled-coil region of SYCP1 as a control ( Figure 3A ) . In preparations incubated with SYCE3 antibodies , gold particles exclusively localized to the CE of the SC . This is in strong contrast to the findings when using an antibody against the coiled-coil region of SYCP1 , where - as expected - TFs between CE and LEs are labeled ( see also [29] ) . For a quantitative analysis of immunogold data we sub-divided the distance between the SC center and the outer edge of LEs into 7 equal sections and counted the immuno-gold particles in each section ( see [36] ) . In the case of SYCE3 the bulk of gold particles ( n = 304 ) localized to the two sections adjacent to the SC center . In contrast , gold particles corresponding to the SYCP1 coiled-coil domain ( n = 135 ) largely localized to the area between LEs and CE ( see also [25]; Figure 3B ) . Since three CE-specific proteins had been discovered previously ( SYCE1 , SYCE2 and Tex12; [28] , [29] ) , we became interested in the localization of SYCE3 in respect to the other CE-specific proteins . As previously described , SYCE1 is distributed rather continuously in pachytene along synapsed areas of homologous chromosomes , whereas SYCE2 and Tex12 localize in a more punctuated pattern . Co-localization of SYCE3 and SYCE1 on spread mouse spermatocytes revealed that in both zygotene and pachytene cells these two proteins co-localize in a rather continuous pattern along the synapsed chromosomes ( Figure 4 , see inset ) . In contrast , double-labeling experiments with SYCE3 and SYCE2 showed that in zygotene and pachytene cells SYCE3 does not necessarily co-localize with SYCE2 ( Figure 4 , see inset ) . Here , SYCE3 was distributed in a more continuous pattern whereas SYCE2 appeared more punctuated . To investigate the possible dependence of SYCE3 localization on the presence of other SC-proteins , we performed immunocytochemistry on spread spermatocytes of mice deficient for SYCP1 [27] , SYCP3 [20] , SYCE1 [30] , SYCE2 [31] and Tex12 [32] . In the absence of SYCP3 , double-labeling for SYCE3 and SYCP1 showed that SYCE3 mimics SYCP1 localization , indicating that SYCE3 is loaded only to synaptic chromosome regions ( Figure 5B ) . In Sycp1−/− mice , SYCE3 is completely absent from the AEs ( Figure 5A ) . In the absence of CE protein SYCE1 , on the other hand , SYCE3 localizes to the AEs in a weak discontinuous pattern which is independent of whether AEs are in close apposition or not ( Figure 5C , and insets therein ) . In mice lacking CE protein SYCE2 , SYCE3 localized to small foci at closely aligned chromosome axes ( Figure 5D , and insets therein ) . As expected , SYCE3 also localized to small foci in mice lacking Tex12 ( data not shown ) . Taken together , the results described in the first part of this study lead to the conclusion that SYCE3 is a novel , meiosis-specific component of the CE of mammalian SCs . Chromosome loading of SYCE3 appears to require SYCP1 , but no other currently known , CE-specific proteins . To gain deeper insights into the function of SYCE3 , we generated a mouse strain lacking the SYCE3 protein . To this end , we replaced the two exons coding for the full-length protein with a neomycin cassette by electroporating a modified pKSloxPNT vector for gene-replacement into R1/E ES cells ( Figure S2A; [37] ) . Using PCR and Southern blot we identified one positive ES cell clone ( Figure S2B , S2C ) , which was injected into blastocysts of a C57BL/6 mouse to generate chimeric animals . Mating of chimeras resulted in heterozygote animals , which produced offspring with the mutated locus in Mendelian ratio . Correct deletion of Syce3 in heterozygote and homozygote animals was confirmed by Southern blot analysis ( Figure S2D ) . Syce3−/− mice displayed no overt somatic phenotype , but repeated mating attempts of wild-type with both Syce3−/− male and female mice did not produce offspring implying that both male and female Syce3−/− mice were infertile . Consistent with this , Syce3−/− testes of adult mice have a clearly reduced size compared to their wild-type littermates as previously reported for other infertile mice ( Figure S2E; i . e . [27] ) . In addition , many TUNEL-positive meiotic cells ( Figure S3A ) and no postmeiotic cells were found in histological sections of Syce3−/− testes , indicating a defect during meiosis resulting in programmed cell death at stage IV ( Figure 6 and Figure S3B ) . Compared to wild type females , ovaries of Syce3−/− littermates showed a sharp size reduction and lacked mature follicles suggesting a disruption of oogenesis ( Figure 6 ) . To address the question why Syce3 knockout leads to disruption of meiosis , we performed immunofluorescence analysis on spread mouse spermatocytes in which AEs were labeled with SYCP3 and TFs with SYCP1 as a marker for synapsis . As expected , in wild-type spermatocytes homologs were aligned in close juxtaposition during zygotene ( data not shown ) and full synapsis was achieved during pachytene ( Figure 7A ) . In Syce3−/− pachytene-like spermatocytes the vast majority of AEs of the homologs were paired and aligned along their entire lengths . Incorrect alignment of autosomes was very rare ( Figure 7A ) and was most probably due to harsh chromosome spreading . Sex chromosomes on the other hand were frequently unpaired and appeared as univalents mimicking the phenotypes previously described for other CE proteins [27] , [30]–[32] . In clear contrast to the wild-type situation , AEs completely failed to synapse in pachytene-like Syce3−/− spermatocytes ( Figure 7A ) . To further investigate the effects of SYCE3 depletion on synapsis , we compared SYCP1 localization in Syce3−/− , Syce1−/− and Syce2−/− spermatocytes . As previously described , in Syce2−/− cells SYCP1-staining is confined to regions where homologous chromosomes are in closer association ( see Figure S4B and [31] ) . In contrast , in Syce1−/− cells SYCP1 localizes to the chromosome axes in a weak discontinuous pattern regardless of whether they are closely aligned or not ( see Figure S4A and [30] ) . A similar distribution of SYCP1 was observed in cells deficient for SYCE3: here , too , SYCP1 was distributed in a weak , discontinuous pattern along the AEs , irrespective of whether they were closely aligned or not ( Figure 7A ) . This is in strong contrast to wild-type spermatocytes , where SYCP1 localizes only to the synapsed areas of homologous chromosomes , but not to aligned AEs ( Figure 7A ) . This suggests that SYCP1 is able to bind to the AEs via its C-terminus , and that the N-terminal interactions are impaired in the absence of SYCE3 . Furthermore , these data clearly show that immunofluorescence analysis obtained under our experimental conditions can reproduce previously obtained weak immunofluorescence signals ( compare Figure S4 and [30] , [31] ) and thus allow precise comparison of the different CE-mutant phenotypes . We also investigated the localization of other CE proteins in Syce3−/− mice . To this end , we labeled spread Syce3−/− spermatocytes with SYCP3 as an AE marker in combination with either SYCE1 or SYCE2 . Interestingly , both SYCE1 and SYCE2 were completely absent from the axes in cells lacking SYCE3 ( Figure 7A ) . These results provide clear evidence that SYCE3 is required for loading of the other CE proteins . To obtain more detailed information about synapsis defects in the Syce3 knockout mice , we performed electron microscopic analysis on Syce3−/− testis . In wild-type spermatocytes , normal SCs composed of LEs with attached chromatin and a CE were observed . In contrast , we found partially aligned AEs in Syce3−/− spermatocytes , but no CE or CE-like structures at all ( Figure 7B ) . This phenotype resembles the situation found in Sycp1−/− and Syce1−/− mice , but differs from that of Syce2−/− and Tex12−/− mice in which synapsis appears to initiate due to the assembly of short CE-like structures [30]–[32] . During leptotene , meiotic recombination is initiated by the introduction of DNA DSBs . These sites become marked by histone γH2AX . During leptotene and zygotene , γH2AX is located in large domains around the DNA breaks , but as meiotic prophase I progresses , it becomes restricted to the sex chromosomes [38] . In Syce3−/− spermatocytes , however , immunostaining of γH2AX revealed altered dynamics . While distribution of γH2AX in early ( leptotene , zygotene ) mutant spermatocytes resembles that of early wild-type cells ( data not shown ) , γH2AX is not restricted to the sex chromosomes during the pachytene-like stage ( Figure 8 ) . Instead , γH2AX remains associated with most of the chromosomes in a cloud-like manner . The persistence of γH2AX-staining up to the advanced stages of prophase I suggests that DSBs are formed in Syce3-deficient spermatocytes , but that they are not efficiently repaired . To gain additional insights into further processing of DSBs in mutant cells , we performed immunostaining for proteins that are specific for different recombination nodules . In wild-type mice , one distinguishes between early nodules ( ENs ) which appear prior to synapsis and assemble at sites of DSBs [39] , [40] , transitional nodules ( TNs ) during zygotene [41] and late recombination nodules ( RNs ) which mark sites of future crossover events [42] . ENs assemble at the leptotene stage and are made up of the two RecA homologs RAD51 and DMC1 . They form numerous foci along chromosome cores and catalyze strand exchanges between homologous DNA molecules as a first step during processing of DSBs to crossover events [41] , [43] . During zygotene of wild-type mice , RAD51 and DMC1 are gradually replaced by RPA . Thereby ENs are transformed into TNs which appear isochronously to synapsis . It is likely that TNs are involved in stabilization or resolution of early recombination intermediates [41] , [43] . In Syce3−/− spermatocytes , RAD51 and RPA form numerous foci , localized to chromosome cores during zygotene ( data not shown ) . This pattern of distribution resembles that of wild-type spermatocytes in early stages of prophase I ( data not shown ) . These observations are consistent with the notion that early DNA-DNA interactions can be mediated by RAD51 and RPA in the absence of SYCE3 . However , the following processing steps are likely to be disturbed as RAD51 and RPA remain associated with chromosomes in pachytene-like spermatocytes ( Figure 8 ) . MLH1 marks presumed future crossover sites [43]–[45] . Correspondingly , in wild-type pachytene spermatocytes each bivalent displayed one or two MLH1 foci . In contrast , male littermates lacking SYCE3 have no MLH1 foci , pointing to a disruption of crossover formation ( Figure 8 ) . To rule out the possibility that the persistence of RPA on chromosome axes and the lack of MLH1 foci in Syce3-deficient spermatocytes is due to their arrest in pachytene and their subsequent elimination by apoptosis rather than reflecting a direct function of SYCE3 in homologous recombination we additionally performed a close examination of recombination in Syce3−/− oocytes at 19 . 5 dpc ( days post coitum ) . In concordance with earlier reports [46] , [47] the majority of 19 . 5 dpc oocytes were staged at late pachytene or diplotene . Consistent with this , SCYP3 labeled AEs were fully synapsed in wild-type pachytene oocytes . As already described above for Syce3−/− spermatocytes synapsis was completely abolished in SYCE3-deficient pachytene-like oocytes . Compared to the situation in the male ( see Figure 7A and Figure 8 ) , however , pairing and alignment in 19 . 5 dpc Syce3−/− oocytes seemed to be more strongly affected ( Figure 9A ) . Whether this finding reflects a female–specific function of SYCE3 in homologous pairing and/or alignment or is rather a secondary effect caused by defective synapsis cannot be judged at present . Careful analysis of recombination markers in 19 . 5 dpc oocytes strongly supported the notion that SYCE3 is essential for progression of meiotic recombination ( Figure 9A ) . Consistent with our findings in pachytene-like Syce3−/− spermatocytes , in 19 . 5 dpc oocytes of SYCE3-deficient animals γH2AX stayed associated with chromosome axes , unlike in oocytes of wild-type littermates ( Figure 9A ) . This suggests the persistence of unrepaired DSBs . Moreover , as judged by quantitative analysis of RPA and MLH1 dynamics , processing of TNs to RNs appeared to be considerably affected by the absence of SYCE3: In wild-type oocytes the median of RPA foci per cell was 22 during pachytene and decreased to 4 at diplotene stage . At the same time , the median number of MLH1 foci per cell was 22 at pachytene and 7 . 5 at diplotene stage in the controls . Thus , both the dynamics of RPA and MLH1 foci as well as their relative abundance found in our wild-type controls were consistent with data reported previously [41] . In clear contrast , the median number of RPA foci in SYCE3-deficient pachytene-like oocytes was significantly higher compared to wild-type controls ( 67 vs . 22; p<0 . 001 ) . Additionally , RPA foci persisted on chromosome axes of Syce3−/− oocytes at diplotene-like stage ( median , 27 vs . 4; p<0 . 001 ) indicating that processing of TNs to RNs was impaired in the absence of SYCE3 ( Figure 9B ) . In line with this notion , MLH1 was virtually completely absent from chromosome axes of Syce3−/− oocytes at both late pachytene- ( median , 0 ) and dilpotene-like stage ( median , 0 ) ( Figure 9B ) . To rule out the possibility that the considerable alterations of RPA and MHL1 dynamics in SYCE3-deficient oocytes could be caused by a delay of meiotic progression per se , we quantified the ratio of pachytene ( WT , n = 44; Syce3−/− , n = 53 ) and diplotene ( WT , n = 33; Syce3−/− , n = 35 ) stages in 19 . 5 dpc oocytes of both Syce3−/− and wild-type females . Here , no significant difference could be observed ( p = 0 , 688; Figure 9C ) . Together , our data strongly argues that while progression through meiotic stages per se remains unaffected by the loss of SYCE3 in females , progression of recombination ( i . e . processing of recombination intermediates into MLH1-marked late RNs , which are presumed markers of future crossovers in the wild-type ) is critically depending on its presence . We demonstrated that SYCE3 is important for fertility , initiation of synapsis and for a correct progression of meiotic recombination . Hence , we were interested in identifying binding partners of SYCE3 that would provide mechanistic insights into SYCE3 function . As the localization of the other CE components , SYCE1 , SYCE2 and Tex12 , is severely altered in Syce3−/− mice , they appear to be good candidates for being SYCE3 binding partners . Therefore , we performed co-transfection/immunoprecipitation experiments in somatic cells that do not express meiosis-specific proteins according to the approach described by Stewart-Hutchinson et al . [48] . COS-7 cells were transfected with EGFP- or myc-tagged fusion constructs of SYCE1 , SYCE2 , Tex12 , SYCP1 N-terminus , SYCP1 C-terminus ( as a control ) and SYCE3 . We used either a myc or an EGFP-specific antibody for precipitation . We found that SYCE3 interacts with SYCE1 ( Figure 10A , 10B ) , which is consistent with the highly similar spatial-temporal expression of these two proteins . Interestingly , we show that SYCE3 also binds to SYCE2 although these two proteins do not exactly colocalize in pachytene spermatocytes ( Figure 10C , 10D ) . However , under these experimental conditions , we could not detect an interaction between SYCE3 and Tex12 or the N- or C-terminus of SYCP1 ( Figure 10E , 10F , 10G , see also below ) .
Detailed analyses of Syce2 and Tex12 knockout mice showed that loss of each of these two genes causes infertility in both sexes . In both null mice , meiotic chromosomes align , and AEs form , but they do not synapse . Although synapsis between homologs is initiated at multiple positions along the axes , it fails to propagate along the entire chromosomes . SYCP1 and SYCE1 colocalize at these sites of synapsis initiation and , as revealed at the electron microscopical level , short CE-like structures become assembled . Furthermore , correct progression of meiotic recombination is altered in cells lacking SYCE2 or Tex12 . Altogether , these data and the direct interaction of SYCE2 and Tex12 lead to the model that both SYCE2 and Tex12 are required for the longitudinal polymerization of SYCP1 filaments along the axial elements and thus for the propagation of synapsis along the homologs [31] , [32] . Elimination of SYCE1 - the other currently known CE protein - displays a different phenotype . The absence of SYCE1 leads to the alignment of homologous chromosomes with a disrupted synapsis , but , in contrast to SYCE2 and Tex12 knockout mice , SYCE1-deficient mice display no sites of synapsis initiation and no CE-like structures at all . In these mice , SYCP1 is located in a weak discontinuous pattern along AEs , whether they are closely aligned or not . This indicates that under physiological conditions SYCP1 alone is insufficient for formation of stable head-to-head polymers for which CE proteins are required . Despite this striking difference to SYCE2 and Tex12 null mice , SYCE1-deficient mice also exhibit a disturbed progression of meiotic recombination [30] . Taken together , analysis of these central region knockout mice leads to three main conclusions: ( 1 ) The organization of the central region appears to be highly complex , ( 2 ) disruption of any currently known protein component causes defective synapsis leading to severe meiotic defects and infertility and ( 3 ) the correct assembly of the central region is also required for normal progression of meiotic recombination [27] , [30]–[32] . Here , we have demonstrated that in wild-type cells SYCE3 distribution closely resembles SYCE1 localization ( Figure 4; [28] , [29] ) . Thus , we expected that SYCE3 is likely involved in a complex with SYCE1 . Consistent with this assumption , no CE-like structures were detected in the EM analysis of Syce3−/− mouse testis sections ( Figure 7B ) . The complete absence of any CE-like structure strongly suggests that , in contrast to SYCE2 and Tex12 , SYCE3 is required for initiating synapsis . Furthermore , we have shown that SYCE1 and the SYCE2/Tex12 complex do not localize to chromosomes in Syce3−/− spermatocytes ( Figure 7A ) . The lack of CE-like structures and CE-specific proteins indicates that SYCE3 is required upstream of SYCE1 , SYCE2 and Tex12 and downstream of SYCP1 during the initiation of CE assembly . A possible function of SYCE3 could be to enable recruitment of SYCE1 and the SYCE2/Tex12 complex to SYCP1 N-termini . Loading of SYCE1 and SYCE2 to SYCP1 in turn would stabilize SYCP1 N-termini as proposed by Costa et al . [28] . This hypothesis is consistent with our results obtained from co-immunoprecipitation assays with transfected COS-7 cells , revealing that SYCE3 is capable of binding SYCE1 as well as SYCE2 ( Figure 10 ) . However , whether the role of SYCE3 in CE assembly is of a structural or regulatory character still needs to be clarified . Taken together , we conclude that SYCE3 is essential for initiating synapsis and for chromosomal loading of SYCE1 and the SYCE2/Tex12 complex . According to our present knowledge the behavior of SYCP1 and SYCE3 in meiotic cells lacking certain CE-specific proteins remains somewhat puzzling . In wild-type cells SYCP1 localization is restricted to synapsed chromosome axes . By contrast , in the absence of SYCE1 , SYCP1 is observed in a weak , discontinuous pattern at the unsynapsed chromosome axes , no matter whether they were closely aligned or not . However , this is not the case in the absence of SYCE2 or Tex12 , as under these conditions SYCP1 exclusively localizes to small foci at the sites of synapsis initiation [31] . These observations led to the hypothesis that SYCE1 is required to restrict SYCP1 to synapsed areas [30] . Our finding that SYCE3 also localizes to unsynapsed chromosome axes in SYCE1 null spermatocytes suggests that SYCE3 restriction to synapsed areas also depends on SYCE1 ( Figure 5C ) . Nevertheless , we presently have no satisfactory explanation as to how this restricted localization can be accomplished . Another aspect that would require further investigation is the relationship between SYCP1 and SYCE3 . As mentioned above , chromosomal loading of SYCE3 requires SYCP1 . On the other hand , in our co-transfection/immunoprecipitation studies we obtained no evidence for an interaction between SYCP1 and SYCE3 ( Figure 10 ) . One possibility might be that interaction between these two proteins requires higher order structures of SYCP1 ( for example dimerization or N-terminal association of dimers ) that cannot occur under the conditions of the experimental assays . However , the existence of additional , yet undiscovered CE proteins that would mediate this binding cannot be ruled out a priori . In early spermatocytes of Syce3-deficient mice γH2AX is distributed as in wild-type animals suggesting that homologous recombination is initiated in a wild-type manner . However , in later stages of prophase I γH2AX shows altered dynamics in spermatocytes and oocytes . In these cells it remains associated with chromosomes ( Figure 8 and Figure 9 ) . Thus we suggest that induced DSBs are repaired inefficiently in Syce3 knockout meiocytes . This assumption is further supported by the analysis of components of the recombination machinery that assemble at the sites of DSBs as a further step during the process of meiotic recombination [41] , [43] . In the absence of SYCE3 , both Syce3−/− spermatocytes and oocytes reveal differences to the wild-type: RAD51 - as well as RPA foci - stay associated with chromosomal cores and virtually no MLH1 signal is detectable ( Figure 8 and Figure 9 ) . The observed defects indicate that the exchange of homologous DNA strands ( catalyzed by RAD51 ) and the formation of early recombination intermediates can take place in the absence of SYCE3 , but further processing of the recombination sites fails to occur . As a consequence , no crossovers are formed in Syce3−/− meiocytes ( see also [41] , [43] ) . Interestingly , disruption of the CE by eliminating either Syce1 , Syce2 or Tex12 results in a similar recombination phenotype . Our observations confirm and extend the notion that CE assembly and CE-specific protein components are not necessary for recombination initiation but essential for recombination progression [27] , [30]–[32] . What is the physical link between the recombination machinery and the SC ? In mammals , early recombination events and formation of ENs take place before SC central region assembly [27] . Early recombination events also appear to be independent of AE assembly during leptotene/zygotene as shown in mice lacking SYCP3 . In these mice , AE assembly is impaired . However , they show normal levels of DMC1 foci formation and synapsis occurs between homologous chromosome regions [21] , [22] . Recombination progression and crossover formation can also take place in the absence of AE assembly . As shown in female Sycp3−/− mice , oocytes reveal the presence of chiasmata but at a lower level than in the wild-type , which results in a reduction but not a complete loss of fertility [21] . In contrast , assembly of the SC central region is required for recombination progression and crossover formation as shown in mice lacking TF protein SYCP1 or any of the CE-specific proteins ( [27] , [30]–[32]; this study ) . Previous electron microscopical studies revealed a close contact between RNs and SC components ( e . g . [49] , [50] ) . RNs seem to coalesce with the CE and fibers were shown to connect RNs and LEs [50] . These and other observations lead to the proposal that components of RNs might play a role in SC assembly . On the other hand , the SC central region can be seen as a platform for attachment and organization of RN components required for proper recombination progression ( see [50] , [30] and references therein ) . Our knowledge about the protein-protein interactions between components of the recombination machinery and the SC of mammals is rather fragmentary . Interactions have been reported between RAD51 and TF protein SYCP1 as well as CE protein SYCE2 [30] , [51] , and between TN component Tex11 ( ZIP4H ) and LE protein SYCP2 [52] . Although still preliminary , the emerging picture leads to the proposal that SCs and RNs are held together through a network of protein-protein interactions . Therefore , the recombination phenotype described here might be caused by the inability of RN components to bind SYCP1 and SYCE2 due to the defective localization and assembly of these proteins . However , a direct involvement in these interactions of additional CE element proteins ( including SYCE3 ) cannot be excluded at present . Despite the progress in recent years , elucidation of the mutual dependence between SC assembly and meiotic recombination events would require additional experiments .
All animal care and experiments were conducted in accordance with the guidelines provided by the German Animal Welfare Act ( German Ministry of Agriculture , Health and Economic Cooperation ) . For the generation of Syce3 knockout mice we obtained approval from the Landesdirektion Dresden ( 24-9168 . 11-9/2005-1 ) . Animal housing and breeding was approved by the regulatory agency of the city of Würzburg ( Reference ABD/OA/Tr; according to §11/1 No . 1 of the German Animal Welfare Act ) . All aspects of the mouse work were carried out following strict guidelines to insure careful , consistent and ethical handling of mice . Taking the mouse SYCE3 protein sequence ( GenBank accession number: NP_001156354 ) as query we analyzed predicted coiled-coil domains using PSORTII ( http://psort . hgc . jp/ ) [53] and predicted phosphorylation sites ( http://www . cbs . dtu . dk/services/NetPhos/ ) [54] . The mouse Syce3 cDNA sequence also served as query for searching all GenBank sequences with the BlastN and TBlastN algorithm ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) . Multiple sequence alignments were performed online with CLUSTALW ( http://www . ebi . ac . uk/Tools/clustalw2/index . html ) [55] . Whole RNA from mice testes from different ages and from various tissues of adult mice was extracted using peqGOLD TriFast™ ( Peqlab , Erlangen , Germany ) according to the manufacturer's protocol . cDNA was synthesized from 1 µg of RNA by reverse transcription with Oligo ( dT ) primers and M-MLV reverse transcriptase ( Promega , Mannheim , Germany ) . Reverse transcribed cDNA samples were stored at −20°C before they were used in a polymerase chain reaction . Specific primers used for RT-PCRs and respective PCR conditions are listed in Table S1 . In order to clone SYCE3 cDNA ( GenBank accession number: HQ130280 ) , we amplified full-length SYCE3 from cDNA derived from a reverse transcription of total testis RNA , as described above and using the same oligonucleotides and PCR conditions described for SYCE3 RT-PCR ( Table S1 ) . To raise SYCE3 specific antibodies , we generated and purified a GST-SYCE3 fusion protein using the vector pGEX-5X-1 ( Amersham Pharmacia Biotech , Braunschweig , Germany ) and the Bulk GST Purification Module ( Amersham ) according to the manufacturer's protocol . Anti-SYCE3 antisera were raised by immunizing a rabbit and a guinea pig with the purified GST-SYCE3 fusion protein ( Seqlab , Göttingen , Germany ) . Specificity of both affinity-purified antibodies was validated by testing them on separate Western blots with protein lysates from wild-type , Syce3+/− and Syce3−/− littermates ( generation of Syce3−/− mice is described below ) . The presence of the expected band of 12 kDa in wild-type and Syce3+/− testis lysates and moreover the absence of the aforementioned band in Syce3−/− testis lysates confirmed the specificity of both antibodies ( see Figure S1A ) . Protein samples derived from co-immunoprecipitation analysis were separated on 10%–15% polyacrylamide gels [56] . Separation of protein samples from testicular cells of adult mice was carried out by using tricine-SDS-PAGE ( 16% separating gel/6 M urea; [57] ) . Proteins were transferred to nitrocellulose membranes using the semi-dry Western blotting system described by Matsudaira [58] . The membranes were blocked overnight at 4°C in TBST buffer ( 10 mM Tris/HCl , pH 7 . 4 , 150 mM NaCl , 0 . 1% Tween 20 ) containing 5% milk powder . Incubation with the respective primary antibody was carried out in blocking solution for 1 h at room temperature: guinea pig anti-SYCE3 ( 1∶1000 ) , rabbit anti-SYCE3 ( 1∶1000 ) , mouse anti-myc ( 1∶2000; R950-25 , Invitrogen , Darmstadt , Germany ) , mouse anti-GFP ( 1∶200; sc-9996 , Santa Cruz Biotechnology , Heidelberg , Germany ) , guinea pig anti-SYCE1 ( 1∶1000 ) [29] , guinea pig anti-SYCE2 ( 1∶400 ) [29] , mouse anti-actin ( 1∶10000 , A4700 , Sigma-Aldrich , Munich , Germany ) . Peroxidase-conjugated secondary antibodies were applied as specified by the manufacturer ( Dianova , Hamburg , Germany ) . Bound antibodies were detected with the enhanced chemiluminescence system ( Amersham ) . Western blots were stripped by incubating the membranes for 30 min in 0 . 1 M glycine buffer ( pH 2 . 5 ) followed by a 30 min incubation in 0 . 1 M Tris buffer containing 2% SDS and subsequently washing three times in TBST . Spread-preparations of meiocytes were produced as described by de Boer et al . [59] and in each experiment spread-preparations were immunostained at the same time with the same mixture of the appropiate affinity-purified primary antibodies: rabbit anti-SYCE3 ( 1∶50 ) , guinea pig anti-SYCE3 ( 1∶100 ) , guinea pig anti-SYCE1 ( 1∶1000 ) [29] , guinea pig anti-SYCE2 ( 1∶200 ) [29] , guinea pig anti-SYCP1 ( 1∶150 ) [60] , rabbit anti-SYCP1 ( 1∶200 ) [26] , guinea pig anti-SYCP3 ( 1∶150 ) [61] , rabbit anti-SYCP3 ( 1∶200; NB300-232 , Acris , Herford , Germany ) mouse anti-γH2AX ( 1∶500; 05-636 , Millipore , Schwalbach/Ts . , Germany ) , mouse anti-RPA ( 1∶40; NA19L , Calbiochem , Darmstadt , Germany ) , rabbit anti-RAD51 ( 1∶30; PC130 , Calbiochem ) , mouse anti-MLH1 ( 1∶30; 551091 , BD Pharmingen , Heidelberg , Germany ) . Secondary antibodies were applied as specified by the manufacturer ( Dianova ) . Histology was performed on 5 µm sections of paraffin-embedded testis or ovary tissue fixed overnight in 4% formaldehyde according to standard protocols . Staging of mouse seminiferous tubule cross-sections was done according to Ahmed and de Rooij [62] . A TUNEL assay was carried out on 10 µm sections of paraffin embedded testis or ovary tissue fixed overnight in 4% formaldehyde using the ApopTag Peroxidase In Situ Apoptosis Kit ( Millipore ) according to the manufacturer's protocol . Statistically significant differences in the medians of RPA and MLH1 foci comparing Syce3−/− and control cells were verified by Mann-Whitney U test . Numbers of pachytene and diplotene stages from control and Syce3−/− 19 . 5 dpc oocytes were statistically compared using a chi-square test . Fluorescence microscopy was carried out by using a Zeiss Axiophot fluorescence microscope ( Zeiss , Munich , Germany ) equipped with a Plan-NEOFLUAR 40×/0 . 75 or a Plan-NEOFLUAR 20×/0 . 5 objective and the AxioCam MRm ( Zeiss ) camera . Digital images were pseudocoloured using the AxioVs40 V4 . 7 . 1 . 0 software release and processed using Adobe Photoshop ( Adobe Systems , San Jose , CA ) . Light microscopy was carried out using the stereo microscope MZ FLIII ( Leica ) . Confocal laser scanning microscopy was performed with a Leica TCS-SP2 confocal laser scanning microscope ( Leica , Bensheim , Germany ) equipped with a 63×/1 . 40 HCX PL APO lbd . BL oil-immersion objective . All confocal images are pseudocoloured using the Leica TCS-SP2 software and are two-dimensional projections calculated from a series of sequenced optical sections using the maximum projection algorithm ( Leica ) . Imaging of wild-type and mutant SYCE1 , SYCE2 , SYCP1 , SYCP3 and SYCE3 cells was performed using the same microscope settings . Digital images were processed with the same settings in Adobe Photoshop ( Adobe Systems ) . Electron microscopy was performed using ultra thin sections of testis tissue fixed in 2 . 5% glutaraldehyde and 1% osmium tetroxide as described previously [22] . For immunoelectron microscopy 10 µm cryosections of rat testis were fixed with acetone for 10 min at −20°C and air-dried . Incubation with primary antibodies ( guinea pig anti-SYCE3 ( 1∶500–1∶1500 ) ; mouse anti-SYCP1 ( 1∶50 ) [26] was carried out in a humidified box for 4 h at room temperature . After rinsing twice in PBS , sections were fixed for 10 min in 2% formaldehyde and blocked with 50 mM NH4Cl . 6 nm gold conjugated secondary antibodies were incubated overnight at 4°C and samples were washed subsequently in PBS . Samples were fixed for 30 min in 2 . 5% glutaraldehyde and postfixed in 2% osmium tetroxide . After rinsing three times with H2O , samples were dehydrated in an ethanol series and embedded in Epon . Ultrathin sections were stained with uranyl acetate and lead citrate according to standard procedures [22] . We deleted the Syce3 gene by replacing the entire coding sequence of SYCE3 ( exons 2 and 3 ) with a neomycin cassette in reverse orientation using a modified pKSloxPNT vector . The vector for homologous recombination was designed as follows ( see also Figure S2A ) : a 1 . 1 kbp genomic fragment ( F1 ) containing part of intron 1 was cloned into the SalI restriction site downstream of the neomycin cassette and a 4 . 1 kbp fragment ( F2 ) containing part of intron 3 was ligated into the EcoRI restriction site located in between the thymidine kinase and neomycin cassette . Electroporation of the modified replacement vector into the R1/E embryonic stem cells , laser assisted microinjection into 8-cell C57BL/6 morula and transfer of morula into CD1 ( outbred ) foster mice was performed at the transgenic core facility of the Max Planck Institute of Molecular Cell Biology and Genetics , Dresden . After electroporation and selection , we identified one positive ES cell clone by PCR with external primers ( oligonucleotide sequence for genotyping: see Table S2 ) and confirmed correct targeting by Southern blot . For Southern blot analysis 10 µg BstEII digested DNA derived from ES cells ( or tail tips of Syce3+/+ , Syce3+/− and Syce3−/− mice ) was loaded on a 0 . 8% agarose gel , subsequently transferred to a nylon membrane and correct insertion was tested with both external and neomycin probes . Blastocyst injection of the ES cell clone produced germline transmitting chimeras . Mating of chimeras with C57BL/6 mice gave rise to wild-type and Syce3+/− mice . Intercrossing of Syce3+/− mice produced offspring with all genotypes in Mendelian ratio . To confirm the absence of SYCE3 we performed PCR , Southern blot , immunofluorescence analysis on spread Syce3−/− spermatocytes and Western Blot analysis on Syce3−/∼ testis tissue with polyclonal antibodies raised against the full-length SYCE3 protein ( data not shown and Figure S1A ) . In order to express full-length SYCE3 , SYCE1 , SYCE2 , Tex12 as well as SYCP1 N-terminal ( aa 1–120 ) and SYCP1 C-terminal ( aa 820–997 ) fusion constructs in the culture cell line COS-7 ( green monkey kidney ) for co-immunoprecipitation analysis , the respective cDNAs were inserted into pEGFP ( Clontech , Heidelberg , Germany ) or pCMV-Myc ( BD Bioscience ) vectors [60] . The fusion-constructs used ( including the oligonucleotides used for cloning ) are summarized in Table S3 . Cells were transfected with the respective constructs using the effectene system according to the manufacturer's instructions ( Qiagen , Hilden , Germany ) . Co-immunoprecipitation experiments were performed as described by Stewart-Hutchinson et al . [48] with the following modifications: ( 1 ) myc and EGFP constructs were immunoprecipitated with 0 . 5 µg of mouse anti-myc ( R950-25 , Invitrogen ) or 0 . 5 µg of mouse anti-GFP ( sc-9996 , Santa Cruz Biotechnology ) antibody per 60-mm dish . ( 2 ) Immune complexes were pulled down by protein G dynabeads ( 100-03D , Invitrogen ) . | Meiosis is a special type of cell division that takes place in the germ line of sexually reproducing diploid organisms . Major events during meiosis are the pairing , recombination , and segregation of homologous chromosomes . As a consequence , daughter cells are haploid and genetically diverse . Therefore , meiosis is of utmost importance for the life of sexually reproducing species as it maintains the species-specific chromosome number and generates genetic diversity within a species . Proper segregation of homologous chromosomes during meiosis requires homolog pairs to be physically linked . The synaptonemal complex ( SC ) , a meiosis-specific structure conserved in evolution , is essential for this process . Defective assembly of the SC has deleterious effects on germ cells and can cause infertility in mice and humans . Here , we report on a newly identified protein component of the mammalian SC that we have named SYCE3 . SYCE3 is strongly conserved among mammals . Using the mouse as a model system , we demonstrate that loss of SYCE3 leads to infertility in both sexes . Infertility is caused by disruption of meiosis due to the inability of Syce3−/− mice to assemble the central element of SCs . Our findings provide new insights into the complexity of SC assembly and its relevance to mammalian fertility . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"meiosis",
"chromosome",
"biology",
"biology",
"genomics",
"genetics",
"and",
"genomics"
] | 2011 | A Novel Mouse Synaptonemal Complex Protein Is Essential for Loading of Central Element Proteins, Recombination, and Fertility |
Tropical coral reefs feature extraordinary biodiversity and high productivity rates in oligotrophic waters . Due to increasing frequencies of perturbations – anthropogenic and natural – many reefs are under threat . Such perturbations often have devastating effects on these unique ecosystems and especially if they occur simultaneously and amplify each other's impact , they might trigger a phase shift and create irreversible conditions . We developed a generic , spatially explicit , individual-based model in which competition drives the dynamics of a virtual benthic reef community – comprised of scleractinian corals and algae – under different environmental settings . Higher system properties , like population dynamics or community composition arise through self-organization as emergent properties . The model was parameterized for a typical coral reef site at Zanzibar , Tanzania and features coral bleaching and physical disturbance regimes as major sources of perturbations . Our results show that various types and modes ( intensities and frequencies ) of perturbations create diverse outcomes and that the switch from high diversity to single species dominance can be evoked by small changes in a key parameter . Here we extend the understanding of coral reef resilience and the identification of key processes , drivers and respective thresholds , responsible for changes in local situations . One future goal is to provide a tool which may aid decision making processes in management of coral reefs .
Tropical coral reefs are highly productive but also fragile ecosystems that provide habitats for the coastal fauna and multiple services to local human communities [1] . Due to their high biodiversity , they exhibit a complex pattern of interactions between organisms and their environment with feedback loops within and between trophic as well as different hierarchical levels [2] , and thereby facilitate a framework of non-linear dynamics which complicates a holistic analysis . Although extensive knowledge of corals , their responses to environmental change [3] and interaction with other organisms [4] , and reef resilience [5] has been gained in the last few decades , the understanding of coral reef functioning is still far from being complete [6] . Reefs are increasingly under threat and many coral species are in danger of becoming extinct [7] , due primarily to anthropogenic influence . Globally , coral reef systems are subject to rising sea surface temperatures which increase their susceptibility to bleaching , and to ocean acidification which erodes CaCO2 structures . Both stressors are chronically increasing and can be attributed to climate change [3] , [5] . Additionally it is predicted that extreme weather events ( e . g . el Niño or hurricanes ) will strike with increasing frequency [8] , [9] . Directly imposed human pressure upon coral reefs can have physical – e . g . by the use of destructive fishing techniques [10] , [11] , sedimentation [12] , or anchorage [13]–[15] – or chemical – e . g . nutrients , sewage , pollution [11] , [16] consequences . The overall tendency of coral reef systems to react to changes in environmental conditions and anthropogenic influences can be described by the term resilience . It “… determines the persistence of relationships within a system […] and is a measure of the ability of these systems to absorb change […] and still persist . ” [17] . In a coral reef it may be determined by species diversity , functional redundancy , life history of reef organisms , species functioning at different spatial and temporal scales , and connectivity to other reefs or habitat types [18] , [19] . Reduced resilience can impose catastrophic regime shifts [20] , [21] and in a reef often lead to a phase shift from coral dominated systems to alternative states; i . e . dominance of macroalgae [16] , [22] , [23] or of other benthic organisms [24] , but see [25] . During the last two decades a series of ecological models have been applied to coral reef ecosystems . Among these we can find applications on various spatial and temporal scales . While Kleypas et al . [26] sought to approximate the possible geographic range for coral reefs to exist globally , other applications focus on conservation [27] or sustainable fishing regimes [28] , [29] . There are yet other models at the regional , local and/or small scale [30]–[34] with the purpose to explore the influence of environmental conditions on spatial processes and interactions of coral reef community dynamics , and some of these models , like [35] are designed to aid management decisions . Individual-based models ( IBM ) have proven to be an exceptional tool to tackle ecological questions with adequate detail [36]–[38] because properties of investigated ecological systems can be described very close to reality . By including , for example , heterogeneously varying individual interactions and spatial heterogeneity , IBMs considerably extend the range of ecological modelling [39] . In this study we focus on individual benthic organisms and their interaction with the environment because these processes and the spatial configuration of a community are the basis for environmental responses to perturbations in reality . There is a lot of knowledge on properties of individual coral colonies of various species; e . g . which symbionts they possess , how they grow , and how they react to thermal stress [40] , [41] , upon changing environmental settings in general , or if faced with other benthic organisms within their local neighborhood [42]–[45] . All of these factors are relevant for the understanding of coral reef functioning and should be included in an analysis of local reef dynamics . To date , the application of individual-based models in the context of coral reefs has been somewhat limited , but interesting models have been developed for some investigations . Yniguez et al . [46] described the three-dimensional growth pattern of Halimeda tuna , a common macroalga in Florida Key reefs . Sleeman et al . [47] utilize an individual-based model to analyze different spatial arrangements of coral transplants in order to improve reef restoration measures . Koehl et al . [48] simulated larval transport in turbulent waters , and Brandt and McManus [49] investigated the spread of the white plague disease in various coral populations . Tam and Ang [50] present a strictly theoretical 3-dimensional model in which they describe disturbance-induced changes in a coral community with three different hypothetical coral growth patterns . Here , we present a generic multi-species individual-based coral reef model in which scleractinian coral species and algae compete for space . This tool enables the analysis of key functions for coral reef resilience and the identification of major causes of phase shifts for local situations . In our example we apply a basic system with a standard parameterization for a typical Western Indian Ocean reef system . In order to improve the understanding of how climate change and different modes of human interference affect the benthic composition of specific reef sites and their resilience we examine community responses under various environmental settings . Hence , we apply ( 1 ) different frequencies of temperature-induced bleaching , ( 2 ) two levels of mechanical disturbance regimes: a smaller one , which represents , for example , direct anchor damage or a smaller boat hitting the coral reef at low tide and a larger one representing , for example , damage by abrasion due to anchor chains , boat crashes or from fishing nets , and ( 3 ) both perturbations acting together to test the influences on the benthic community .
At standard settings , the massive P . lutea dominated , followed by the branching A . muricata . The other two species both leveled at around 5% coverage . A . muricata exhibited the highest fluctuations in relative cover and P . lobata the lowest . Surveyed exclusively , both perturbation modes triggered similar responses in the benthic community; not only changes in overall benthic coverage but also alterations in community composition . For low frequencies of rare events , massive growth forms clearly dominated the system with high total benthic cover . Macroalgal cover is negligibly low . If perturbations occurred at intermediate frequencies , overall cover decreased and space was nearly evenly distributed between the two different growth morphologies . At the highest perturbation levels , the total benthic cover was very low , the relative fraction of algae increased strongly and massive species were displaced by branching ones . Nevertheless , single effect testing of major bleaching events and mechanical disturbances revealed that both perturbation types triggered differential responses of the benthic community . These are discussed further below . Without the influence of mechanical disturbances , there was nearly no visible change in the total benthic cover ( which was always ≥99% ) , if major bleaching events occurred in intervals of 16 years or higher . The community was dominated by P . lutea with a relative coverage of ∼80% or larger ( Fig . 4 ) , followed by P . lobata ( ∼18% ) . When extreme temperatures occurred every 10 to 15 years , major changes in relative cover arose mainly for P . lutea which decreased from 76 . 1 to 28 . 1% coverage , and P . lobata which decreased from 17 . 0 to 6 . 0% coverage . For the dominance of P . lutea we identified a threshold of 8–9 years between major bleaching events ( Fig . 4 ) . At very high bleaching frequencies the total coral cover did not exceed 2% coverage and macroalgae dominated the benthic community . Branching corals dominated if bleaching events occurred every 8 years or more often . Within this range the cover of P . damicornis increased gradually and A . muricata only dominated if extreme temperature events happened between 7 and 9 year intervals . Without the influence of extreme temperature events both applied disturbance intensities triggered nearly the same community responses , although the frequency of the smaller size events had to be far higher for a similar effect ( Fig . 5 ) . Increasing disturbance frequencies abetted dominance shifts from massive to branching growth forms , resembling the pattern of the single effect of bleaching . The interface from the dominance of P . lutea to A . muricata , where the community featured the highest evenness , was restricted to a small range of configurations . At standard frequencies ( i . e . for smaller intensities every 12 , and for larger ones every 60 months ) the total benthic cover was 117% ( see section ‘Environmental settings and scenario conditions’ ) where P . lutea clearly dominated , followed by P . lobata , the two branching coral species covered together ∼3% , and macroalgae covered 0 . 5% . Under highest applied frequencies of mechanical disturbance events the total benthic cover amounted to 16 . 2% , and massive corals nearly disappeared altogether ( <0 . 1% cover ) with only the faster growing branching corals still present in the system . Under the regime of both applied perturbations , where different frequencies of bleaching events were tested under standard mechanical disturbance levels the effect of bleaching was amplified ( Fig . 6 ) . Similar to the single effect scenario the total coral cover was low and macroalgae dominated at very high frequencies of bleaching events . At 20 year intervals , the total benthic cover did not exceed 63% , and as in the assessment of the sole bleaching effect , P . lutea dominated the community , while all other species' coverage stayed below 10% . P . lobata did not exceed 10% cover within any tested frequency . Also contrasting is the behavior of A . muricata . It dominated the community at extreme temperature intervals between 8 and 14 years but its cover decreased tremendously at higher frequencies . P . damicornis again increased its relative cover gradually at high frequencies , but then stayed more or less constant at levels around 5% coverage if major bleaching occurred every 9 years or more seldom . The ratio of massive and branching corals was nearly 4∶5 at 15 year intervals for major bleaching , and the dominance threshold for P . lutea was shifted from 8–9 year intervals up to 14–15 year intervals in the combined perturbations scenarios .
The model outcome reflects findings of empirical studies in many regards and provides interesting insights on the influence of multiple perturbations on coral reef communities . Massive corals are generally slow growing but exhibit a strong physical structure . Provided their tissue is healthy and the defense intact , they are quite resistant to overgrowth by other organisms , like branching coral species or macroalgae which competitively mainly rely on their faster growth rates [63] . In addition , both of the observed massive species feature low susceptibilities to bleaching among the tested corals , and a low bleaching-induced mortality . The combination of slow growth and high endurance implies small fluctuations in relative coverage . As yet another consequence of the above mentioned properties , massive species outcompete their benthic opponents gradually if perturbation levels are low , where P . lutea overrules P . lobata and dominates the community due to its higher growth rate ( see Fig . 4 and Fig . 5 at low perturbation levels ) . This effect is also pronounced by the applied stock-recruitment relationship which leads to a self-enhancing process; i . e . individual colonies grow fast , reach maturity earlier , and produce many propagules again resulting in a higher recruitment rate and new colonies . The branching species A . muricata exhibits the highest fluctuations in population size and relative coverage . It has the fastest growth rate of all simulated species and , given there is enough space , can dominate the benthic community within a few years due to the self-accelerating process produced by the stock-recruitment relationship described above . On the other hand , its bleaching vulnerability to thermal stress is the highest within the tested community , which leads to considerable losses due to extreme temperature events . Acropora and Pocillopora are genera with many species which feature high susceptibility to bleaching [64] . Accordingly , A . muricata is the only species in our model which shows bleaching responses in years when temperatures did not rise as high as in 1998 . For P . damicornis the situation is different . In our model , this species is the least susceptible of all tested species in terms of bleaching , but when a colony bleaches it nearly always dies . Therefore , extreme weather events have nearly analogous effects on the mortality for both branching species . The long term survival of a species is hence strongly dependent on larval input from the outside and thence influenced by asynchronies on the regional and trans-regional scale . The results show that different types and levels of perturbations can lead to very diverse community responses and thereby reef fate . In a mechanical disturbance event , the only reason for an individual to be affected is if it is located at the wrong spot at the wrong time . At higher frequencies of these events the growth rate and recolonization speed of an organism decides winners and losers . Extreme temperature events feature a more selective pattern . Populations of higher susceptibility to thermally induced bleaching suffer the highest losses . Their lower abundance evokes a reduced reproductive output for the next spawning event which , in the long run , might constitute a disadvantage over less susceptible species . Generally , mechanical disturbances occur locally and do not affect the regional coral populations , so that neighboring reefs can serve as a source for new recruits and the local population can recover comparatively fast . An increase of the sea surface temperature is affecting populations at a larger spatial extent and evokes a regional effect . Conspecifics of a sensitive species are affected similarly region-wide if they are not protected from the high temperatures in some way . In such a scenario the larval support between reefs can be hampered tremendously and the risk of extinction increases [65] . Structural complexity of a coral reef is a very sensitive emergent property and high coral cover does not thoroughly imply high ecosystem function ( e . g . 3-dimensional framework ) [66] . Perturbations at too high or too low frequencies cause a loss of biodiversity and thereby structural complexity . Very low frequencies and/or intensities of mechanical disturbance events mostly lead to dominance of species with high competitiveness or endurance and it is just a matter of time until individuals of these species overgrow and displace inferior organisms . In our example , massive corals dominate under very low disturbance regimes . They do not provide as much structure as branching corals and so structural complexity is lost which might cause unfavorable conditions for other reef associated organisms which are deprived of hiding places . One example is Changuu reef , close to the city of Zanzibar , a very exploited site by the means of fishing and under strong influence of waste water , and hence pollutants and nutrients from the town [67]–[69] . Muhando et al . [70] found that corallimorpharians covered 14% of the reef , mainly on the reef crest and flat . Here a large part of the ( still persisting ) coral cover is made up by Galaxea astreata and Porites rus [71] , both of which seem to be quite resistant to environmental change and possess strong competitive traits over other taxa ( e . g . corallimorpharians ) but which facilitate scarce structure . Therefore this site exhibits low biodiversity of fish and other associated organisms . Under very high physical disturbance levels the overall coral cover is generally very low . These conditions are detrimental for massive corals and thus provide free settling ground for branching species . On the other hand branching corals are affected as well , because without shelter from surge or strong currents , they might break off . Nevertheless , due to their high growth rates they play a crucial role in post-disturbance recolonization [72] and reef building event though recovery periods may be too short for slower growing massive coral species . Although branching coral species are displaced by massive species in Indo-Pacific reef sites [73] , this trend cannot only be attributed to the two perturbation types , addressed in this study , but is also owed to the fact that massive species are more resistant to other threats , like pollution [74] or more selective mechanical disturbances like hurricanes , which invoke strong surge and currents [72] . In the given context and despite the limited number of represented species , the principles of the intermediate disturbance hypothesis [75] are well resembled by our model results and in line with previous studies [30] , [50] . At high and low disturbance regimes one species dominates , whereas the evenness is highest at intermediate disturbance levels . In order to investigate the sole effects of perturbations on long term dynamics of a reef without the effect of extinctions , we made two major assumptions . First , we kept the grazing intensity on algae more or less constant , assuming that the herbivore community is completely independent of the structural complexity that the reef is providing . The general view is that the abundance of herbivorous fish and other associated reef organisms is directly influenced by the availability of sheltering places provided by the reef structure [19] , [76] , [77] , and especially by the abundance of branching corals [66] . A loss of a reef's structural complexity therefore reduces herbivore abundance , macroalgae are not grazed efficiently and their populations can proliferate freely . Resulting stands of macroalgae decrease coral recruitment success [78] , [79] which can then produce a feedback in which macroalgae may take over a once coral dominated system . Secondly , we assume a constant larval supply from the outside , on top of the applied stock-recruitment relationship . According to McClanahan et al . [80] the extinction risk of all species tested here as a response to bleaching seems to be very low in the Western Indian Ocean region , with Pocillopora and Acropora showing a probability of 12 and 11% , respectively , and the massive Porites of 7% . In contrast to the observations that McClanahan et al . [81] made in Kenya , where they observed that Pocillopora was completely depleted at protected sites and close to gone from unprotected ones , P . damicornis survives in our simulations . This is due to our second assumption that nearby source reefs were not as much affected by high temperature influences as the focal reef , either because they are located in deeper waters where the heating effect is alleviated or by mixing of water masses . These two assumptions implicate limitations for model extrapolations . To transfer the model to other sites , specific adaptations to the local conditions and to local species parameters have to be made . This also applies to the other explicit and implicit assumptions , underlying the model development , such as the 2-dimensional spatial configuration , the choice of reef components , and their rules of interaction and competition . Additionally , the applied parameter specifications , most of all the relative rates of growth , the reaction to bleaching and larval recruitment , are important for the model outcome . Although there are many types of perturbations , in this article we concentrated on two of the main ones , which concern coral reefs and most probably will have increasing impact in the future; bleaching as a result of global climate change and anchor and boat damage due to an increased demand for food and in consequence more fishing on reefs . The framework of our model allows for the addition of other substantial threats , like ocean acidification , nutrient input ( which hampers coral fitness but may have a positive effect on algae ) , coral diseases , and sedimentation , all of which will affect resource allocation in virtual corals and thereby decrease growth rates and/or competitive strength . Intensive fisheries constitute a fundamental problem for future coral reefs which should be treated with special attention . Enhanced fishing pressure depletes stocks , which most probably results in either higher effort ( i . e . more fishing trips ) and/or the utilization of more efficient fishing gear that often has destructive capacities . Both practices increase the risk of reef degradation ( i . e . by more frequent anchorage and boat damage , or directly through fishing gear ) . In the longer run this might evoke a downward spiral . While we are still far away from the point in which we can represent all features of these highly complex biota , our study extends model capabilities from former coral reef models . This extension improves the accuracy of involved processes as well as on the spatial scale and resolution of the simulated reef system . The model reflects important coral reef dynamics and allows us to test different scenarios relevant for resilience research . Its generic and modular structure and the potential to parameterize different reef components and different coral species , as well as environmental influences provides the possibility for adapting it to represent many reef sites worldwide . The model clearly shows that even when perturbation regimes are kept constant the system never reaches a stable state and stochasticity produces a continuously adapting and fluctuating community response . Pulse events can have a strong influence on ecosystem functioning , especially if they amplify each other's effects or stress the ecosystem in addition to a prevailing chronic disturbance [21] such as increasing nutrient loads or sedimentation . This may cause a shift of general system properties , and lead to , for example , a coral-algae phase shift [82] . Several studies that have quantified resilience with a small number of key parameters have already provided interesting insights [83]–[85] . Nevertheless , measuring resilience in such complex systems as coral reefs is hard to accomplish because the high number of site specific relations and components complicates generalization and extrapolation of specific findings considerably . However , management has a high interest in , and demand for , tools which aid in tracing and identifying distinct drivers for the decline of specific coral reefs . Only a clear recognition of drivers , mechanisms and causes makes it possible to establish or introduce adequate protection measures . Generic frameworks , like the model presented here , allow a fairly easy integration of site specific features and may serve as an appropriate basis for management support tools . | The degradation of coral reefs is a major threat for tropical coastal environments , worldwide . For this reason we developed a spatially explicit model which simulates competition in a benthic reef community under the influence of various environmental factors . Here we highlight the impact of two major perturbation types ( mechanical disturbance events and temperature-induced bleaching events ) on the long-term dynamics of a standard coral reef off Zanzibar Island , Tanzania . While mechanical disturbances are more non-specific and affect all organisms of the reef similarly , temperature-induced bleaching causes selective impact among coral species within the benthic community . Our results show clearly that complex systems which are organized of a multitude of diverse entities and hence feature complex emergent properties need to be analyzed on different integration levels rather than seen as a black box . Our tool may help to disentangle the combined effects of different perturbations and to analyze their respective impact on the benthic community of a coral reef . Hence , it will help to direct future research foci and to coordinate management measures for distinct site specific contexts . | [
"Abstract",
"Introduction",
"Results",
"Discussion"
] | [
"simulation",
"languages",
"community",
"structure",
"computerized",
"simulations",
"marine",
"biology",
"corals",
"coral",
"reefs",
"marine",
"and",
"aquatic",
"sciences",
"biology",
"community",
"ecology",
"computer",
"science",
"programming",
"languages",
"ecology",
"marine",
"ecology",
"species",
"interactions"
] | 2012 | Simulations of Long-Term Community Dynamics in Coral Reefs - How Perturbations Shape Trajectories |
Genome-wide association studies are revolutionizing the search for the genes underlying human complex diseases . The main decisions to be made at the design stage of these studies are the choice of the commercial genotyping chip to be used and the numbers of case and control samples to be genotyped . The most common method of comparing different chips is using a measure of coverage , but this fails to properly account for the effects of sample size , the genetic model of the disease , and linkage disequilibrium between SNPs . In this paper , we argue that the statistical power to detect a causative variant should be the major criterion in study design . Because of the complicated pattern of linkage disequilibrium ( LD ) in the human genome , power cannot be calculated analytically and must instead be assessed by simulation . We describe in detail a method of simulating case-control samples at a set of linked SNPs that replicates the patterns of LD in human populations , and we used it to assess power for a comprehensive set of available genotyping chips . Our results allow us to compare the performance of the chips to detect variants with different effect sizes and allele frequencies , look at how power changes with sample size in different populations or when using multi-marker tags and genotype imputation approaches , and how performance compares to a hypothetical chip that contains every SNP in HapMap . A main conclusion of this study is that marked differences in genome coverage may not translate into appreciable differences in power and that , when taking budgetary considerations into account , the most powerful design may not always correspond to the chip with the highest coverage . We also show that genotype imputation can be used to boost the power of many chips up to the level obtained from a hypothetical “complete” chip containing all the SNPs in HapMap . Our results have been encapsulated into an R software package that allows users to design future association studies and our methods provide a framework with which new chip sets can be evaluated .
The International HapMap project [1] , [2] documented the strong correlations between alleles at polymorphic loci in close physical proximity along human chromosomes . As a consequence it is necessary to genotype only a subset of loci to capture much of the common variation in the genome . Combined with recent technological innovations this observation has made the concept of genome-wide association ( GWA ) studies a reality [3] , [4] . Over the few last years these studies have been very successful in uncovering new disease genes for many different complex diseases [5] . Well over 300 such loci have already been published and many more studies are currently being planned . In the design of such studies two fundamental decisions have to be made: which loci to genotype , and in how many individuals . Both decisions have practical constraints . For example it is currently not possible to assay all known variation in the human genome at a reasonable cost and choices must be made between a set of commercially available genotyping chips . Similarly , sample sizes are often limited by the number of well characterized clinical samples . Therefore , ultimately , the researcher and funding bodies must ask how to use the financial and practical resources available in order to best further the understanding of the genetics of the disease or trait of interest . A primary consideration should be the power of the study: the probability of detecting a variant assumed to be causal . In comparing chips for GWA studies it has been common to ask what proportion of SNPs not directly genotyped are “captured” or “tagged” by the chip , i . e . are well predicted , via LD , by a SNP , or combination of SNPs , on the chip . To do so it is necessary to define the level of prediction required , or equivalently to set a threshold for the required level of correlation . Although arbitrary , this has often been set at 0 . 8 [6] , [7] , [8] . The resulting proportion of SNPs captured at this level is often referred to as the coverage of the chip . Having specified the threshold it is possible to estimate the coverage of a particular chip from HapMap data , although we note that some care is required to account for SNPs not in HapMap [8] . Here we focus instead on the power of particular chips to detect causal variants of different effect sizes , and the way in which this varies with study size and/or study cost and when using genotype imputation methods . Although coverage is straightforward to estimate , power is a complicated function of the set of SNPs on the chip , effect size , and sample size , and can only be assessed by simulation . It turns out that differences in coverage between chips are often not reflected in substantial differences in power and that the use of genotype imputation further reduces these differences . Study power is routinely used throughout science in experimental design and we argue that it should be the primary consideration in designing GWAs . This approach was used in settling several design questions in the Wellcome Trust Case Control Consortium [5] . Our results have been encapsulated in a user-friendly R package that allows the power of different chip and sample size combinations to be assessed given a total budget for the study . Knowledge of study power is also invaluable when analysing data from a study . Assessment of whether positive results at a particular significance level are “real” or due to chance requires knowledge of power [5] , and the practical decision of how far down the list of potential associations one should go in replication studies should be informed by power considerations . Other comparisons of chips have been carried out but have either focussed exclusively on estimating coverage [8] , have been limited in scope of which chips have been evaluated [9] or have used analytical calculations that do not properly take into account the complex LD structure of the human genome [10] , [11] or failed to assess the impact of imputation correctly [11] . A recent paper [12] has used chip data to assess the performance of the chips but the small sample size ( N = 359 ) means that these results cannot be used to assess power of new study designs of more realistic sizes . In addition , the simulations of quantitative phenotypes used the Signal to Noise Ratio ( SNR ) to measure effect size of the causal SNP which is non-standard and difficult to interpret . For binary traits , simulations assumed a disease prevalence of 25% , a relative risk of 3 and a sample size of only 75 cases and 75 controls . These parameter settings are not realistic for genome-wide association studies or useful when designing new studies .
Study power depends on assumptions about the underlying disease model , in addition to effect sizes and sample sizes . When the true causative SNP is not on the genotyping chip there will typically be several SNPs on the chip which are correlated with it . One or more of these could give a signal of significant association and hence allow detection of the locus . The LD structure of the human genome is sufficiently complicated that this effect cannot be captured analytically . It must be assessed via simulation studies . Nonetheless , there is one very simple situation for which analytical calculation is possible and helpful: that of the simplest disease model in which only a single SNP , correlated with the causal variant , is genotyped . For a design with the same number of cases and controls , under the disease model in which disease risk changes multiplicatively with the number of copies of the risk allele carried by an individual ( this model is often referred to as the additive model because risk increases additively on the log scale ) , there is a known analytical relationship [13]: ( 1 ) where χ2 is the chi-squared test statistic , the number of cases and controls , γ the effect size , p the allele frequency of the risk variant and γ2 is the correlation between the marker and causal SNP . Although the real problem is much more complicated than this setting , Equation 1 does provide some useful intuition . Firstly , when the relative effect size is large ( ) the correlation between the marker and causal SNP may only need to be weak ( r2≪0 . 8 ) for the association to be detected ( the expected test statistic is big ) . Equally , if the relative effect size is small ( ) then even strong or complete association ( 0 . 8<r2≤1 ) may not generate sufficient power to reject the null hypothesis of no association . Assessment by simulation of the power of a particular chip requires simulation of large sets of case and control samples which mimic the LD patterns in human populations ( see Figure S1 for an example ) . The approach we use , implemented in a software package called HAPGEN is conceptually simple and is illustrated in Figure 1 . We have previously used this approach to compare different analysis methods and has been briefly describe before [14] . In this paper we provide full details of the approach and these are given in the Methods section . Informally , the required samples are built up from the known haplotypes in HapMap . Consider first the simulation of control samples in a region of the genome . A particular control individual is simulated by separately simulating its two haplotypes in the region . Each of these haplotypes is made up as mosaics of the known haplotypes in HapMap , with the mechanism for constructing these mosaic haplotypes based on population genetics theory . Fine-scale estimates of recombination rates are used to calculate the probability of breaks in the mosaic pattern as one moves along the region . For a given SNP assumed to be causal under a particular disease model and effect sizes , it is straightforward to calculate the genotype frequencies in cases at that SNP . Case samples are simulated separately by first simulating the genotype of the case at the causative SNP and then working outwards in each direction to construct the haplotypes carrying the alleles simulated at the causative SNP . Loosely , this process will result in oversampling of HapMap chromosomes which carry the risk allele , with the effect dropping off as one moves away ( in genetic distance ) from the causative locus ( see Figures S2 , S3 , and S4 for examples ) . We apply the method here to an assessment of the power of different chips , but we note that there are many other settings which require simulations of large case-control samples . These include comparisons of analysis methods [14] and tagging approaches [15] , assessments of parameter estimates , and design questions for follow-on studies such as resequencing and fine mapping of associated regions . We assessed the power of commercially available chips via simulation . Each simulation assumed a particular SNP in HapMap was causative , with a given effect size and used the HAPGEN package to simulate case and control samples of different sizes . In the simulated data we then restrict attention to the genotypes at only the SNPs on the chip in question and ask whether analysis of these would yield a significant result for any of the SNPs on the chip . An estimate of power is obtained by repeating the simulation over a large number of putative disease SNPs across the genome and using the proportion of simulations in which we find a significant test statistic . For definiteness , in the results presented below we simulate data under the additive disease model , and in analysis of the data consider each SNP separately and apply the so-called trend , or Cochran-Armitage test [16] , a chi-squared test with one degree of freedom . We fix a significance level of 5×10−7 , and vary the number of cases and controls in the simulated study . There are various other versions of these assumptions which could be made . We explicitly look at one set of multi-marker tests below and also carry-out a limited set of simulations to assess the impact of genotype imputation . There are two somewhat different perspectives that could be adopted regarding genome-wide association studies . One is to regard the GWAS as a self-contained experiment in its own right with the statistical inference being a formal hypothesis test of the null hypothesis of no association . From this perspective , the goal at the conclusion of the GWAS is to decide whether particular SNPs are , or are not , associated with the phenotype of interest . But this is not what happens in practice . There is a strong consensus in the field that the results of association studies should not be relied upon without additional ( statistically significant ) evidence from analyses in independent replication samples [17] , and many major journals have policies which preclude publication of GWAS studies by themselves , without such replication evidence . Common practice is thus to regard the GWAS as an experiment to highlight SNPs of interest , and then to take as many as possible of the interesting SNPs into replication studies . We adopt the second perspective throughout this paper , and our power calculations are for the probability that for each of the genotyping chips considered , there will be SNPs reaching a prespecified , low , p-value , under specific assumptions about the underlying genetic effects . Given current practice , we believe the right quantity to calculate would be the probability , for the respective chips , effect sizes , and sample sizes , that the experiment would give rise to SNPs showing enough signal to be taken forward for replication . This is ( inevitably ) ill-posed , so we focus instead on a surrogate for it , namely the probability that at least one SNP will have a p-value below a very stringent threshold . In this context there is nothing special about the choice of p-value threshold , and it is now well understood , for example from meta-analyses , that SNPs well down any ranked list of hits from the GWAS associations can still be genuine associations . For definiteness , we focus throughout on the threshold of p<5×10−7 ) . This is deliberately set so that false positive rates will be low – for example , most SNPs with trend test p-values passing this threshold in GWAS studies , including all of those in the WTCCC experiment , have had associations confirmed in replication studies ( see [18] and the NHGRI Catalog of Published Genome-Wide Association Studies at http://www . genome . gov/GWAStudies/ ) . Choice of a different p-value threshold changes the numerical value of the power we calculate , but does not affect the relative performance of the chips , or the relative effect of sample size ( data not shown ) . If one were to adopt the first of the two perspectives on a GWAS study , namely that it is a formal statistical hypothesis test in its own right , then power comparisons become more complicated , at least under a frequentist statistical perspective: for a given nominal per-SNP significance level , the overall GWAS experiment will have somewhat different false positive rates for the different commercial chips , because they have different SNP sets , or when some SNP genotypes are imputed , depending on the number of imputed SNPs , for the same reason . Actually , even for a fixed chip , overall false positive rates will differ depending on the population in which the GWAS is conducted , because of differing patterns of LD between the SNPs on the chip ( and hence different effective numbers of independent tests ) . We do not pursue this approach here , principally because it does not reflect the way GWAS experiments are typically used in practice: regardless of the genotyping chip used , whether or not genotype imputation is employed , and the population studied , researchers tend to focus on the most significant SNPs after the GWAS and try to confirm that they are real in replication studies . In addition , as noted above , overall GWAS false positive rates are low , for any of the commercial chips , at the very low per-SNP significance level we consider . Nonetheless , in what follows , readers should be aware that we are comparing power , defined here as the probability that at least one SNP reaches a fixed p-value threshold under specific assumptions about design and effect sizes , across settings in which these very low false positive rates will differ between chips ( and across populations ) . In calculating power , as thus defined , we simulate data under the assumption that a particular allele is causal and then look to see whether any SNPs on the respective genotyping chip , within a large region around the causal SNP attain the specified significance level . In ignoring the SNPs on the chip elsewhere in the genome , this approximation will underestimate the probability of there being a SNP meeting the significance threshold , but at the very low threshold , the probability of there being a SNP elsewhere in the genome meeting the threshold is extremely small , so that effect of this approximation will be minimal and our power calculations based on only on SNPs within the 1Mb region containing the causal SNP will be very close to the true values . We simulated putative disease loci at SNPs in phase II of the HapMap within twenty-two one megabase regions on each of the autosomes , a total of nearly 50 , 000 SNPs , which together are typical for the genome in terms of SNP coverage and recombination rates ( see Figure S5 and Text S1 for details ) . We investigated the power afforded by seven different genotyping chips: the 100 k , 500 k and 6 . 0 chips from Affymetrix ( www . affymetrix . com ) and the 300 k , 610 k , 650 k and 1 M chips from Illumina ( www . illumina . com ) . These chips sets differ in the way in which the SNPs are chosen and the total number of SNPs assayed . As technology develops and genotyping chips become denser it is a natural question to ask how much power would be gained by genotyping additional SNPs or by using genotyping imputation methods [14] . To facilitate such comparisons we evaluated the performance of a hypothetical chip that contains all the SNPs in HapMap to act as a point of reference in our results . The performance of this ‘complete’ chip is shown as a solid black line in all of the figures showing power . Since the simulations we carry out only use HapMap SNPs as causal SNPs this analysis approximates the scenario in which we have a chip which types all possible SNP variation . We return below to consideration of results for studies in the Yoruban population . Focussing now on the power curves in the top row of Figure 2 several features are evident . The first is the profound effect of sample size . Effect sizes of 1 . 5 or smaller might be typical of what would now be expected for most variants affecting susceptibility to common human diseases [5] . For effect sizes at the top of this range ( 1 . 3–1 . 5 ) very large studies ( say 2 , 000–3 , 000 cases and the same number of controls ) are needed to have reasonable power , while for smaller effect sizes even studies of 5000 cases and 5000 controls have very little power . This ties in with growing empirical evidence . For example , for Crohn's disease , the WTCCC study , of 2000 cases and 3000 controls found 9 loci with p<5×10−7 , whereas several smaller studies published around the same time each found only one or two of the loci , with little overlap across these smaller studies , consistent with each having modest power for the larger set of loci . Further , recent meta-analyses of 4 , 539 cases for type 2 diabetes and 3 , 230 for Crohn's disease have been needed to discover further loci with estimated effect sizes in the range 1 . 1–1 . 2 . Even for a disease not previously studied by GWA , studies with fewer than 2000 cases and 2000 controls will have low power , except in special circumstances , for example if there are loci with larger effect sizes than has been typical across many other diseases . A second general feature of the power curves for Caucasian studies in Figure 2 is that aside from the Affymetrix 100 K chip ( which is no longer available ) , there are not major differences in power across the other seven chips . For Caucasian samples the chips are typically ordered ( with decreasing power ) : Illumina 1M , Illumina 650 k , Illumina 610 k , Affymetrix 6 . 0 , Illumina 300 k , Affymetrix 500 k , but the absolute difference in power between the best and worst of these chips is often no more than around 10% . Put another way , for effect sizes in the range 1 . 3–1 . 5 , a study with the Affymetrix 500 K chip would have the same power as one with the Illumina 1 M chip if its sample size were larger by 10–20% , with smaller increases in sample sizes giving studies with other chips the same power . Further , in Caucasian studies , power for all chips other than the Affymetrix 100 K chip is quite close to the best which could be obtained , namely by directly genotyping the causative SNP . Equation 1 makes clear the dependence of power on the frequency of the risk allele . The results in Figure 2 are averaged over putative causative SNPs with a risk allele frequency ( RAF ) in the range 5–95% . Figure 3 shows that this hides quite different behaviour depending on whether the putative disease SNP is rare or common , and that the conclusions in the preceding subsection apply principally for common causative SNPs . The Figure shows a substantial difference in power for common and rare alleles with the same effect size and that power is minimal for the rare alleles when the effect size is small . These results refer to single-SNP analyses . While there are definitely more powerful analysis methods for rare alleles [14] , this is not a major factor in the loss of power , and neither is the incomplete coverage of the SNPs on the commercially available chips: even using a sample size of 3000 cases and controls and genotyping the causal locus directly ( black line ) is unlikely to lead to a test statistic which will reach the small levels of significance thought appropriate for GWAS . There is an open question as to whether rarer causal alleles might have larger effect sizes than common causal alleles . If this were though plausible , then in assessing power overall for a particular chip , one could focus in Figure 3 on particular ranges of effect sizes for common causative alleles and a different range of effect sizes for rarer causative alleles . It is becoming clear that many loci harbouring common alleles affecting common diseases will have effect sizes in the range 1 . 1–1 . 2 , and our simulations demonstrate that there is almost no power to detect these in studies of the size currently underway . As has already been shown empirically [19] , [20] these loci can be found by meta-analyses and follow up in larger samples of GWA findings . Slightly larger relative risks do become detectable in large samples . For example the power to detect an effect of size 1 . 3 jumps from almost zero with 1000 cases and 1000 controls to over 50% in a study three times the size . Figure 3 also demonstrates that chip sets differ in the power they offer to detect associations at different frequencies . Most noticeably , when averaged over common alleles the Illumina 300 k chip set offers more power than Affymetrix 500 k . For rare alleles , the opposite is true with the Affymetrix 500 k chip having more power than the Illumina 610 k chip . This is most likely due to the way in which the Illumina SNP sets have been designed to specifically tag the common variation present in the HapMap panels . Immediately apparent is how close , for studies in Caucasian populations , the genotyping chips track the power afforded by the ideal “Complete chip” in a given study design and disease model . Figure 3 illustrates that the potential benefits of increasing SNP density on the chips or from using imputation [14] are greatest for low frequency SNPs . When focusing on common alleles , the potential benefits are greatest for the Affymetrix 100 k and 500 k chips and the Illumina 300 k chip and we show this when specifically consider imputation below ( see Table 1 ) . However , a clear consequence of these results is that for any of the chips in current use , increasing sample size is likely to have a bigger effect on power than increasing SNP density . A striking feature of Figures 2 and 3 is that substantial differences in coverage between different chips do not translate into big differences in power . Put another way , coverage is often a poor surrogate for power . As an example , the coverage in the CEU HapMap population ( r2≥0 . 8 ) provided by the Affymetrix 500 k and Illumina 610 k chips are 65% and 87% respectively , a difference of 22% . On the other hand , the difference in power e . g . for relative risk 1 . 5 and 1500 cases and controls , is only 7% ( 66% and 73% respectively ) . In one sense this shouldn't be surprising . Coverage is measured to a hard threshold: so if SNP has r2 of 0 . 85 to its best proxy on one chip and 0 . 75 to its best proxy on another chip , it will be counted as “covered” by one chip but not by the other , whereas the difference in power is small . Coverage statistics also do not depend on study size or disease model . Figure 4 illustrates the differences in correlation structure for two chips . For each HapMap SNP we found it's best “tag” ( the SNP on the chip with which it has the highest r2 ) and generated a histogram of these maximized r2 values . To recover coverage we simply count the proportion of SNPs for which the best tag r2 is ≥0 . 8 , coloured red in the bottom row of figure 4 . In this sense , informally , it is useful to think of coverage as assuming that there is power one for every “tagged” SNP and no power for every other SNP . This is of course false , in ways which help to explain why coverage differences do not translate into power differences . When a SNP is common and the effect size is moderate or large , there will still be good power to detect it even if the best SNP on the chip only has r2 = 0 . 5 or less . At the other extreme , for rare SNPs , unless the effect size is very large , power would be low even if the SNP had a perfect proxy on the chip . Thus even if these SNPs were well covered by one chip and completely missed by another they would not contribute to a difference in power between the chips because both chips would have power close to zero for them . The top row of Figure 4 shows the average power for SNPs in each LD bin . For the Affymetrix 500 K chip , there is a greater contribution to power from the sets of SNPs which are not well “covered” , than for the Illumina chip , and hance a smaller difference in power than in coverage . For several reasons it is of interest to study the power of commercially available chips in different populations . Firstly the Illumina 100 k , 300 k and 610 k chips are aimed at capturing variation in the CEU population , whereas the Affymetrix 500 k chip is not designed with a specific population in mind . Furthermore the Illimina 650 k chip has a subset of SNPs targeted at capturing variation in the HapMap YRI ( Yoruba , Africa ) population . LD will not extend as far in the YRI collection [1] as in the CEU , reducing the coverage of a given set of SNPs . Figures 2 and 3 show the results of power calculation using the distribution of diversity in both the HapMap CEU and YRI populations . The results show that the increased ancestral recombination leads to a loss of power and coverage across all chips for a range of study designs . The difference between the power available from commercial genotyping chips and that achievable by exhaustively assaying all SNPs shows that increasing marker density may yield a better return than a similar approach in non-African populations . The Illumina 650 k chip , with the YRI fill-in illustrates these potential benefits , showing a marked increase in power over the 610 k . However the performance of the Illumina 300 k chip , designed using the CEU HapMap , falls below the Affymetrix 500 k when genetic diversity is modelled on the YRI HapMap panel . It is not yet clear how closely patterns of diversity and LD in other African populations mimic those in the Yoruba , and hence to what extent the power results will translate to studies in other populations . One general point is that the Illumina 650 k chip was designed specifically to capture common Yoruban variation , so one might expect power for this chip to decrease in other African populations , for which it is not specifically designed . On the other hand , the Affymetrix 500 k chip was not designed using this data , so there would be not a systematic effect changing power estimates for other African populations . As a consequence , differences in power between the Illumina 650 k chip and Affymetrix chips may well be smaller in other African populations . Multi-marker methods , which use combinations of SNPs , have been suggested as an efficient way to increase both coverage and power [15] . Figures S6 and S7 show the results of simulations that implement the multi-marker tests . In these figures the dotted lines , which represent coverage , are higher for all chips in comparison to single marker approaches ( Figures 2 and 3 ) consistent with previous observations . We find that multi-marker approaches also increase statistical power to detect disease loci , but that the increase is modest relative to coverage , and the broad conclusions above are not much affected . Interestingly , when comparing across genotyping platforms , we find for example that the Affymetrix 500 k chip gains more by combining SNPs than the Illumina 300 k chip . Genotype imputation methods [21] , [14] are now being widely used in the analysis of genome-wide association studies [5] and meta-analysis of such studies [22] , [20] . These methods can be thought of as a more sophisticated version of Multi Marker tests but are relatively much more computationally demanding . We carried out an evaluation of the boost in power that can be gained by imputation using the program IMPUTE [14] . For our simulations with a sample size of 2000 cases and 2000 controls and a relative of the causal SNP or 1 . 3 we ran IMPUTE on the genotype data from each of the chips under study using the CEU HapMap as the basis for imputation . We then carried out a test of association at all the imputed SNPs in addition to the SNPs on each chip . We used our program SNPTEST to carry out tests of association at imputed SNPs to properly account for the uncertainty that can occur at such SNPs[14] . The results of the simulations are shown in Table 1 and shows that the use of IMPUTE provides a noticeable boost in power over testing just the SNPs on each chip or using Multi Marker tests ( as defined in [15] ) . This agrees with our previous results [14] . It is also very noticeable that imputation reduces the differences in power between the chips and that the use of imputation produces a level of power that is almost as high as our hypothetical ‘complete’ chip . We also note that the boost in power is more substantial than that estimated in another recent study [11] . A close look at the details of this other study shows that the only imputed SNPs used were those ( a ) which had real genotype data from one of the other chips , and ( b ) the imputed and real data at the SNP agreed with an r2>0 . 8 . So for example , for the Affy 500 k chip only genotypes at 427 , 838 imputed SNPs were used , rather than all those available from HapMap ( approximately 2 . 5 milion SNPs ) , as normal practice when carrying out imputation . Using such a filter clearly creates a bias towards imputed SNPs that are almost perfect tags for SNPs on the chip so it is not surprising that this study shows such small increases in power when using imputation . One option open to researchers who would like to increase power in the context of limited case series is just to increase the control collection . This strategy might include using cases for one disease as extra controls for another ( assuming suitably different disease aetiologies and similar population history ) . We investigated the utility of such an approach by performing simulations with 1000 cases and an increasing number of control ( Figure S8 ) . Although the gains are not as strong as increasing both the case and control sample sizes ( Figure 2 ) , the ability to reject the null hypothesis of no association increase considerably with the size of the control panel . For example , adding an extra 2000 controls to a case-control study with sample size 1000–1000 increases power to detect an effect of 1 . 5 typically by 20% . Subject to care in their use , the growing availability of genotyped sets of controls promises to make this a possibility worth investigating for many studies . The results of our simulations can be used to assess the power of a range of possible designs for a given budget and have been encapsulated in a user friendly R package for this purpose ( see Software section ) . Table 2 shows the study size and power that can be achieved on a budget of $2 , 000 , 000 for each of the chips assuming the disease causing allele of has a relative risk of 1 . 5 , a risk allele frequency of at least 0 . 05 and that a p-value threshold of 5×10−7 is used to define power . Since the different chips vary in their prices and their per sample processing costs we obtained quotes from service providers for the various chips and averaged them ( see Text S1 ) . The prices were based on quotes for 4000 chips and quotes were converted to US dollars using current exchange rates where necessary . We obtained 5 different quotes for the Affymetrix chips and 6 different quotes for the Illumina chips . The results show that in this scenario the Illumina 300 k chip produces the most powerful design ( 82 . 1% ) primarily due to its relatively cheap price compared to the other chips . Using the same sample size ( 2653 cases and controls ) the ‘Complete’ chip has a power of 88 . 1% . It is also notable that the power of thie Illumina 300 k chip is nearly 17% greater than the power that can be achieved by the Illumina1 M chip ( 63 . 5% ) which has approximately 3 times the SNP density . These result further illustrate the deficiencies in using coverage as a measure of chip performance as sample size is not factored into the calculation . Although these results are interesting we advise against using them directly in the design of a new study . There were noticeable variations in the quotes we obtained from the service providers and prices are likely to change through time . We encourage new studies to re-calculate power of various designs based on a set of up to date and competitive prices and to take into account the general effect that genotype imputation can have on these power estimates .
Because of the complexity of human LD patterns , many questions of interest cannot be addressed analytically . We have described in detail our simulation method , HAPGEN , for generating large samples of case and control data at every HapMap SNP , which mimic the patterns of diversity and LD present in the HapMap data . The software can simulate case data under a single causal disease SNP model for specified genotypic relative risks . We have used the method here to assess the power of various commercially available genotyping chips for case-control genome-wide association studies , but note that it could be utilised to assess other design questions , in the evaluation of analytical methods , and in considering follow-on studies such as resequencing and fine-mapping . In Caucasian populations the differences in power afforded by current-generation genotyping chips are not large , and the power of these chips is close to that of an optimal chip which always directly genotyped the causal SNP . Listed in order of decreasing power for the CEU population , averaged over all potential disease SNPs with RAF ≥5% , the chips we considered were: Illumina 1M , Illumina 650 k , Illumina 610 k , Affymetrix 6 . 0 , Illumina 300 k , Affymetrix 500 k and Affymetrix 100 k . In line with our previous work we have shown that imputation can boost the power of each chip substantially and that the resulting power will approach that which could be obtained by a hypothetical ‘complete’ chip that types all the SNPs in HapMap . One limitation of the approach we ( and others [9] , [10] , [12] , [11] ) have used is that the causal SNP is assumed to be one of those SNPs in the HapMap panel and this will not always be true . Other studies [1] have shown that the majority of SNPs not in HapMap will be highly correlated with the SNPs that are in HapMap and this is especially true for the more common SNPs . This means there is a slight bias in our power results for each chip and for the use of imputation but we do not expect it to be large . A consequence of this point is that the power we estimate for the ‘complete’ chip approximates the power we might obtain if we had a chip which typed all the SNPs that exist in the human genome . A main conclusion from our analysis is that study size is a crucial determinant of the power to detect a causal variant . Increasing study size typically has a larger effect on power than increasing the number or coverage of SNPs on the chip , at least amongst chips currently available . Even for effect sizes at the larger end of those estimated to date for common human diseases ( RRs of 1 . 3–1 . 5 ) quite large sample sizes , at least 2000 cases and 2000 controls and ideally more , are needed to give good power to detect the causal variant . When case numbers are limited , there are still non-trivial gains in power available from increasing just the number of controls . Care is needed in assessing the appropriateness of a set of controls , but as larger sets of control genotypes are made publicly available this strategy has considerable appeal , whatever the number of available cases . SNPs with smaller effect sizes are unlikely to be detected even in studies of the sizes currently undertaken , but as has been shown empirically for several diseases , these can be found by meta-analyses which combine different GWAs , or by follow-up in large samples of SNPs which look promising in the original GWA but fail to meet the low levels of significance thought appropriate for GWAS . When the causal SNP is rare ( MAF<10% ) , all chips have low power unless its effect is large and sample sizes are large . This conclusion would hold even if the chip directly genotyped the causal SNP . The relative ordering of different chips , on the basis of power , also changes in this context . As would be expected , power is also lower for all chips for samples which match the patterns of LD seen in the Yoruba HapMap sample , and again the relative ordering of chips changes in this setting . It is not yet clear how well the results for the Yoruba would extend to other African populations . An often-quoted metric in assessing chips is the coverage of each chip: an estimate of the proportion of SNPs which have r2>0 . 8 with at least one SNP on the chip . Although relatively simple to calculate ( and even simpler to miscalculate ) , not least because it does not depend on study size , our results show that coverage can be a poor surrogate for power , and that relatively large differences between chips in coverage do not translate to large differences in power . The sets of SNPs on Illumina chips are chosen in part to maximize particular criteria , such as coverage , for certain populations , typically those in HapMap . One difficulty of analyses such as those in this paper is that these resources are also the natural ones with which to assess properties of the chips . Thus when Illumina chips “tuned” to one population ( say the 610 K chip for CEU ) are used in other populations , power might be systematically lower than the levels assessed here . In contrast , SNP sets of Affymetrix chips are chosen largely in a non-population specific way . While power is likely to vary in populations other than those we have considered here , there is not the same systematic effect which would lead to a decrease in power . A quantitative assessment of this phenomena will be possible when dense genotype data is available for other populations , such HapMap Phase 3 . We have assumed here that accurate genotypes are available for all SNPs on each chip . In practice some SNPs on each chip will fail QC tests and not be available for analyses . As a consequence , our study will overestimate power , though this effect is unlikely to be large . We are only able to use SNPs in HapMap as potential disease SNPs . These may not be systematically representative of all potential disease SNPs . HapMap SNPs have systematically higher MAFs than do arbitrary SNPs [2] , but for SNPs within a particular range of MAF , it seems unlikely that their LD properties will differ systematically , so , for example , we would expect our results for common SNPs to extend beyond those in HapMap . We have focussed on the most common GWA design , namely of a single-stage study , and the simplest disease model . The flexibility of the simulation approach allows many other practical aspects of study design to be incorporated into power calculations . These include more complex disease models , two-stage strategies ( the starting point for our work was a comparison of power for one- and two-stage designs in the context of the WTCCC study [5] ) , genotyping errors , QC filters , misidentification of cases as controls and simple types of population structure . The HAPGEN software also provides a useful tool for the development and comparison of more sophisticated multi-marker approaches to detecting disease association ( e . g . imputation [14] ) . We therefore believe that simulations are an essential tool in the design of association studies by allowing a focus on study power and an assessment of the affect on power of following a given study design . We hope that this method will continue to find use and can be extended to new catalogs of genetic variation such as the 1000 Genomes Project http://www . 1000genomes . org/ . As in other areas of science , power seems a central consideration in study design and choice of genotyping chip . But other issues may also play a role . These include coverage of particular genes , or genomic regions of interest; the utility of GWA data for directing downstream studies such as resequencing and fine mapping; data quality for particular chips; and the extent to which a chip reliably assays other forms of genetic variation such as copy number polymorphisms . Adding data to existing studies is straightforward if the same chip is used , but the success of imputation methods , in particular in meta-analyses [19] , [20] means that this is not essential . In general , Affymetrix chips have more redundancy than do Illumina chips , in the sense of containing sets of SNPs which are correlated with each other . The immediate consequence of this is lower coverage and lower power for the same number of SNPs , but there can be advantages to this redundancy: loss of a particular SNP to QC filters may not be as costly; and signals of association are likely to include more SNPs , thus making them easier to distinguish from genotyping artefacts . Ultimately power can only be calculated under an alternative model . Thus on a practical level the optimal choice of assays and sample sizes will actually depend on the researcher's belief regarding the unknown distribution of effect sizes and models relating genotype and phenotype . In particular we show that one might adopt different strategies depending on the expected frequency of disease causing variant , the effect size and even the population from which cases and controls are sampled ( Figure 3 ) . In the continuing search to better understand the genetic basis of common human diseases , numerous study designs can be adopted which may involve combining data sets , imputing missing SNPs [14] , distilling signals of association over multiple experimental stages , and so on . In this complex setting study power will remain a central criterion in study design , and the kinds of approaches developed here will continue to allow informed decision making by experimenters .
We adopt the model introduced by [23] ( denoted LS from now on ) , who described a new model for linkage disequilibrium , which enjoys many of the advantages of coalescent-based methods ( e . g . it directly relates LD patterns to the underlying recombination rate ) while remaining computationally tractable for huge genomic regions , up to entire chromosomes . Their model relates the distribution of sampled haplotypes to the underlying recombination rate , by exploiting the identity ( 2 ) where h1 , … , hn denote the n sampled haplotypes , and ρ denotes the recombination parameter ( which may be a vector of parameters if the recombination rate is allowed to vary along the region ) . This identity expresses the unknown probability distribution on the left as a product of conditional distributions on the right . LS substitute an approximation for these conditional distributions into the right hand side of ( 3 ) , to obtain an approximation to the distribution of the haplotypes h given ρ ( 3 ) If h1 , … , hn are n sampled haplotypes typed at S bi-allelic loci ( SNPs ) LS modelled the distribution of the first haplotype as independent of ρ , i . e . all 2S possible haplotypes are equally likely , so . For the conditional distribution of hk+1 given h1 , … , hk , LS modelled hk+1 as an imperfect mosaic of h1 , … , hk through the use of a Hidden Markov Model ( HMM ) . That is , at each SNP , hk+1 is a ( possibly imperfect ) copy of one of h1 , … , hk at that position where where the transition rates between the hidden copying states are parameterized in terms of the underlying recombination rate . The transition rates are different for each of the conditional distributions in such a way so as to mimic the property that as we condition on an increasingly larger number of haplotypes we expect to see fewer novel recombinant haplotypes . A parameterisation for the mutation rate ( or emission probabilities of the HMM ) is used that has similar properties ( see [23] for more details ) . The simulation of a new set of haplotypes for control and case individuals is proceeds using the following algorithm . 1 . Pick a locus from the set of markers in the real dataset as the disease locus . The disease locus is chosen at random from all those loci with a minor allele frequency ( MAF ) within some specified range [l , u] . We use to denote the disease locus , a and A to denote the major and minor alleles at the disease locus and use p denote the sample minor allele frequency at this locus . 2 . For a given disease model simulate the alleles at the disease locus of the new individual conditional upon case-control status . At the disease locus we use a general genotype model in which the frequencies of the genotypes aa , Aa and AA in control individuals are given by ( 1−p ) 2 , 2p ( 1−p ) and p2 respectively . This assumes that the control individuals are so-called population controls ( as used by the WTCCC study [5] ) rather than individuals who have been selected to specifically not have the disease . For case individuals the genotype frequencies are determined by specification of the two relative risks ( 4 ) where denotes the probability that an individual is a case conditional upon having genotype g . Under this model ( 5 ) where γ = ( 1−p ) 2+2αp ( 1−p ) +βp2 . As an example , if p = 0 . 1 , α = 2 and β = 4 the control and case genotype frequencies are ( 0 . 81 , 0 . 18 , 0 . 01 ) and ( 0 . 67 , 0 . 30 , 0 . 03 ) respectively . Assuming we have a set of k known haplotypes , the generation of a case ( control ) starts by simulating a genotype g using the case ( control ) genotype frequencies . This simulated genotype specifies the alleles on the the two haplotypes of the new individual at the disease locus . For example , if g = Aa then hk+1 , d = 1 and hk+2 , d = 0 . 3 . This step involves the simulation of two new haplotypes for the individual conditional upon the alleles simulated at the disease locus in Step 2 and conditional upon the fine-scale recombination map across the region . This involves simulating the rest of hk+1 and hk+2 . We only describe the generation of sites right flanking of the disease locus as the generation of the left flanking markers is virtually identical . Also the simulation of hk+2 follows directly from our description of how the rest of hk+1 is simulated . Let Xj be the hidden state of the HMM that denotes which haplotype hk+1 copies at site j ( so that ) . This state variable is initialized at the disease locus as follows ( 6 ) The value of , as with LS , is Watterson's point estimate ( Watterson , 1975 ) ( 7 ) Simulation of the hidden state of the HMM then proceeds using the following transition rule ( 8 ) where zj is the physical distance between markers j and j+1 ( assumed known ) ; and , where is the effective ( diploid ) population size , and cj is the average rate of crossover per unit physical distance , per meiosis , between sites j and j+1 ( so that cjzj is the genetic distance between sites j and j+1 ) . This transition matrix captures the idea that , if sites j and j+1 are a small genetic distance apart ( i . e . cjzj is small ) then they are highly likely to copy the same chromosome ( i . e . Xj+1 = Xj ) . To mimic the effects of mutation the copying process may be imperfect: with probability k/ ( k+θ ) the copy is exact , while with probability θ/ ( k+θ ) a mutation will be applied to the copied haplotype . Specifically , 4 . Return to step 2 to generate another individual or terminate . Illustrations of the HAPGEN method in practice and details of the testing the method against coalecent simulations are given in Text S1 . We used release 21 of the HapMap data for which phased haplotypes are available in NCBI b35 coordinates . The SNPs that occur on each genotyping chip were obtained from the websites of Affymetrix and Illumina respectively . Some of the SNPs in these sets do not occur in the HapMap phased haplotype data due to QC measures applied to the raw genotype data . For the Affymetrix 6 . 0 and Illumina 1 M chips 90 . 8% and 88 . 1% of the SNPs on these chips respectively are in this release in HapMap . This will have the effect of making our estimates of power slight underestimates of the true power . We simulated data for twenty-two one megabase regions chosen at random , one from each autosome . To ensure that the regions used to approximate genome-wide power were representative of the genome at large we their SNP density . Figure S5 plots the distribution of inter-SNP distances within the 22 analysis regions and across the whole genome for three of the genotyping chips analyzed . The close match between the distribution , both on the physical scale and in terms of genetic distance suggests that our results are insensitive to the regions we chose to simulate , and can be used to make comparisons of genotyping chips genome-wide . We used data from the HapMap project Phase II to estimate coverage . Single marker coverage was defined to be the proportion of all variation ( with minor allele frequency greater than 5% ) in r2 with a SNP on the genotyping chip above 0 . 8 . Using this definition we achieved very similar estimates to previous studies which used the whole genome ( we use twenty two representative megabases ) . Multi-marker coverage was calculated by an aggressive search of all 2-SNP and 3 SNP haplotypes within 250kb of the SNP being tagged [6] . The SNP was tagged if any of these multi-marker tags had r2 above 0 . 8 , the rule defining the haplotype was also stored and added to the list of multi-marker tests . Single marker tests ( Cochran-Armitage test ) were performed at each SNP on the genotyping chip where information were simulated from the relevant HapMap panel . Multi-marker tests of association were performed in an identical fashion with the marker being formed by the multi-marker haplotypes known to tag HapMap variation . To avoid over estimation of power , multi-marker tags chosen to tag the current putative disease SNP in the simulations were excluded from the test set . Tests at imputed SNPs took account of the uncertainty in genotypes through a missing data likelihood as described in [14] . The HAPGEN software is freely available for academic use from the website http://www . stats . ox . ac . uk/̃marchini/software/gwas/gwas . html . In addition , the results of the power calculations for the 7 commercially available genotyping chips have been included in an R package called GWASpower available from http://www . stats . ox . ac . uk/̃marchini/#software . This package allows the user to determine the most powerful study design for a given budget . As new commercial genotyping chips become available we will update the package to include results of new chips . The package works by fitting a Generalised Linear Model to the results of the simulation study and using the model fit to predict the power for a given number of cases and controls . | Genome-wide association studies are a powerful and now widely-used method for finding genetic variants that increase the risk of developing particular diseases . These studies are complex and must be planned carefully in order to maximize the probability of finding novel associations . The main design choices to be made relate to sample sizes and choice of commercially available genotyping chip and are often constrained by cost , which can currently be as much as several million dollars . No comprehensive comparisons of chips based on their power for different sample sizes or for fixed study cost are currently available . We describe in detail a method for simulating large genome-wide association samples that accounts for the complex correlations between SNPs due to LD , and we used this method to assess the power of current genotyping chips . Our results highlight the differences between the chips under a range of plausible scenarios , and we demonstrate how our results can be used to design a study with a budget constraint . We also show how genotype imputation can be used to boost the power of each chip and that this method decreases the differences between the chips . Our simulation method and software for comparing power are being made available so that future association studies can be designed in a principled fashion . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/complex",
"traits",
"genetics",
"and",
"genomics/bioinformatics",
"genetics",
"and",
"genomics/population",
"genetics"
] | 2009 | Designing Genome-Wide Association Studies: Sample Size, Power, Imputation, and the Choice of Genotyping Chip |
Anti-tumor necrosis factor alpha ( anti-TNF ) biologic therapy is a widely used treatment for rheumatoid arthritis ( RA ) . It is unknown why some RA patients fail to respond adequately to anti-TNF therapy , which limits the development of clinical biomarkers to predict response or new drugs to target refractory cases . To understand the biological basis of response to anti-TNF therapy , we conducted a genome-wide association study ( GWAS ) meta-analysis of more than 2 million common variants in 2 , 706 RA patients from 13 different collections . Patients were treated with one of three anti-TNF medications: etanercept ( n = 733 ) , infliximab ( n = 894 ) , or adalimumab ( n = 1 , 071 ) . We identified a SNP ( rs6427528 ) at the 1q23 locus that was associated with change in disease activity score ( ΔDAS ) in the etanercept subset of patients ( P = 8×10−8 ) , but not in the infliximab or adalimumab subsets ( P>0 . 05 ) . The SNP is predicted to disrupt transcription factor binding site motifs in the 3′ UTR of an immune-related gene , CD84 , and the allele associated with better response to etanercept was associated with higher CD84 gene expression in peripheral blood mononuclear cells ( P = 1×10−11 in 228 non-RA patients and P = 0 . 004 in 132 RA patients ) . Consistent with the genetic findings , higher CD84 gene expression correlated with lower cross-sectional DAS ( P = 0 . 02 , n = 210 ) and showed a non-significant trend for better ΔDAS in a subset of RA patients with gene expression data ( n = 31 , etanercept-treated ) . A small , multi-ethnic replication showed a non-significant trend towards an association among etanercept-treated RA patients of Portuguese ancestry ( n = 139 , P = 0 . 4 ) , but no association among patients of Japanese ancestry ( n = 151 , P = 0 . 8 ) . Our study demonstrates that an allele associated with response to etanercept therapy is also associated with CD84 gene expression , and further that CD84 expression correlates with disease activity . These findings support a model in which CD84 genotypes and/or expression may serve as a useful biomarker for response to etanercept treatment in RA patients of European ancestry .
Rheumatoid arthritis ( RA ) is an autoimmune disease characterized by chronic inflammation of the synovial lining of the joint [1] . If left untreated , outcome varies from self-limited disease in a small proportion of RA patients to severe disease resulting in profound structural damage , excess morbidity and disability , and early mortality [2] . In the last twenty years , disease activity has been controlled in many patients by treatment with disease-modifying anti-rheumatic drugs ( DMARDs ) , such as methotrexate , and the more recently developed biologic DMARDs that block inflammatory cytokines such as tumor necrosis factor-alpha ( TNFa ) [3] . Unfortunately , these medications are not effective in all RA patients , with up to one-third of patients failing to respond to any single DMARD [1]–[3] . Moreover , the biological mechanisms underlying treatment failure are unknown , which limits the development of clinical biomarkers to guide DMARD therapy or the development of new drugs to target refractory cases . There are two classes of anti-TNF therapy: the TNF receptor fusion protein ( etanercept ) , which acts as a soluble receptor to bind circulating cytokine and prevent TNF from binding to its cell surface receptor , and monoclonal antibodies that bind TNF ( adalimumab , infliximab , certolizumab , and golimumab ) . There are undoubtedly shared mechanisms between the two drug classes ( e . g . , downstream signaling factors ) , as illustrated by similar effects on the change in inflammatory cytokines , complement activation , lymphocyte trafficking , and apoptosis [4] , [5] , [6] . Similarly , there are likely to be different biological factors that influence response: infliximab and adalimumab are approved for treatment of Crohn's disease; infliximab and adalimumab bind to transmembrane TNF on the surface of activated immune cells , whereas etanercept only binds soluble TNF [7]; and etanercept also binds a related molecule , lymphotoxin alpha ( LTA ) , whereas infliximab/adalimumab do not [8] . Pharmacogenetics of response to anti-TNF therapy in RA remains in its early stages , with no single variant reaching an unambiguous level of statistical significance . Candidate gene studies suggest associations of TNFa or TNF receptor alleles , RA risk alleles or other SNPs with response to anti-TNF therapy [9] , [10] , [11] . Two GWAS in small sample sets ( largest was 566 patients ) have been performed , which identified loci with suggestive evidence for association [12] , [13] . Therefore , GWAS of large sample sizes may yet uncover genetic factors associated with response to anti-TNF therapy in RA , and larger cohorts enable separate analyses of the different types of anti-TNF drugs . Here we report a GWAS of 2 , 706 samples with anti-TNF treatment response data collected from an international collaboration , including previously published GWAS data [12] , [13] . Our primary outcome measure was the change in disease activity score based on a joint count in 28 joints ( DAS28 ) from baseline to 3–12 months after initiating anti-TNF therapy . Our secondary outcome measure was European League Against Rheumatism ( EULAR ) responder status [14] , [15] , where patients are classified as EULAR good responders , moderate responders or non-responders based on follow up DAS28 after treatment and overall change in DAS28 . We found a highly significant association for a variant that we also show is also a strong expression quantitative trait locus ( eQTL ) for the CD84 gene . Our findings suggest that CD84 genotype and/or expression may prove to be a biomarker for etanercept response in RA patients .
Clinical and GWAS data were compiled for 2 , 706 individuals of European ancestry from 13 collections as part of an international collaboration . Table 1 shows sample sizes , phenotypes and clinical variables for the four collections that were the units of analysis ( additional details are shown in Table S1 ) . Disease activity score based on a 28-joint count ( DAS28 ) were collected at baseline and at one time point after anti-TNF therapy administration ( mean 3 . 7 months , range 3–12 months ) . We defined our primary phenotype as a change in DAS28 ( ΔDAS ) from baseline ( so that greater ΔDAS corresponded with better response to therapy; overall mean and standard deviation of 2 . 1±1 . 3 ) , adjusted for baseline DAS . A secondary phenotype was used based on European League Against Rheumatism ( EULAR ) response criteria . EULAR ‘good response’ was defined as ending DAS<3 . 2 and ΔDAS>1 . 2; ‘non-response’ was defined as ΔDAS <0 . 6 or ΔDAS≤1 . 2 , and ending DAS >5 . 1; and ‘moderate response’ is in between [15] . We limited our secondary analysis to a dichotomous outcome , EULAR good responders ( n = 998 for all patients treated with anti-TNF therapy ) versus EULAR non-responders ( n = 655 ) , excluding the moderate category based on the hypothesis that a more extreme phenotype of response would yield improved discrimination . Clinical variables were examined for association with phenotype , and therefore possible confounding in genetic association tests . In multivariate models ( Table S2 ) , only baseline DAS was strongly associated with the ΔDAS phenotype . As previously shown [11] , age and gender showed univariate associations that were attenuated in the multivariate analysis . Accordingly , we used only baseline DAS as a clinical covariate , as this allowed us to maximize sample size given clinical variable missing data in some cohorts . We performed quality control ( QC ) filtering and data processing of GWAS data for each of eleven genotyping batches . Genotyping array platforms are described in the Methods . HapMap2 imputation allowed us to test for association at >2 M SNPs with imputation quality scores >0 . 5 . Genotype data were merged across several genotype batches to create four collections for genome-wide association testing . We performed linear regression association tests using baseline DAS and three principal components as covariates , and performed inverse-variance weighted meta-analysis to combine results across the four collections . Quantile-quantile plots with genomic control λGC values are shown in Figure S1 . We found no evidence of systematic inflation of association test results , and no evidence of deflation for imputed versus genotyped SNPs . As a final filter , we excluded SNPs that showed strong evidence of heterogeneity across collections ( Cochran's Q P<0 . 001 ) . We first analyzed all samples together ( n = 2 , 706 ) , regardless of drug type . We found no clear evidence of association with treatment response measured by ΔDAS ( Figure 1A ) . Similar results were obtained using the binary phenotype of EULAR responder versus EULAR non-responder status ( Figures S1 and S2 ) . We next separately analyzed patients treated with either etanercept ( n = 733 ) , infliximab ( n = 894 ) or adalimumab ( n = 1 , 071 ) ( Figure 1B–1D ) , under the hypothesis that different genetic loci affect response to the different drugs based on their mechanism of action or other biochemical properties . GWAS results are publicly available for all SNPs tested at the Plenge laboratory and RICOPILI Web sites ( see URLs ) . GWAS results for all SNPs achieving P<10−6 from any analysis are detailed in the Table S3 . For etanercept-treated RA patients , a locus on chromosome 1q23 achieved near-genome-wide significance ( rs6427528 , PMETA = 8×10−8 ) ( Figure 1B , Figure 2A , and Figure 3 ) , but not in the infliximab or adalimumab subsets ( P>0 . 05 ) ( Figure S3 ) . SNPs in linkage disequilibrium ( LD ) showed consistent association results ( rs1503860 , P = 1×10−7 , r2 = 1 with rs6427528 in HapMap; three perfect-LD clusters of SNPs exemplified by rs3737792 , rs10908787 and rs11265432 respectively; P<5×10−6; r2 = 0 . 83 , 0 . 63 and 0 . 59 with rs6427528 , respectively ) . No single collection was responsible for the signal of association , as the effect size was consistent across all collections ( Figure S4 ) . The top SNP rs6427528 was genotyped in the ReAct dataset ( Illumina Omni Express genotyping chip ) , and was well imputed across all other datasets ( imputation quality score INFO ≥0 . 94 , which is an estimate of genotype accuracy; the range of INFO scores is 0–1 , where 1 indicates high confidence ) . All of these SNPs had minor allele frequencies ranging from 7–10% . The SNP explains 2 . 6% variance in response to etanercept treatment . For patients treated with infliximab , we observed a suggestive result on chromosome 10p14 ( rs12570744 , P = 2×10−7 ) . No highly significant or suggestive results were observed for the ΔDAS phenotype in patients treated with adalimumab ( PMETA>10−5 ) . Qualitatively similar results were attained in the analysis of our secondary phenotype , EULAR good responder vs non-responder status ( Figures S1 and S2 ) . For SNPs at the 1q23 locus , the pattern of association with responder/non-responder status ( etanercept-treated patients ) was consistent with the results for ΔDAS ( P = 6×10−3 for rs6427528 and rs1503860 ) . We also identified potential novel associations , with suggestive results for infliximab ( rs4336372 , chromosome 5q35 , P = 8×10−7 ) and adalimumab ( rs940928 , chromosome 2q12 , P = 2×10−6 ) . For each SNP with P<10−6 identified by our GWAS ( n = 6 independent SNPs ) , we searched for biological evidence to support a true positive association . We used genome-wide sequence data from the 1000 Genomes Project to search for putative functional variants in LD with the index SNP ( defined as SNPs predicted to change protein-sequence or mRNA splicing ) . We also used genome-wide expression data to search for an expression quantitative trait locus ( eQTL ) in public databases and in peripheral blood mononuclear cells ( PBMCs ) in 228 non-RA patients and in 132 RA patients . While we did not identify any variants disrupting protein-coding sequences or mRNA splicing , we did find that the 1q23 SNP associated with response to etanercept therapy was a strong eQTL in PBMCs ( Figure 2A and Figure 3 ) . In an analysis of 679 SNPs for cis-regulated expression of five genes in the region of LD ( SLAMF6 , CD84 , SLAMF1 , CD48 , and SLAMF7 ) , we found that rs6427528-CD84 ( and SNPs in LD with it ) was the top eQTL of all results ( n = 228 subjects; Figure 2A ) . This SNP was specifically associated with CD84 expression , and was not an eQTL for other genes in the region ( P>0 . 36 for the other genes ) . We replicated our eQTL finding in 132 RA patients with both GWAS data and genome-wide expression data . PBMC expression data were available from RA patients in the Brigham RA Sequential Study ( BRASS ) and Autoimmune Biomarkers Collaborative Network ( ABCoN ) collections . We observed a significant association between rs6427528 genotype and CD84 expression ( linear regression adjusted for cohort P = 0 . 004 , rank correlation P = 0 . 018 ) . The direction of effect was the same as in the PBMC samples from 228 non-RA patients . A combined analysis of RA patients and the non-RA patient eQTL data ( described above ) yielded rank correlation P = 3×10−10 ( n = 360 total individuals ) . We searched sequence data to determine if rs6427528 , or any of the SNPs in LD with it , were located within conserved , non-coding motifs that might explain the eQTL data . We used HaploReg [16] to examine the chromatin context of rs6427528 and 26 SNPs in LD with it ( at r2>0 . 50 ) . We found that 5 SNPs occur in strong enhancers inferred from chromatin marks ( Figure 2B ) [17] . Two of these 5 SNPs , rs10797077 and rs6427528 ( r2 = 0 . 74 to each other ) , are predicted to disrupt transcription factor binding sites , and rs10797077 occurs at a site that shows conservation across mammalian genomes [18] . Figure 2C shows the DNA sequence position weight matrices of the transcription factor binding sites changed by rs10797077 ( the minor allele creates a stronger binding site for the AIRE transcription factor ) and rs6427528 ( the minor allele creates a binding site for KROX and SREBP ) . Because the genetic data demonstrates that the allele associated with better response is associated with higher CD84 expression , this suggests that CD84 expression itself may serve as a useful biomarker of disease activity or treatment response . We tested both hypotheses using PBMC expression data from the BRASS and ABCoN collections . First , we tested if CD84 expression is associated with cross-sectional DAS , adjusting for age , gender and cohort ( Figure 4 ) . We observed a significant inverse association between CD84 expression and cross-sectional DAS in 210 RA patients ( beta = −0 . 3 , P = 0 . 02 , r2 = 0 . 02 ) . That is , higher CD84 expression was associated with lower DAS , regardless of treatment . Second , we tested CD84 for association with our primary treatment response phenotype , ΔDAS . The sample size for this analysis was smaller than for the cross-sectional analysis , as we required that patients be on anti-TNF therapy and have pre- and post-treatment DAS . We found that CD84 expression levels showed a non-significant trend towards an association with ΔDAS in 31 etanercept-treated patients ( beta = 0 . 2 , r2 = 0 . 002 , P = 0 . 46 ) and in all 78 anti-TNF-treated patients ( beta = 0 . 14 , r2 = 0 . 004 , P = 0 . 4 ) . The effect is in the same direction one would predict based on the genetic association at rs6427528: the allele associated with better response is also associated with higher CD84 expression ( Figure 3 ) , and in 31 RA patients , higher CD84 expression ( regardless of genotype ) is associated with a larger ΔDAS ( i . e . , better response; Figure 4 ) . Since most of the samples available to us as part of our international collaboration were included in our GWAS , few additional samples were available for replication . In addition , the remaining samples available to us were from different ethnic backgrounds . Nonetheless , we sought to replicate the associations of rs6427528 with ΔDAS in these additional samples . We genotyped 139 etanercept-treated patients from a rheumatoid arthritis registry in Portugal ( Reuma . pt ) and 151 etanercept-treated patients from two Japanese collections ( IORRA , n = 88 patients on etanercept and Kyoto University , n = 63 on etanercept ) . Replication sample sizes , clinical data and results for these two SNPs are shown in Table S4 . Based on the observed effect size in the GWAS and observed allele frequency in the replication samples , we had 32% power to replicate this finding in the Portuguese samples and 17% power to replicate this finding in the Asian samples at P<0 . 05 . The same association analysis as for GWAS was carried out: linear regression assuming an additive genetic model and using ΔDAS as phenotype , adjusted for baseline DAS . Replication results are shown in Figure 5 . While the SNPs fail to replicate in these patient collections at P<0 . 05 , the direction of effect is the same in the Portuguese and Kyoto replication samples as in our GWAS . In a combined analysis limited to subjects of European-ancestry ( GWAS data and Portuguese replication samples ) , rs6427528 remained highly suggestive ( P = 2×10−6 ) . Including the Japanese subjects , the overall GWAS+replication combined meta-analysis P-value remained suggestive ( P = 5×10−4 ) .
Here we present the largest GWAS to date on anti-TNF therapy response in 2 , 706 RA patients . We find a significant association at the 1q23/CD84 locus in 733 etanercept treated patients ( P = 8×10−8 ) , but not in RA patients treated with drugs that act as a monoclonal antibody to neutralize TNF ( infliximab or adalimumab ) . The allele associated with a larger ΔDAS ( i . e . , better response ) was associated with higher CD84 expression in PBMCs from non-RA patients ( P = 1×10−11 ) and in RA patients ( P = 0 . 004 ) . We first conducted a GWAS of both categories of anti-TNF drugs ( the soluble receptor drug , etanercept , and two monoclonal antibody drugs , infliximab and adalimumab ) . However , this analysis revealed no strongly associated SNPs . When we subset our GWAS by each of the three individual drugs , several SNPs in the 1q23 locus were highly significant in etanercept-treated patients , and SNPs in three other loci ( 10p15 , 5q35 and 2q12 ) were associated in infliximab or adalimumab subset analyses . Furthermore , the top SNPs for each analysis ( Table S3 ) showed little correlation across the three anti-TNF drugs . This simple observation suggests that genetic control of treatment response may be different for different drugs . This finding is consistent with the clinical observation that RA patients who fail one anti-TNF drug may still respond to a different anti-TNF drug , albeit at lower rates of response [19] . If confirmed in larger samples and more comprehensive analyses , then this could have major implications for how physicians prescribe these drugs . The most significant finding from our GWAS was a set of equivalent SNPs in LD with each other from the 1q23 locus in etanercept-treated RA patients ( Figure 1 and Figure 2A ) . While the top SNP did not reach genome-wide significance in predicting treatment response , it did reach genome-wide significance as an eQTL in PBMCs ( P = 1×10−11; Figure 2A ) . This finding indicates that the SNP ( or another variant in LD with it ) is indeed biologically functional in a human tissue that is important in the immune response . Two SNPs , rs10797077 and rs6427528 , disrupt transcription factor binding sites , and represent excellent candidates for the causative allele to explain the effect on CD84 expression ( Figure 2C ) . Our findings suggest that CD84 genotype and/or expression could be a biomarker for etanercept treatment response among individuals of European ancestry . The genetic and expression data predict that CD84 expression should be positively associated with treatment response ( i . e . , higher expression is associated with better response; Figure 3 ) . While we did not observe a significant association between CD84 levels and ΔDAS , we did observe a trend consistent with this prediction ( Figure 4 ) . Importantly , we note that power was extremely limited with the small sample sizes for which we had CD84 expression as well as drug response data ( n = 31 RA patients treated with etanercept ) . The CD84 gene is a compelling candidate for immune response , belonging to the CD2 subset of the immunoglobulin superfamily . It has been implicated in T-cell activation and maturation [20] . CD84 localizes to the surface of CD4+ and CD8+ T cells , and acts as a costimulatory molecule for IFN-gamma secretion [21] . CD84 is also expressed in B-cells , monocytes and platelets . CD84 has not been previously implicated in genetic studies of RA risk , disease activity , disease severity , or treatment response . A limitation of our study is the small sample size available for replication ( n = 290 etanercept-treated patients ) , and the lack of replication observed for the top CD84 SNP ( rs6427528 ) among patients of Portuguese and Japanese ancestry . The simplest explanation is that our original observation in the GWAS data represents a false positive association . However , the eQTL and gene expression data argue against this possibility . Explanations for a false negative finding in our replication collections include: ( 1 ) lack of power , especially if the effect size observed in the GWAS represents an over-estimate of the true effect size ( the Winner's Curse ) – we estimate that we had 32% and 17% power ( at P = 0 . 05 ) to detect an association in the Portuguese and Japanese sample collections , respectively; ( 2 ) clinical heterogeneity , which is always a possibility in pharmacogenetic studies , especially those conducted in different countries; and ( 3 ) ethnic differences , including different patterns of LD between the underlying causative allele ( which is as yet unknown ) and marker SNPs tested in our study . We did observe subtle differences in local patterns of LD between Asians and Europeans using genetic data from the 1000 Genomes Project ( Figure S5 ) . We note that the rs6427528 minor allele A has a frequency of ∼5–10% in European and East Asian populations , and ∼50% in the African YRI population ( HapMap2 and 1000 Genomes ) ; therefore , it may be of interest to test African American samples in replication . What are the options for increasing sample size in pharmacogenetic studies , thereby providing an opportunity to replicate our CD84 genetic and expression findings ? While it might seem trivial to collect more samples through traditional registries , this is extremely challenging for phenotypes pertaining to treatment efficacy . To underscore this point , we highlight our study design , where we organized samples and clinical data from 16 different collections across 7 different countries in order to obtain the samples for the current study . Going forward , non-traditional strategies to collect biospecimens linked with clinical data ( e . g . , online registries , electronic medical records ) may be required to achieve clinical collections of sufficient size to discover pharmacogenomic predictors of efficacy . In conclusion , we conducted the largest GWAS to date for response to anti-TNF therapy in RA patients . Our genetic and expression data suggest that CD84 genetic variants and/or expression levels could be developed as predictive biomarkers for etanercept treatment response in RA patients of European ancestry .
All patients met 1987 ACR criteria for RA , or were diagnosed by a board-certified rheumatologist . In addition , patients were required to have at least moderate disease activity at baseline ( DAS>3 . 2 ) . All patients gave their informed consent and all institutional review boards approved of this study . A total of 13 collections from across 5 countries were included in GWAS [11] , [12] , [13] , [22]: Autoimmune Biomarkers Collaborative Network ( ABCoN ) from the U . S . ( N = 79 ) ; the Genetics Network Rheumatology Amsterdam ( GENRA , N = 53 ) ; the Dutch Behandelstrategieen voor Rheumatoide Arthritis ( BeSt , N = 85 ) ; the U . K . Biological in Rheumatoid arthritis Genetics and Genomics Study Syndicate ( BRAGGSS , N = 140 ) ; the U . S . Brigham Rheumatoid Arthritis Sequential Study ( BRASS , N = 55 ) ; the Swedish Epidemiological Investigation of Rheumatoid Arthritis ( EIRA , N = 298 ) ; the Immunex Early Rheumatoid Arthritis study ( eRA N = 57 ) ; the Swedish Karolinska Institutet study ( KI , N = 77 ) ; the Netherlands collection from Leiden University Medical Center ( LUMC , N = 43 ) ; and the U . S . Treatment of Early Aggressive RA ( TEAR , N = 109 ) . We refer to these collections as the American College of Rheumatology Research and Education Foundation ( REF ) collection , as funding for GWAS genotyping was provided by the “Within Our Reach” project . We included additional samples from BRAGGSS ( N = 595 ) [12]; the Dutch Rheumatoid Arthritis Monitoring registry ( DREAM ) in the Netherlands , and the ApotheekZorg ( AZ ) database ( which facilitates the Dutch distribution of adalimumab; N = 880 ) [23] , [24] , together referred to as DREAM; and the French Research in Active Rheumatoid Arthritis ( ReAct , N = 272 ) [25] . Additional samples were collected for replication of SNPs in the 1q23 locus . These included the Rheumatic Diseases Portuguese Register ( Reuma . pt , N = 378 ) from the Portuguese Society of Rheumatology ( SPR ) , which captures more than 90% of patients treated with biological therapies and managed in rheumatology departments across Portugal [26] . Additional replication samples ( N = 374 ) of East Asian ancestry were included from the IORRA and Kyoto University Hospital registries , part of the Japanese Genetics and Allied research in Rheumatic diseases Networking consortium ( GARNET ) [27] . Clinical data were collected in each cohort , including disease activity scores at baseline and at least one time point after treatment , gender , age , methotrexate use , as well as autoantibody status ( RF or CCP ) . The composite disease activity scores for 28 joints ( DAS28 ) included laboratory values for erythrocyte sedimentation rate ( ESR ) for most samples and C-reactive protein ( CRP ) for 191 samples in the REF collection ( ABCoN , BRASS and eRA cohorts ) . DAS28 values were available at baseline and at 3–12 months after initiating anti-TNF therapy . Our primary phenotype was defined as ΔDAS = baseline DAS - end DAS , and responder status was also determined according to EULAR criteria for start and end DAS [15] . Clinical variables were assessed for association with phenotype in multivariate linear or logistic regression models for both the ΔDAS and EULAR responder-status phenotypes . Clinical variables that were significant in these analyses were retained as covariates in genetic association tests , except for methotrexate co-therapy . Including a covariate for methotrexate co-therapy reduced sample size substantially due to missing clinical data , so results were compared for our primary analysis and a secondary analysis with the covariates ( and with reduced sample size ) and the results were verified not to be impacted ( not shown ) . A total of eleven genotyping batches were processed separately . ( 1 ) BRASS samples were genotyped using Affymetrix 6 . 0 chip [28]; ( 2 ) WTCCC samples were genotyped on Affymetrix 500K chip [12] . All other cohorts were genotyped using Illumina platform arrays ( see Table 1 ) . Our American College of Rheumatology Research Education Fund ( REF ) collection was made up of smaller cohorts from throughout North America and Europe , including BRASS samples . Also included in REF: ( 3 ) ABCoN [13] and ( 4 ) EIRA [29] were separately genotyped on the Illumina 317K genotyping array; ( 5 ) eRA on the Illumina 550K chip; and ( 6 ) GENRA , BeSt , BRAGGSS ( a subset of N = 53 samples ) , KI and LUMC were genotyped in one batch , and ( 7 ) BRAGGSS ( N = 87 ) and TEAR were genotyped in a second batch , both using Illumina 660k chips , at the Broad Institute ( 8–10 ) . DREAM and AZ samples were genotyped in three batches , one on 550K chip and two on 660K chips ( manuscript in preparation ) , and ( 11 ) ReAct samples were genotyped on Illumina OmniExpress chips . Quality control ( QC ) filtering was done in each genotyping batch , including filtering individuals with >5% missing data , and filtering SNPs with >1% missing data , minor allele frequency ( MAF ) <1% and Chi-squared test of Hardy Weinberg equilibrium PHWE<10−5 . We then used individual-pairwise identity-by-state estimates to remove occasional related and potentially contaminated samples . Data processing and QC were performed in PLINK [30] . Principal Components Analysis ( PCA ) was performed using EIGENSTRAT [31] ( default settings ) on the combined dataset using 20 , 411 SNPs genotyped across all datasets . Ethnicity outliers including all individuals of non-European decent were identified and removed , and the first three eigenvectors were used as covariates in GWAS . Imputation was conducted on each of eleven datasets separately , using the IMPUTE v1 software [32] and haplotype-phased HapMap Phase 2 ( release 22 ) European CEU founders as a reference panel . Imputation of BRASS and EIRA was previously reported [28] , [33] , and we followed the same imputation procedures for the remaining datasets . Imputation yielded posterior genotype probabilities as well as imputation quality scores at SNPs not genotyped with a minor allele frequency ≥1% in HapMap CEU . We removed imputed SNPs with imputation ‘info’ scores <0 . 5 or MAF <1% in any of the datasets . Gene expression levels were quantified using mRNA derived from peripheral blood mononuclear cells ( PBMCs ) using Affymetrix Human Genome U133 Plus 2 . 0 , for 255 multiple sclerosis patients in the Comprehensive Longitudinal Investigation of MS at the Brigham and Women's Hospital [34] , either untreated ( N = 83 ) or treated with interferon-beta ( N = 105 ) or glatiramer acetate ( N = 67 ) . The raw intensity values were subject to quality control based on the recommended pipeline available in the simpleaffy and affyPLM R Bioconductor packages , and were then normalized using GCRMA ( N = 228 ) . The data are available on the Gene Expression Omnibus website ( GSE16214 ) . Expression levels for 17 , 390 probes mapping to 9 , 665 Ensembl transcripts were adjusted for confounding factors including age , gender , drug and batch using principle components and Bayesian factor analysis [35] , and used in eQTL association analyses . Genotype data were collected on the Affymetrix 550K GeneChip 6 . 0 platform as a part of a previously published study [36] . Allelic dosages from imputed data ( HapMap Phase II CEU samples; >2 million SNPs , MACH imputation quality >0 . 1 and MAF> = 0 . 05 ) were used for association analysis . Cis-eQTLs were identified +/−1 Mb of transcription start sites ( TSS ) in the 1q23 locus region . Significance was evaluated by 10 , 000 permutations per gene , and false discovery rates were calculated based on cis-eQTL analyses in the total of 9 , 665 genes [37] . Additional expression profile data were available for subsets of samples that were part of two cohorts in our GWAS . Expression data from patients enrolled in the BRASS registry have been previously published [38] . Expression data were collected on Affymetrix Gene Chip U133 Plus 2 microarrays . BRASS patients had either cross-sectional expression data ( n = 132 , assayed at the time the patient was enrolled in BRASS ) or pre- and post-treatment expression data ( n = 17 samples , 8 treated with etanercept ) . Of these , n = 87 patients had expression and GWAS data . For patients with pre- and post-treatment data , we used the “baseline” pre-treatment expression data for cross-sectional analysis . In ABCoN , 65 RA patients ( n = 23 treated with etanercept ) had both pre- and post-treatment expression data , as well as ΔDAS clinical data [39] , and n = 45 patients had expression and GWAS data . As with BRASS , we use the “baseline” pre-treatment expression data for cross-sectional analysis . For ABCoN expression profile data were collected on Illumina Human WG6v3 microarrays and were quantile normalized according to Illumina recommended protocols . Within both BRASS and ABCoN , expression data were normalized to the mean and standard deviation within each collection . For prospective analyses of expression data and ΔDAS , we combined BRASS and ABCoN to include 31 etanercept-treated patients and 78 anti-TNF-treated patients . In our primary GWAS analysis , we tested each SNP for association with ΔDAS using linear regression adjusted for baseline DAS and the first 3 PCA eigenvectors in each collection . In our secondary GWAS analysis , we modeled SNPs predicting EULAR good response versus EULAR non-response using logistic regression , again adjusting for start-DAS value and the first three eigenvectors . Association analysis was done using SNPTEST [32] assuming an additive genetic model . Genomic control λGC values [40] for genotyped SNPs only and all SNPs were calculated , and no inflation or deflation was observed in the distributions of association test results . We then conducted inverse variance-weighted meta-analysis to combine results across the four datasets , and conducted Cochran's Q tests for heterogeneity using the β coefficients [41] . We further divided samples into 3 subsets according to drug ( etanercept , infliximab or adalimumab ) . GWAS analysis for each group followed the same analysis procedure . Meta-analysis and heterogeneity tests were conducted using SAS . Expression analyses utilized linear regression or Spearman rank correlation , also using SAS . We tested for effects of cohort , age , gender and concurrent methotrexate , and results are shown using significant covariates as indicated . | There are no genetic predictors of response to one of the most widely used classes of drugs in the treatment of rheumatoid arthritis—biological modifiers of the inflammatory cytokine tumor necrosis factor-alpha ( or anti-TNF therapy ) . To identify genetic predictors , we performed the largest genome-wide association study ( GWAS ) to date as part of an international collaboration . In our study , which included 2 , 706 RA patients treated with one of three anti-TNF drugs , the most significant finding was restricted to RA patients treated with etanercept ( P = 8×10−8 ) , a drug that acts as a soluble receptor to bind circulating cytokine and prevents TNF from binding to its cell surface receptor . The associated variant influences expression of a nearby immune-related gene , CD84 , whose expression is correlated with disease activity in RA patients . Together , our data support a model in which genomic factors related to CD84 expression serve as a predictor of disease activity and response to etanercept therapy among RA patients of European ancestry , but not anti-TNF therapies that act through different biological mechanisms or potentially in RA patients of other genetic ancestries . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"gene",
"expression",
"genetics",
"biology",
"genetics",
"and",
"genomics"
] | 2013 | Genome-Wide Association Study and Gene Expression Analysis Identifies CD84 as a Predictor of Response to Etanercept Therapy in Rheumatoid Arthritis |
Meiotic mapping of quantitative trait loci regulating expression ( eQTLs ) has allowed the construction of gene networks . However , the limited mapping resolution of these studies has meant that genotype data are largely ignored , leading to undirected networks that fail to capture regulatory hierarchies . Here we use high resolution mapping of copy number eQTLs ( ceQTLs ) in a mouse-hamster radiation hybrid ( RH ) panel to construct directed genetic networks in the mammalian cell . The RH network covering 20 , 145 mouse genes had significant overlap with , and similar topological structures to , existing biological networks . Upregulated edges in the RH network had significantly more overlap than downregulated . This suggests repressive relationships between genes are missed by existing approaches , perhaps because the corresponding proteins are not present in the cell at the same time and therefore unlikely to interact . Gene essentiality was positively correlated with connectivity and betweenness centrality in the RH network , strengthening the centrality-lethality principle in mammals . Consistent with their regulatory role , transcription factors had significantly more outgoing edges ( regulating ) than incoming ( regulated ) in the RH network , a feature hidden by conventional undirected networks . Directed RH genetic networks thus showed concordance with pre-existing networks while also yielding information inaccessible to current undirected approaches .
Interrogating genome-scale datasets is a necessary step to a systems biology of the mammalian cell [1] , [2] . Networks have been constructed using various approaches . In the transcriptome , coexpression networks have been constructed by linking genes whose correlations exceed a selected p-value based on transcript profiling data across different samples [3] . In the proteome , genes can be linked if their corresponding proteins bind each other based on yeast two-hybrid ( Y2H ) or co-affinity immunoprecipitation assays [4] , [5] . Protein-protein interactions can also be ascertained from literature-curated ( LC ) databases [6] , [7] . The Human Protein Reference Database ( HPRD ) consists of ∼8 , 800 proteins and ∼25 , 000 interactions and was constructed using Y2H , co-affinity purification and LC data [6] . Genes can also be linked by virtue of membership of a common pathway [8] , [9] , an example being the Kyoto Encyclopedia of Genes and Genomes ( KEGG ) pathway [10]–[12] . Networks constructed using these various approaches are correlated , with some exceptions . While a single dataset often has a large number of false positives and false negatives and reflects only one facet of gene function , accessing multiple independent datasets increases the reliability of gene functional annotation . Integrating diverse gene networks has been shown predictive of loss-of-function phenotypes in yeast [8] , [13] and Caenorhabditis elegans [9] . Recently transcriptional networks have been constructed using expression data from genetically polymorphic individuals [14]–[16] . This approach allows the identification of quantitative trait loci ( QTLs ) regulating expression , or eQTLs . Mapping of eQTLs relies on expression perturbations due to naturally occurring polymorphisms . These sequence variants may be lacking in critical pathways because of selective pressure , rendering inaccessible important regions of the genetic network . A disadvantage of most currently available networks is that it is difficult to infer functional relationships between interacting genes . Consequently , the edges between genes are undirected and have no regulatory hierarchy . This is also true of eQTL networks where , because of limited mapping power , genotype information has been generally ignored and coexpression networks have been constructed instead [17] . Causality between expression and clinical traits has been inferred from eQTL data using conditional correlation measures [18] and structural model analysis [19] , [20] . However , this approach has been restricted to a small subset of markers and traits and cannot be easily extended to constructing gene networks . Radiation hybrid panels have been used to construct high resolution maps of mammalian genomes [21]–[23] . Fragmenting a mammalian genome using radiation yields many more breakpoints than meiotic mapping and hence greatly enhanced resolution . The T31 mouse-hamster hybrid panel was constructed by lethally irradiating mouse cells harboring the thymidine kinase gene ( Tk1+ ) [22] . These cells were then fused to Tk1− hamster A23 cells . Selection for the Tk1+ gene using HAT medium resulted in a panel of 100 hybrid cell lines , each of which contained a random sampling of the mouse genome . Mouse autosomal genes retained in a hybrid clone have two hamster copies plus one mouse copy , compared to two copies otherwise . We recently used the T31 RH panel for high-resolution mapping of QTLs for gene expression [24] . The QTLs regulate expression because of copy number changes and they are therefore called copy number expression QTLs or ceQTLs . We re-genotyped the T31 panel at 232 , 626 markers using array comparative genomic hybridization ( aCGH ) . The average retention frequency of mouse markers in the panel was 23 . 9% and the average length of the mouse fragments was 7 . 17 Mb . We also analyzed the panel using expression microarrays interrogating 20 , 145 genes . Using regression , we found 29 , 769 trans ceQTLs regulating 9 , 538 genes at a false discovery rate ( FDR ) = 0 . 4 in the T31 panel . At the same FDR threshold , we also found 18 , 810 cis ceQTLs . Consistent with the average fragment length , a ceQTL was identified as trans if >10 Mb from a regulated gene and cis otherwise . The interval for the ceQTLs was <150 kb , thus localizing them to an average of only 2–3 genes . In this paper we evaluate gene networks constructed from ceQTL mapping . In contrast to undirected networks from meiotically mapped eQTLs and protein binding approaches , the high resolution mapping and dense genotyping of ceQTLs in the RH panel allowed the use of genotype information to construct directed networks . This directionality permits insights that cannot be obtained from undirected networks .
We previously analyzed a mouse-hamster radiation hybrid panel , T31 [24] . The donor cells were male primary embryonic fibroblasts from the inbred mouse strain 129 and the recipient cells were from the A23 male Chinese hamster lung fibroblast-derived cell line [22] . A total of 99 cell lines from the original panel were available . RH clones with retained autosomal mouse genes in the panel have two hamster copies plus usually one extra mouse copy , compared to two hamster copies otherwise . The variation in gene dosage drives changes in mRNA expression . Transcript abundance and marker dosage were measured by mouse expression arrays and comparative genomic hybridization arrays ( aCGH ) , respectively . A total of 20 , 145 transcript levels were assayed by the expression arrays and 232 , 626 markers by the aCGH . We mapped ceQTLs by regressing the expression array data on the aCGH data . Mouse and hamster genes were detected with comparable efficiency and behaved equivalently in terms of regulation [24] . To construct the RH network , the copy number of each gene was estimated by linear interpolation using the two neighboring aCGH markers . The linear interpolation based estimation is reasonable , considering the high density of aCGH markers . Measured transcripts were denoted by , where and are gene and RH clone index , respectively . The estimated gene copy number was denoted by for gene in RH clone . For each ordered pair of genes and , a Pearson correlation coefficient between and was calculated from the 99 observations . In a linear model , where and are regression parameters , the correlation coefficient can be viewed as a standardized slope and measures the goodness of fit for the linear model . A significantly large positive value implies induction and a significantly large negative value implies repression . Previously , we used an F-statistic , which is monotonic in the absolute value of the correlation coefficient , to test for significant association in a context of the linear model [24] . Here we preserved the sign and used the correlation coefficient as a test statistic . We found that yielded more significant overlaps with other biological datasets than ( below ) . The number of directed edges and number of nodes with ≥1 edge for right-tailed , left-tailed and both-tailed thresholding are shown in Table S1 and Figure S1 ( see Methods ) . We constructed an adjacency matrix by assigning to its entry , which gives information on whether gene regulates gene , either directly or indirectly . Since has real number entries and is not symmetric , the network represented by is weighted and directed . We used the correlation coefficients for thresholding and calculated the statistical significance of similarities to existing biological datasets . This is in contrast to transforming the correlation coefficients into FDR ( false discovery rate ) corrected p-values and then performing statistical thresholding [24] . Our strategy in this study is similar , in spirit , to the integration approach taken in [8] , [9] where the reliability of each dataset is measured by comparing with a benchmark dataset . Since nearly all genes show a copy number increase in a portion of the RH panel , the bulk of genes ( 94% ) also showed a cis ceQTL [24] . To remove these cis ceQTLs as an artifactual source of edges in the RH network , we omitted all markers within 10 Mb of the gene being considered . Thus , only trans ceQTLs were employed in the analysis . We examined the similarity of our network to existing datasets including protein-protein interactions from HPRD ( Human Protein Reference Database ) [6] , the KEGG ( Kyoto Encyclopedia of Genes and Genomes ) pathway database [10]–[12] , Gene Ontology ( GO ) annotations [25] and a coexpression network obtained from the SymAtlas microarray database of normal mouse tissues [26] ( see Methods ) . We used two different approaches to compare the directed RH and undirected networks . In the first approach , we discarded the edge directions of the RH network and calculated an overlap of undirected edges between the RH and existing networks . It is not uncommon to disregard directions in a network for modeling and analysis purposes [27]–[33] and projecting a directed network onto a space of undirected networks by forgoing information on edge directions seems reasonable . In the second approach , we assumed a hidden directed random network for each undirected existing network and estimated the resulting overlap of directed edges . We examined whether upregulation in the RH data , represented by positive correlation coefficients , , showed a different significance of overlap with existing datasets than downregulation , represented by . We defined an unweighted adjacency matrix by left-tailed thresholding of the RH data , where if for a given correlation coefficient threshold , and otherwise . This network emphasized downregulation in the RH data . We also defined by both-tailed thresholding , where if , and otherwise . This network gave equal weight to up- and downregulation in the RH data and is equivalent to previous datasets produced from F-tests [24] . The unweighted adjacency matrix for right-tailed thresholding is defined as if , and , emphasizing upregulation in the RH data . Unweighted RH networks obtained from left-tailed thresholding , which emphasized downregulation , did not show any significant overlap ( FDR-corrected ) with existing datasets ( Figure S3 , Dataset S1 ) , except the GO cellular component annotation . Even this significance was modest . Unweighted networks obtained by both-tailed thresholding , which equally weighted up- and downregulation , also did not show any significant overlap ( FDR-corrected ) with existing datasets , except the GO biological process annotation ( Figure S4 , Dataset S1 ) . Figure 2E compares the maximum significance over correlation coefficient thresholds for the different thresholding approaches . Overall , the results suggest upregulation in the RH network yields more significant overlap with existing datasets than downregulation . This may reflect the fact that if a gene represses another gene in trans the two protein products are unlikely to co-exist in the cell and hence unlikely to interact . A corollary is that protein binding methods such as yeast two-hybrid and co-affinity immunoprecipitation may miss negative regulatory interactions . Our finding is reminiscent of the observation that interacting protein pairs have significantly higher transcript abundance correlations than chance [43] , [44] . The overlap analysis based on edge-comparison may fail to capture some indirect interactions or other topologies . We therefore compared the topological properties of the RH and HPRD networks . The degrees ( number of edges for each node , or connectivity ) of the weighted ( unthresholded ) RH and HPRD networks were significantly correlated ( Spearman's correlation coefficient = 0 . 055 , ) . However , the similarity to the HPRD network disappeared when we used absolute values of the correlation coefficients of the RH network in the adjacency matrix , ( Spearman's correlation coefficient = −0 . 0081 , ) . These observations imply that the degree distribution for upregulated but not downregulated edges in the RH network is significantly correlated with the HPRD network . This is consistent with the notion that repressive relationships are not well represented in HPRD . Next , we compared the betweenness centralities of the RH and HPRD networks . The betweenness centrality measures the total number of nonredundant shortest paths going through each node , representing the severity of bottlenecks in the network [45] , [46] . The betweenness centralities of the RH and HPRD networks were significantly correlated ( FDR = 0 . 05 ) when the right-tailed correlation coefficient thresholds for RH network were between −0 . 1 and 0 . 1 ( Figure 2F ) . We calculated the diameters ( average minimum distance between pairs of nodes ) of the RH and HPRD networks . The diameter of a giant connected component , consisting of 5 , 433 nodes with 20 , 859 undirected edges excepting self-loops , of the HPRD network was 4 . 13 . For the RH network , we considered those 5 , 433 genes that were in the HPRD network and used a right-tailed threshold of 0 . 37544 , yielding 20 , 859 undirected edges , to make its size ( node and edge numbers ) comparable to the HPRD network . The diameter of the RH network was 4 . 11 , close to that ( 4 . 13 ) of the HPRD network . We also compared the clustering coefficients of the RH and HPRD networks , a measure of local cliqueness [47] , but found no significant positive correlation . In summary , the RH network showed similarities with the HPRD network in terms of connectivity , betweenness centrality and diameter , but not cliqueness . Previous studies in other networks showed that essentiality is positively correlated with connectivity and betweenness centrality [9] , [46] , [48]–[56] . However , some authors have questioned the association between essentiality and connectivity , attributing it to dataset bias [6] , [57] . We tested whether essentiality is associated with connectivity and betweenness centrality in the RH network . Essential genes had significantly more edges than non-essential genes for a range of right-tailed correlation coefficient thresholds from −0 . 12 to 0 . 16 ( FDR = 0 . 01 ) using a one-sided Wilcoxon rank-sum test [34] ( Figure 4A ) . This range is similar to that for significant overlaps with existing datasets . Also , the fraction of essential genes was positively correlated with the degree of the weighted RH network ( Pearson's correlation coefficient = 0 . 70 , ) ( Figure 4B ) . Similarly , essential genes had significantly larger betweenness centralities for a range of right-tailed correlation coefficient thresholds from −0 . 14 to 0 . 16 ( FDR = 0 . 01 ) using a one-sided Wilcoxon rank-sum test ( Figure 4C ) . Figure 4D shows that the fraction of essential genes was positively correlated with betweenness centrality for the RH network constructed from a typically optimal right-tailed correlation coefficient threshold for overlap of 0 . 1 ( Pearson's correlation coefficient = 0 . 72 , ) . It is natural to suppose that transcription factors would have more outgoing than incoming edges since transcription factors regulate other genes . This proposition cannot be tested in conventional undirected networks , but can be tested in the directed RH network . Using a one-sided paired signed rank test [34] we found that transcription factors had significantly more outgoing edges than by chance ( FDR = 0 . 01 ) for a range of correlation coefficient thresholds from 0 . 23 to 0 . 46 ( Figure 5A ) . We also used a one-sided Fisher's exact and chi-square test to evaluate the association between transcription factors and genes having ≥1 outgoing edge in the RH network . The significance of the association was modest but significant ( FDR = 0 . 05 ) ( Figure 5B ) . In contrast , the association between transcription factors and genes having ≥1 incoming edge was not significant ( FDR = 0 . 05 ) ( Figure 5B ) . Together , these results imply that transcription factors are more likely to regulate other genes than be the target of regulation and suggest transcription factors have a privileged role in genetic networks .
We used high resolution mapping of ceQTLs in an RH panel to create a directed genetic network . There was significant overlap with existing networks such as HPRD , KEGG , GO annotation and a SymAtlas coexpression network . The RH network also showed similar topological properties to the HPRD network in connectivity , betweenness centrality and diameter . The RH network showed maximum significance of overlap with existing networks at relatively low positive correlation coefficient thresholds between 0 and 0 . 2 . The low thresholds were not simply by chance , since randomly permuted RH networks did not show any significant overlap with existing networks . Also , the low values did not seem to be caused by noise in the array measurements or by lack of sufficient numbers of RH clones , since the correlation coefficient thresholds giving maximum overlap significance remained nearly constant for varying clone number , although the sensitivity of overlap increased with the number of clones . This may reflect the orthogonal nature of the RH network compared to existing networks , suggesting the RH approach will yield complementary information on mammalian genetic networks . Novel and replicated edges in the RH network may thus be balanced in the low correlation coefficient threshold range . The overlap between the RH network and existing interaction networks was greater for edges possessing upregulation than downregulation . This observation may be because the corresponding proteins are unlikely to interact if one gene represses another , since the proteins will not be present in the cell at the same time . It also implies that protein-protein interaction networks may fail to uncover valid edges between genes if they have a repressive relationship . Previous studies found significant associations of essentiality with connectivity and/or betweenness centrality in protein-protein interaction networks [39] , [46] , [48]–[52] , coexpression networks [53] , [56] , Bayesian integrated gene networks [9] and transcriptional regulatory networks [46] , [50] , [54] . Most investigations focused on yeast , worm and fly and there have been only a few studies of mammalian gene networks [6] , [9] . Some authors have questioned the association of essentiality and connectivity [6] , [57] . Coulomb et al . found that essentiality was poorly related to connectivity when biases in protein interaction databases were taken into account [57] . Yu et al . also found related problems due to bias in a yeast two hybrid dataset [39] . In contrast , the RH network is free of biases that may exist in protein interaction datasets . The significant positive correlation between essentiality , connectivity and betweenness centrality in the RH network adds to the evidence of the centrality-lethality rule in the mammalian setting . We also showed that transcription factors were likely to have more outgoing rather than incoming edges . While this finding is not unexpected and helps validate the RH network , a recent study using naturally occurring polymorphisms in yeast suggested that transcription factors are no more likely to reside close to eQTLs than chance [58] . The discrepancy between the RH and yeast studies may be because an increase in copy number in the RH cells is a more reliable way to perturb gene networks than naturally occurring alleles . In contrast , polymorphisms may be under selective pressure to minimize disruptions in potentially critical nodes in gene networks , such as transcription factors . We thresholded the adjacency matrix at different correlation coefficients to compare unweighted RH networks with existing unweighted datasets . However , we chose to leave the RH network weighted rather than finalizing an unweighted form at an optimal threshold . Such an operation is irreversible and would lose information on linkage strength and sign . In other studies , the sensitivity of a coexpression network was limited by thresholding [56] and weighted coexpression networks were more robust than unweighted networks [53] . Indeed , weighted networks are widely used in various applications . In probabilistic integrated gene networks , linkages between genes are represented by weighted sums of log likelihood score ( LLS ) values [8] , [9] . Weighting was also used for a Bayesian gene network [13] and a scientific collaboration network [59] . In addition , weighted coexpression networks have been extensively studied [53] , [60] and it is straightforward to incorporate a weighted network into a probabilistic integrated network by a Bayesian LLS approach [8] , [9] . We constructed a directed gene network from radiation hybrids and found it concordant with existing networks . We also showed that RH networks have the potential to provide new insights reflecting orthogonal aspects of gene regulation . The RH networks will be refined as more panels , including those available for other species , are analyzed resulting in improved power and sensitivity .
Details on the analysis of the T31 RH panel cells and the preprocessing of aCGH and expression array data can be found in [24] . The microarray and aCGH data have been deposited in NCBI Gene Expression Omnibus ( GEO ) database under accession number GSE9052 . The directed RH network was constructed as described in Results . The copy number for each gene was estimated from the aCGH data by linear interpolation as follows . Let denote the array measurement for aCGH marker in RH clone . For gene , suppose marker is nearest to the gene from the left on the same chromosome and marker is nearest from the right . The copy number for gene in clone was estimated by where , and denote the genome coordinates in bp for gene and markers and , respectively . If gene did not have any marker to the left or right on the chromosome , the array measurement for the nearest marker was taken instead . A protein-protein interaction network was constructed from HPRD ( Human Protein Reference Database ) [6] by generating an adjacency matrix , where if the proteins corresponding to annotated mouse genes and interact with each other and otherwise . Note that is symmetric and the HPRD network is undirected . The HPRD network had 6 , 015 nodes and 20 , 957 undirected edges , excepting self-loops . A network was constructed from the KEGG ( Kyoto Encyclopedia of Genes and Genomes ) pathway database [10]–[12] by generating an adjacency matrix such that if genes and participated in the same pathway and otherwise . The KEGG pathway network had 1 , 629 nodes and 139 , 664 undirected edges except self-loops . A network was constructed from the GO ( Gene Ontology ) database [25] by generating an adjacency matrix where if genes and belong to a common GO term and otherwise . Only GO terms with ≤200 genes were considered . Similarly , , and were constructed considering only the GO molecular function terms , GO biological process terms and GO cellular component terms , respectively . The undirected GO , GO-molecular function , GO-biological process and GO-cellular component networks had 10 , 442 nodes with 786 , 928 edges , 7 , 745 nodes with 359 , 006 edges , 7 , 653 nodes with 404 , 641 edges and 3 , 509 nodes with 140 , 904 edges , respectively , excepting self-loops . All edges were undirected . We constructed an mRNA coexpression network from the publicly available SymAtlas microarray database [26] . This database contains transcript profiling data from 61 normal mouse tissues . The Pearson's correlation coefficients of mRNA expression across the mouse tissues were calculated and an adjacency matrix was generated by right-tailed thresholding the correlation coefficients with . The SymAtlas coexpression network had 15 , 190 nodes and 18 , 754 , 380 undirected edges . The significance of overlap between the RH network obtained from thresholding and , for example , the HPRD network was tested as follows . First , for a given threshold , the adjacency matrix of an unweighted RH network was constructed where for right-tailed thresholding , for left-tailed thresholding and for both-tailed thresholding ( see Results ) . Second , for a comparison with the unweighted HPRD network , the adjacency matrix was forced to be symmetric by constructing a symmetric matrix for an undirected RH network such that if or , and otherwise . Third , a two by two contingency table was built showing the relationship between ( 1 or 0 ) and ( 1 or 0 ) , where only pairs of genes in common to both networks are taken . In addition , for all networks , only gene pairs separated by at least 10 Mb on a chromosome or on different chromosomes were selected . This requirement was imposed to remove possible biases due to copy number effects of a gene's own dosage in the RH network and to ensure gene pairs were in trans . Fourth , an overlap was defined as the number of gene pairs such that both and . Then a one-sided Fisher's exact test was performed to evaluate whether the overlap was significant and calculate a p-value . If the expected values in all table cells exceeded 50 , a one-sided chi-square test was used to reduce computational cost . We similarly calculated the significance of overlaps with the KEGG pathway network , the SymAtlas coexpression network and the GO annotations . For each existing undirected dataset , for example , the HPRD network , we assume there is a hidden directed random network with adjacency matrix , whose elements are independent Bernoulli random variables with success probability . We suppose only the undirected version is observed , where ( recall only off-diagonal elements are considered , that is , ) . Then for are independent Bernoulli random variables with success probability . Therefore , using an empirical success probability , the ratio of 1's to the total in , the success probability of the hidden directed random network can be estimated as . The overlap between the unweighted ( thresholded ) directed RH network , represented by , and the hidden directed HPRD network is given by . However , the overlap is not directly observable and instead we calculate the conditional expectation given . Since , it can be seen that Ignoring the constant scaling factor without loss of generality , we define an overlap as ( recall that is symmetric whereas is not ) . To test whether an observed overlap is greater than chance , we calculate a p-value as the probability of the overlap being greater than or equal to the observed value assuming the HPRD network is a random network as described above , where are independent Bernoulli random variables with success probability and and are independent binomial random variables , and , with being the number of unordered pairs such that for or 2 . To reduce the computation cost , is approximated using the normal distribution when , , and . The node degree of the undirected , weighted adjacency matrix where was calculated by . Similarly , the degree of the HPRD network was calculated by . Then we calculated the Spearman's correlation coefficients between and . The betweenness centralities and clustering coefficients of the RH adjacency matrix and the HPRD adjacency matrix were calculated using MatlabBGL ( http://www . stanford . edu/~dgleich ) . When we calculated the betweenness centrality of the RH network , we used a subgraph by taking nodes that were in the HPRD network to reduce computational cost . Then the Spearman's correlation coefficients between the betweenness centralities and also between clustering coefficients for RH and HPRD were calculated . We obtained a list of 1 , 409 essential genes and 1 , 979 nonessential genes from the Mouse Genome Database [6] , [61] . Those 3 , 388 genes were sorted by degree and binned into successive bins of 200 genes and the correlation between mean degree and fraction of essential genes calculated [9] . The betweenness centrality for the RH network was calculated from , taking a subgraph consisting of a total of 3 , 388 genes of interest to reduce computational cost and . Similarly , the 3 , 388 genes were sorted by betweenness centrality and the significance of correlation between the mean betweenness centrality and the fraction of essential genes tested . We obtained a list of 1 , 053 transcription factors by finding genes whose GO description includes a word “transcription . ” The number of outgoing edges was calculated by for gene and the number of incoming edges by for gene . We used a one-sided paired signed rank test [34] to assess whether transcription factors have larger than . The network data are available at http://labs . pharmacology . ucla . edu/smithlab/RHnetwork . html | An important problem in systems biology is to map gene networks , which help identify gene functions and discover critical disease pathways . Current methods for constructing gene networks have identified a number of biologically significant functional modules . However , these networks do not reveal directionality , that is , which gene regulates which , an important aspect of gene regulation . Radiation hybrid panels are a venerable method for high resolution genetic mapping . Recently we have used radiation hybrids to map loci based on their effects on gene expression . Because these regulatory loci are finely mapped , we can identify which gene turns on another gene , that is , directionality . In this paper , we constructed directed networks from radiation hybrid expression data . We found the radiation hybrid networks concordant with available datasets but also demonstrate that they can reveal information inaccessible to existing approaches . Importantly , directionality can help dissect cause and effect in genetic networks , aiding in understanding and ultimately rational intervention . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genetics",
"and",
"genomics/genomics",
"genetics",
"and",
"genomics/animal",
"genetics",
"computational",
"biology/systems",
"biology",
"computational",
"biology/transcriptional",
"regulation",
"genetics",
"and",
"genomics/gene",
"expression",
"molecular",
"biology/bioinformatics",
"computational",
"biology/genomics",
"genetics",
"and",
"genomics/bioinformatics"
] | 2009 | Directed Mammalian Gene Regulatory Networks Using Expression and Comparative Genomic Hybridization Microarray Data from Radiation Hybrids |
Direct-acting antiviral agents ( DAAs ) for hepatitis C treatment tend to fare better in individuals who are also likely to respond well to interferon-alpha ( IFN ) , a surprising correlation given that DAAs target specific viral proteins whereas IFN triggers a generic antiviral immune response . Here , we posit a causal relationship between IFN-responsiveness and DAA treatment outcome . IFN-responsiveness restricts viral replication , which would prevent the growth of viral variants resistant to DAAs and improve treatment outcome . To test this hypothesis , we developed a multiscale mathematical model integrating IFN-responsiveness at the cellular level , viral kinetics and evolution leading to drug resistance at the individual level , and treatment outcome at the population level . Model predictions quantitatively captured data from over 50 clinical trials demonstrating poorer response to DAAs in previous non-responders to IFN than treatment-naïve individuals , presenting strong evidence supporting the hypothesis . Model predictions additionally described several unexplained clinical observations , viz . , the percentages of infected individuals who 1 ) spontaneously clear HCV , 2 ) get chronically infected but respond to IFN-based therapy , and 3 ) fail IFN-based therapy but respond to DAA-based therapy , resulting in a comprehensive understanding of HCV infection and treatment . An implication of the causal relationship is that failure of DAA-based treatments may be averted by adding IFN , a strategy of potential use in settings with limited access to DAAs . A second , wider implication is that individuals with greater IFN-responsiveness would require shorter DAA-based treatment durations , presenting a basis and a promising population for response-guided therapy .
Direct-acting antiviral agents ( DAAs ) are revolutionizing the treatment of chronic hepatitis C virus ( HCV ) infection . Sustained virological response ( SVR ) rates of over 90% have been achieved in recent clinical trials with all-oral DAA treatments lasting as short as 12 weeks , in striking contrast to the combination of pegylated interferon and ribavirin ( PR ) , which elicited SVR rates of only ~50% with 24–48 weeks of treatment [1 , 2] . Indeed , DAAs are rapidly replacing PR as the treatment of choice for chronic HCV infection [2] . An intriguing feature of DAAs is the differential response they elicit in individuals who respond differently to PR: They seem to work better in individuals who also tend to be more responsive to PR . For instance , with the combination of the DAAs ledipasvir and sofosbuvir , SVR rates dropped from nearly 100% in treatment-naive individuals to ~87% in PR-experienced individuals infected with HCV genotype 1b [1] . This differential response appears more significant with the older generation of DAAs than the newer ones , but is evident across clinical trials and across DAAs ( Table 1 ) . Treatment guidelines for those who previously failed PR treatment are different from treatment naïve patients [3] . Interferon ( IFN ) acts primarily by stimulating the innate immune response to HCV [4] . Ribavirin is thought to potentiate the activity of IFN [5 , 6] . DAAs , on the other hand , target specific HCV proteins , independently of host immune responses [7] . Why responsiveness to IFN should improve outcomes of DAA-based treatments is thus puzzling . Here , we hypothesized that the responsiveness of individuals to IFN and DAAs are causally linked . DAAs are susceptible to viral mutation-driven development of drug resistance [8 , 9] . Resistance-associated amino acid variants ( RAVs ) have been identified that possess high level resistance ( >1000-fold increase in EC50 ) to DAAs [9] . Given the rapid turnover of HCV in vivo [10] and its high mutation rate [11] , RAVs are likely to pre-exist in chronically infected individuals [12] and/or arise during treatment [13] , potentially lowering the efficacy of DAAs . Indeed , in retrospective analyses , RAVs were detected more frequently in individuals who failed DAA treatment than in those who achieved SVR [1] . With the combination of ledipasvir and sofosbuvir , for example , 16% of all the patients treated had detectable RAVs at baseline , whereas of those who suffered virological failure , 43% had detectable RAVs at baseline [14] . Although systematic resistance testing is not recommended , current treatment guidelines suggest resistance testing , where such testing is readily accessible and reliable , in the NS5A region to decide appropriate treatment regimens [3] . IFN , a cytokine produced in response to viral infections , triggers the expression of several hundred IFN-stimulated genes ( ISGs ) in infected cells , creating an antiviral state that restricts viral replication [4 , 15] . Higher responsiveness to IFN may thus restrict viral replication to a greater extent , exerting better control on RAVs and leading to improved outcomes of DAA-based treatments . This causal relationship may underlie the positive correlation between responsiveness to PR and DAAs observed in clinical trials . We tested this hypothesis using mathematical modelling and analysis of clinical data . Mathematical models of HCV kinetics have played a crucial role in our understanding of HCV infection and guided treatments [16] . A model to test the hypothesis above had to address the following questions . 1 ) What is the origin of the differential responsiveness of HCV-infected individuals to IFN ? 2 ) How can the IFN-responsiveness of an individual be quantified ? 3 ) Given the IFN-responsiveness of an individual , how does the individual respond to DAAs , assuming the hypothesized causal link above ? 4 ) How is IFN-responsiveness distributed across individuals in a population ? 5 ) Does this latter distribution , coupled with the predicted responses of individuals to treatments , explain the differences in SVR rates between treatment-naive and treatment-experienced populations observed across DAAs and across clinical trials ? Existing models [5 , 10 , 12 , 17–27] have addressed some but not all of these questions . For instance , IFN-responsiveness has been shown recently to be an emergent property of the IFN signaling network in HCV infected cells [17] . Due to the competing interactions between ISGs and HCV [28–33] , the network exhibits bistability , with one steady state responsive and the other refractory to IFN . The proportion of cells in an individual that preferentially assume the responsive state determines the IFN-responsiveness of the individual [17] . Although variations in ISG protein copy numbers and other factors across cells [25 , 34] and across individuals [35 , 36] and effects such as those attributed to the polymorphisms in the IL28B gene locus [37] that collectively result in different levels of IFN-responsiveness in different individuals have been identified , how IFN-responsiveness is distributed across individuals remains unknown . Inspired by the success of models in describing HIV drug resistance [38] , similar models of HCV kinetics incorporating mutations and their fitness effects have been developed to estimate the pre-existing frequencies of RAVs in chronically HCV infected individuals and to predict their growth during treatment with DAAs [12 , 16 , 18–22] . The latter models , however , do not treat IFN-responsiveness as a factor influencing the pre-existence and growth of RAVs and hence treatment outcomes . Finally , no models , barring those invoking empirical correlations [20 , 39 , 40] , have described SVR rates elicited by different DAA-based treatment regimens . Constructing a mathematical model to test the proposed causal relationship thus faced two broad challenges . First , phenomena spanning multiple length and time scales–from the cellular to the population level–had to be integrated into a single mathematical framework . Second , several missing pieces in the puzzle , not considered by existing models , had to be identified . We developed a model that overcame both these challenges . Model predictions captured the vast body of clinical data of the differential response of patients to DAA-based treatments quantitatively , making a strong case for the proposed causal relationship . The model additionally explained several longstanding but poorly understood clinical observations , presenting a far more comprehensive understanding of HCV infection and treatment response than earlier . Finally , using the model , we suggest new strategies , exploiting the causal relationship , to improve DAA-based treatments .
To establish the correlation between the responsiveness of chronically HCV infected individuals to PR and DAAs , we collated data from all ( over 50 ) clinical trials that reported SVR rates achieved with DAA-based treatments in treatment-naïve individuals , SVRnaive , and in previous null responders to PR , SVRnull ( Methods ) . The data are grouped according to treatment regimen and summarized in Table 1 . Individual datasets are listed in S1 Table . We found that SVRnaive>SVRnull with P≈10−59 overall ( using the χ2-test ) . The difference was starker when the analysis was restricted to treatments that included PR ( P≈10−65 ) , but , importantly , was highly significant when interferon-free regimens alone were considered ( P = 0 . 007 ) . The difference remained when only individuals with ( S2 Table; P≈10−5 ) or without ( S3 Table; P≈10−29 ) liver cirrhosis were considered or when the analysis was restricted to studies that did not factor liver cirrhosis ( S4 Table; P≈10−30 ) . The difference was clearer for treatments that elicited <100% SVR than for more recent , stronger treatments that elicited ~100% SVR regardless of treatment experience . Nonetheless , the clinical evidence of a positive correlation between responsiveness to PR and DAAs was overwhelming and suggested a causal relationship between the two . We proposed a mechanistic hypothesis underlying this relationship , where greater IFN-responsiveness exerted better control on RAVs and improved DAA treatment outcomes ( see above ) , and constructed a mathematical model to test it . We present an overview of the model here ( Fig 1 ) . A detailed description of the various components of our model and how we integrated them into a single framework is in Methods . To describe the response of an individual to PR , we employed the formalism we developed previously , where cells were divided into distinct IFN-responsive and IFN-refractory phenotypes based on the properties of the IFN-signaling network in HCV-infected cells [17] . At the cellular level , interferon triggers the expression of several hundred interferon-stimulated genes ( ISGs ) that collectively create an antiviral state in cells [15] . HCV suppresses the interferon response via multiple mechanisms [4 , 41] , the prominent one involving a block in ISG translation it induces via dimerization and autophosphorylation of protein kinase R [30 , 42 , 43] . We constructed a comprehensive model of the IFN signaling network in the presence of HCV , accounting for the competing effects above which amounted to a double negative feedback , and found that the network exhibited bistability [17] . In one steady state , HCV overcame the IFN response and established lasting infection . In the other , IFN cleared HCV . Intrinsic variations of the factors defining the IFN signaling network , which defined the strength of the IFN response relative to the strength of its subversion by HCV , resulted in individual cells admitting either one or both the states . Cells that admitted the first steady state alone were refractory to IFN . Cells that admitted the latter alone were responsive to IFN . Cells that admitted both were bistable and the state they eventually realized depended on whether they were exposed earlier to HCV or IFN . Based on the description above , we divided cells in an individual into three distinct IFN-response phenotypes , IFN-refractory , bistable , and IFN-responsive . With this classification , we constructed a model of viral kinetics that described viral load changes in individuals following the onset of PR treatment . HCV thrived in the IFN-refractory compartment despite exposure to PR . The relative proportion of cells that were of the IFN-refractory phenotype thus quantified the IFN-responsiveness of individuals . By increasing this proportion , which decreased IFN-responsiveness , the formalism was shown to quantitatively capture the observed patterns of viral load decline , from rapid response to null response , in patients undergoing PR treatment [17] . To describe the pre-existence and growth of RAVs under DAA-based treatment , we built on previous models of viral kinetics and evolution [12 , 16 , 18–22] , which have provided good fits to patient data of wild-type and RAV population dynamics [12 , 16] . The models allowed mutations , which occurred during viral replication , at specific loci to confer resistance to DAAs . The mutations , however , came with fitness costs . The models could thus predict the pre-existing frequencies of RAVs and their growth rates during treatment with DAAs . We combined the models above by integrating the distinct cellular phenotypes with viral kinetics and evolution to arrive at a description of the response of an individual to DAA-based treatments and the influence IFN has on the response . The poorer the IFN-responsiveness of an individual , the greater the level of ongoing replication during treatment , and hence the higher the likelihood of the development of resistance to DAAs . The combined model thus yielded threshold levels of IFN-responsiveness required for treatments to succeed . We next developed independent descriptions of the distribution of IFN-responsiveness in populations . We showed that the pre-treatment set-point viral load was directly related to the IFN-responsiveness of an individual , allowing us to quantify the distribution of IFN-responsiveness using measurements of viral load in populations . The fraction of individuals with IFN-responsiveness above the threshold IFN-responsiveness predicted above for any treatment yielded the SVR rates elicited by that treatment . In particular , the threshold for a null response to PR yielded the percentage of null responders and hence , truncating the distribution above to the latter threshold , the distribution of IFN-responsiveness in null responders to PR . Linking the distribution of IFN-responsiveness in populations to the individual-level models of viral kinetics and treatment response thus allowed estimation of SVR rates across different populations , particularly treatment naïve and previous null responders to PR , elicited by different treatment regimens . We applied the model first to examine whether greater IFN-responsiveness lowered the pre-existence of RAVs in infected individuals ( Fig 2 ) . We quantified the IFN-responsiveness of an individual by the fraction of target cells produced in the individual that exhibited the IFN-refractory phenotype [17] , denoted ϕ1p . The smaller the value of ϕ1p , the more IFN-responsive was the individual . Using our model , we estimated the steady state pre-treatment populations of wild-type and RAV-carrying virions , V0 and V1 , respectively , and the frequency , ρ1 = V1/ ( V0+V1 ) , of RAVs , as functions of ϕ1p ( Eqs ( 1 ) – ( 4 ) , Methods ) . A single point mutation was assumed first to confer resistance to the DAA . Mutation occurred at the rate μ and allowed the production of V1 from cells infected with V0 . The mutations came with a fitness cost to the virus , determined by lower values of viral infectivity and/or productivity , γ , relative to the wild-type . We found that ρ1 was independent of ϕ1p ( Fig 2A ) . As ϕ1p increased , fewer cells were IFN-responsive , which resulted in an increase in V1 ( Fig 2C ) . However , of the virions produced ( Fig 2B and 2C ) , a constant fraction , determined by the mutation rate , μ , and the relative fitness of the RAV , γ , carried the RAV , leaving ρ1 unaffected . Further , both V1 and ρ1 but not V0 increased significantly with μ and γ ( Fig 2A–2C ) . We derived analytical approximations ( S1 Text ) that quantitatively explained these variations ( Fig 2A–2C ) . The results were readily extended to multiple loci ( S2 and S3 Texts; Fig 2D–2G ) . Thus , greater IFN-responsiveness did not significantly alter the pre-treatment frequencies of RAVs , although it did lead to lower pre-treatment viral loads and populations of RAVs , indicating greater control over ongoing viral replication . This greater control could influence the growth of RAVs during treatment with DAAs , which we examined next . We predicted the viral load decline under DAA treatment for a range of treatment efficacies against the wild-type , 0≤εDAA0≤1 , and the RAV , 0≤εDAA1≤εDAA0 , and for different fitness penalties , γ , associated with the RAV ( Eqs ( 1 ) – ( 4 ) , Methods ) . We defined the effective relative fitness of the RAV during treatment as γt=γ ( 1−εDAA1 ) / ( 1−εDAA0 ) , combining the intrinsic fitness disadvantage of the RAV and its advantage in the presence of the drug . We defined ϕ1t as the IFN-responsiveness during treatment . ϕ1t depended on the total IFN exposure , the sum of endogenous and exogenous levels [17] . For IFN-free treatments , where exogenous IFN is absent , we let ϕ1t=ϕ1p ( Methods ) . We found that the response to treatment was determined predominantly by εDAA0 , ϕ1p and γt ( Fig 3 ) . With high εDAA0 ( ∼0 . 99 ) , the wild-type could be cleared by the DAA regardless of ϕ1p . Then , for any γt , the decline of the RAVs was faster for lower ϕ1p . Below a critical value of ϕ1p , SVR was achieved , whereas above this critical value , virological breakthrough occurred ( Fig 3A and 3C ) . Similarly , for a fixed ϕ1p , RAVs declined faster for lower γt ( Fig 3A and 3B ) . A locus of points on a ϕ1p−γt plot delineated the region where SVR occurred from that where virological failure resulted due to drug resistance ( Fig 3A ) . For lower ϕ1p , breakthrough occurred at higher γt , indicating that higher degrees of resistance were necessary for virological breakthrough with higher IFN-responsiveness . With lower εDAA0 , the DAA was not potent enough to suppress the wild-type at all ϕ1p . With ϕ1p large and γt small ( <1 ) , the wild-type drove failure ( Fig 3D and 3F ) . The value of ϕ1p above which failure occurred decreased as εDAA0 decreased , indicating that failure occurred even with higher IFN-responsiveness as the DAA efficacy dropped ( Fig 3A and 3D ) . Conversely , poorer IFN-responsiveness placed more stringent demands on the DAA . As γt increased , both the wild-type and RAV co-existed during treatment failure and when γt rose above ~1 , the RAV outcompeted the wild-type and drove treatment failure ( Fig 3D and 3E ) . With exogenous IFN present , ϕ1t<ϕ1p . We therefore let 0≤ϕ1p≤1 and 0≤ϕ1t≤ϕ1p ( Fig 4 ) . For fixed εDAA0 , γt and ϕ1p , RAVs declined faster for lower ϕ1t ( Fig 4B ) . Again , a threshold ϕ1t existed below which SVR was achieved and above which RAVs drove virological breakthrough when εDAA0 was high ( Fig 4A ) . This threshold was weakly sensitive to ϕ1p because pre-treatment variations were rapidly subsumed post treatment initiation by the dynamics dictated by ϕ1t ( Fig 4B ) . The threshold , however , was sensitive to γt . As γt increased , the threshold dropped , indicating that treatment failure occurred even with higher IFN-responsiveness as the RAVs became more fit ( Fig 4C and 4D ) . Similarly , as εDAA0 decreased , failure occurred at lower ϕ1t , indicating again that poorer IFN-responsiveness contributed to the failure of DAAs ( Fig 4E and 4F ) . Further , as γt increased from <<1 to >>1 , failure occurred first due to the wild-type , then the combination of wild-type and RAVs , and finally due to the RAVs alone ( Fig 4G and 4H ) . Thus , with DAA-based treatment , with or without PR , IFN-responsiveness controlled the growth of RAVs and contributed to treatment success . We examined next the implications of these findings at the population level . For this , we first estimated the distribution of IFN-responsiveness across individuals . We recognized that ϕ1p was linked directly to the chronic set-point viral load in our model ( Methods; Eq . ( S1 . 10 ) ) . The set-point viral load has been found to be log-normally distributed in chronically-infected individuals [44] . We therefore let ϕ1p also be log-normally distributed ( Eq ( 5 ) ; Fig 5A ) . Chronic infection was only possible in our model when ϕ1p was larger than a threshold , ϕ1c . When ϕ1p≤ϕ1c , the set-point viral load was zero , marking spontaneous clearance of infection . Using representative model parameter values , we solved our model ( Eqs ( 1 ) – ( 4 ) , Methods ) using different values of ϕ1p and identified ϕ1c as the maximum value of ϕ1p for which the set-point viral load vanished . We thus obtained ϕ1c=0 . 029 . We next fit a truncated log-normal distribution for ϕ1p>ϕ1c ( Eq ( 6 ) , Fig 5B ) to patient data of the distribution of set-point viral load and identified parameter values defining the log-normal distribution of ϕ1p in a treatment-naïve population ( Fig 5D , inset ) . With the resulting distribution , we estimated the percentage of individuals with ϕ1p≤ϕ1c ( Eq ( 8 ) ) , i . e . , the fraction of infected individuals who spontaneously clear the infection , and found it to be ~21% , close to the mean of ~26% obtained from 31 longitudinal studies [45] . We next considered null responders to PR , defined by ϕ1p>ϕnull . To estimate ϕnull , we employed clinical data of telaprevir-based treatments . Using parameter values similar to previous estimates [12 , 46] , εDAA0≈0 . 99 , εDAA1≈0 . 03 , and γ = 0 . 4 , we applied our model ( Eqs ( 1 ) – ( 4 ) ) and estimated ϕSVRDAA=0 . 065 as the value of ϕ1p below which telaprevir monotherapy would elicit SVR ( Methods ) . ( Note that telaprevir monotherapy can induce SVR [47] . ) We next estimated ϕSVRPR+DAA , the value of ϕ1p below which PR+telaprevir triple therapy would yield SVR , by comparing model predictions of SVR rates in treatment-naïve patients ( Eq ( 9 ) ) with corresponding clinical data [48] , SVRnaivePR+DAA=75 . 4±2 . 4% . Given the distribution of ϕ1p , the value of ϕ1p below which a defined percentage of the population lies can be calculated . Thus , the value of ϕ1p below which SVRnaivePR+DAA=75 . 4±2 . 4% of the population lies yielded ϕSVRPR+DAA=0 . 152±0 . 011 . This implied Δϕ=ϕSVRPR+DAA−ϕSVRDAA=0 . 087±0 . 011 was the increase in IFN-responsiveness due to exogenous IFN administered as part of PR treatment . With this Δϕ , we could estimate ϕ1t for any ϕ1p , allowing us to use our model ( Eqs ( 1 ) – ( 4 ) ) to predict the response to PR-containing regimens . In particular , we could predict the response to PR . Solving our model , we thus identified ϕnull as the minimum ϕ1p that yielded a null response to PR , defined as having occurred when <2 log10 decline in viral load resulted from 12 weeks of treatment . We found that ϕnull = 0 . 12±0 . 01 ( Eqs ( 1 ) – ( 4 ) , Methods ) . Truncating the distribution of ϕ1p to values of ϕ1p above ϕnull yielded the distribution of ϕ1p in null responders to PR ( Eq ( 7 ) , Fig 5C ) . With these estimates , we predicted the percentage of null responders to PR in a treatment-naïve population as the fraction of the population with ϕ1p above ϕnull ( Eq ( 11 ) ) and found NULLnaivePR=33% , which was in close agreement with corresponding clinical observations of 32% [49] . Further , we predicted the response of null responders to PR+telaprevir triple therapy ( Eq ( 13 ) ) as the fraction of null responders with ϕ1p below ϕSVRPR+DAA estimated above , and found SVRnullPR+DAA=26% , again in good agreement with the 32% observed experimentally [48] . This quantitative agreement of our model with independent observations gave us confidence in our model and our estimates of the distribution of IFN-responsiveness . In addition , that the same Δϕ captured the observed influence of PR with and without telaprevir implied that the proposed synergy between IFN and DAAs [17 , 50 , 51] may be small in vivo . We next predicted the response of different patient subpopulations to DAA-based treatments . Using the distributions of IFN-responsiveness identified above , we applied our model ( Eqs ( 1 ) – ( 13 ) , Methods ) to predict SVR rates elicited by DAA-based treatments . We varied parameters to mimic the entire spectrum of accessible DAA efficacies and relative fitness values of RAVs . We found that SVRnull was consistently lower than SVRnaive ( Fig 5D ) . Further , our predictions were in good agreement with clinical data ( Fig 5D; Table 1 ) . The predictions employed some parameters estimated above from telaprevir-based therapy . We derived an analytical expression linking SVRnaive and SVRnull that was independent of the DAA and of whether PR was part of the treatment ( Eq ( 18 ) , Methods ) : SVRnaive=1−NULLnaivePR+NULLnaivePRSVRnull . The expression too fit the clinical data well and the fit was close to our predictions above ( Fig 5D ) . The best-fit estimate of NULLnaivePR=40±14% was in agreement with the corresponding clinical estimate [49] of 32% . Further , the latter expression explained more vividly the diminishing difference between SVRnaive and SVRnull as treatment became more potent . It showed that when SVRnaive approached 100% , so did SVRnull , in agreement with observations from recent trials where nearly all patients were cured regardless of their treatment experience ( Table 1 ) . This agreement between our predictions , in two ways , and clinical data demonstrating SVRnull ≤ SVRnaive presents strong evidence supporting our hypothesis of the causal relationship between IFN-responsiveness and the success of DAA-based treatments . We examined ways of exploiting this relationship to improve treatments . IFN-responsiveness could be exploited to improve DAA-based treatments in two ways: 1 ) to prevent failure and 2 ) to shorten the treatment duration ( Fig 6 ) . IFN-free treatment would fail in an individual if ϕ1p in the individual were larger than a threshold . Adding PR would lower ϕ1p by Δϕ . Δϕ is expected to be different for individuals with ( denoted Δϕc ) and without ( denoted Δϕnc ) liver cirrhosis . We fit the analytical expression above ( Eq ( 18 ) ) to SVR data on populations with and without liver cirrhosis separately ( Fig 6A and 6B ) and estimated Δϕc = 0 . 046 and Δϕnc = 0 . 091 ( Methods ) . PR thus appeared only half as effective in improving IFN-responsiveness in cirrhotic individuals as non-cirrhotic individuals . Adding PR would thus induce SVR if the individual had a cirrhotic ( non-cirrhotic ) liver and the ϕ1p were within Δϕc ( Δϕnc ) of the threshold ( Fig 6C , 6D , 6E and 6G ) . If ϕ1p were farther away from the threshold , adding PR alone would prove inadequate ( Fig 6D , 6F and 6H ) . Increasing the DAA dosage or including additional DAAs to lower the effective fitness of the RAV may then be a way to induce SVR ( Fig 6C–6G ) . Of course , adding PR may require a smaller increase in the DAA dosage , rendering the DAA more tolerable . Even where DAA-based treatment is likely to succeed , greater IFN-responsiveness would induce faster viral load decline and allow shorter treatment durations ( Fig 6I and 6J ) . Using parameters representative of daclatasvir ( Methods ) , we found that as ϕ1p decreased from ~0 . 1 to ~0 . 04 , the treatment duration required for SVR dropped from ~12 weeks to ~8 weeks ( Fig 6I and 6J ) . Although daclatasvir is not recommended for use as monotherapy , we use it here for illustration and because the NS5A region , the target of daclatasvir , is the one region where resistance testing may decide the choice of regimen [3] . Thus , individuals highly responsive to IFN present promising candidates for reducing DAA treatment durations . Indeed , we estimated that ~50% of the individuals treated with daclatasvir would achieve SVR in ~10 weeks and ~20% in ~8 weeks , durations expected to decrease further with DAA combinations , presenting a basis and a novel avenue for response-guided treatment .
DAAs , with >90% SVR rates in clinical trials , are bringing hope to the millions of chronically HCV infected individuals worldwide . In the present study , we elucidated a hypothesis underlying the unexpected positive correlation between the response elicited by DAAs and PR , which explains several confounding clinical observations and presents new potential avenues to improve DAA-based treatments . The hypothesis is that greater IFN-responsiveness restricts the replication space available for the virus , inhibiting the development of resistance to DAAs and improving treatment response . We developed a novel multiscale mathematical model to test this hypothesis . Analysis of a large body of clinical data using the model presented evidence in strong support of the hypothesis . The resulting causal relationship between responsiveness to PR and DAAs implied that increased responsiveness to PR could be exploited to prevent DAA failure and/or shorten the treatment duration , potentially positively impacting treatment response , tolerability , affordability and access . Despite the high SVR rates they elicit , access to DAA-based treatments has seen limited so far; <1 . 3% of the ~150 million chronically HCV infected individuals are estimated to have received DAA-based treatment , with the proportion far smaller in resource-limited settings [52 , 53] . To improve affordability and access , DAA-based treatments that would exert the most potent antiviral activity and/or patient subpopulations that would require the shortest durations are keenly being sought [54–59] . Our study informs these efforts by presenting new avenues to optimize DAA-based treatments . The standard strategy to avert DAA failure is to increase the genetic barrier to resistance by including more DAAs in the drug cocktail [12 , 54] . In a recent set of studies , for instance , numerous DAA combinations were evaluated preclinically to identify the “best” candidates for clinical development and 3 DAA combinations were found to be more potent than 2 DAA combinations [54–56] . We suggest that an alternative strategy may often be feasible: improving IFN-responsiveness by adding IFN ( or PR ) . Where additional DAAs remain inaccessible , especially in resource-limited settings , such a strategy may be useful . A previous modeling study also found that the efficacy of combining PR with a DAA significantly improved the treatment efficacy against DAA-resistant strains compared to the DAA alone [12] . Furthermore , our study quantified the gain in IFN-responsiveness due to standard PR dosage in patients with and without liver cirrhosis , representing a key potential step in personalizing the strategy . Response-guided treatment ( RGT ) is being considered now to define reduced treatment durations for select populations [27 , 57 , 58 , 60 , 61] . For instance , in a recent study , patients who achieved an ultra-rapid early viral load decline ( plasma HCV RNA <500 copies/mL by day 2 ) were found to be cured with just 3 weeks of treatment [57] . Our prediction that individuals with high IFN-responsiveness are amenable to shorter treatment durations presents a much sought-after basis and a promising candidate population for reducing treatment durations , informing ongoing efforts to develop RGT protocols . Personalizing treatment based on the avenues above requires estimation of the level of IFN-responsiveness of individuals . For a treatment-experienced individual , this may be achieved through analysis of viral load changes recorded during the previous PR treatment [17] . For a treatment-naive individual , short-term PR exposure and subsequent measurements of viral load may be necessary . Viral load changes as early as 24 h following the start of PR treatment have been argued to be good indicators of eventual response [62 , 63] . Previous modeling studies have suggested a lead-in period of PR to assess the level of ongoing viral replication and the responsiveness to PR in order to decide optimal treatments [12 , 64] . Developing such indicators to quantify the level of IFN-responsiveness , a promising future research direction based on our present study , would allow personalizing the course of DAA treatments also for treatment-naive individuals . Further , a correlation between IFN-responsiveness and the duration of DAA-based treatment required to achieve SVR would present a direct clinical test of our hypothesis . Recent studies present further evidence in support of our hypothesis . In a study involving 240 chronic HCV patients treated with sofosbuvir and either daclatasvir or simeprevir for 12 weeks , slow responders , defined as those with detectable viremia at week 12 , had a much higher representation of treatment ( PR ) -experienced patients than the overall population , viz . , 82% versus 68% , indicating that IFN-free DAA treatments elicited slower viral load declines in individuals with poorer IFN-responsiveness [65] . Another study involving 216 patients treated with sofosbuvir and either daclatasvir or ledipasvir for 12 weeks found that baseline RAVs and treatment experience did not influence SVR in patients without cirrhosis but had a significant influence in patients with cirrhosis [66] . In yet another study [13] , 6 of the 8 patients treated with daclatasvir and PR achieved SVR , of which 4 had RAVs detected pre-treatment , but had favourable IL28B genotypes ( TT/GT ) [37] and were treatment-naive or partial responders to prior IFN therapy . The 2 who failed treatment had unfavorable/partially favorable IL28B genotypes ( GG/GT ) and were null responders to prior IFN therapy . They experienced virological breakthrough due to the growth of RAVs although the RAVs were not detected pre-treatment . Earlier studies with the first generation DAAs provide further evidence . In a pooled study involving a large number of patients treated with boceprevir and PR , SVR rates were 78% and 76% in IFN-responders with and without baseline RAVs , respectively , whereas in poor IFN-responders , the corresponding SVR rates were 22% and 37% [67] . Similarly , with telaprevir , on-treatment virological failure rates were 1% in previous relapsers to PR , 19% in previous partial responders to PR , and 52% in previous null responders to PR [68] . Thus , in all the above cases , treatment failure was due to drug resistance , which did not depend on the pre-existence of RAVs but was facilitated by poor IFN-responsiveness . Conversely , strong IFN responses appeared to prevent the growth of RAVs and avert treatment failure . These findings are consistent with our model predictions . Our study makes key advances in our understanding of HCV infection and treatment . To test the hypothesized causal relationship between responsiveness to IFN and DAAs , we had to construct a model that integrated phenomena across multiple length and time scales , starting from the cellular to the population level . Responsiveness to IFN manifests at the cellular level , defining the fraction of cells that can be rid of HCV by IFN . The consequence at the level of an infected individual is in restricting viral replication and evolution and hence improving responsiveness to DAAs . At the population level , this effect , given the distribution of IFN-responsiveness , determines SVR rates . Integration of phenomena across these scales into a single mathematical framework had not been accomplished thus far . Our model , by doing so , was able to capture the implications of variations at the cellular level , due to drugs , for instance , for the population-level treatment response . This allowed us to describe many clinical observations of which several had long remained unexplained , viz . , 1 ) the percentage of infected individuals who spontaneously clear HCV , 2 ) the percentage of chronically infected individuals who exhibit a null response to PR , 3 ) the percentage of null responders to PR who respond to triple therapy with PR and a DAA such as telaprevir , and 4 ) the relationship between SVR rates in treatment-naive and treatment-experienced patients elicited by different treatments . A far more comprehensive view of HCV infection and treatment than earlier thus emerges . The model we developed is complex . Yet , we only considered phenomena essential to establishing the causal relationship between IFN-responsiveness and DAA-based treatment outcomes . We thus ignored alternative mechanisms of DAA action [23 , 69] , specific intracellular viral replication events [11 , 24] , modes of synergy between IFN and DAAs [17 , 51] , and factors such as race , gender , viral genotype , and IL28B polymorphisms [1] . Furthermore , we did not estimate the IFN-responsiveness of an individual a priori . The key components of the IFN signaling network in cells have been identified [28] , but variations in their levels and interactions across cells in an individual , which would determine the fraction of cells responsive to IFN , remain to be established . Finally , we assumed SVR to have been achieved when the viral load dropped below the “cure boundary” of 1 virion in the ~15 liters of fluid volume in an individual [5 , 16] . With the new DAA combinations , some individuals with detectable viremia at the end of treatment have been found recently to achieve SVR [70] . The origins of this intriguing phenomenon , which may lead to the definition of a new cure boundary , remain poorly elucidated [71–73] . By employing the “stricter” cure boundary , our model yields conservative estimates of SVR rates .
We let SVRnaive and SVRnull be the response to any given DAA-based treatment in a treatment-naïve and a previous null responder population , respectively . We derived an analytical expression linking SVRnaive and SVRnull as follows . From our model above , it followed that SVRnaive=∫ϕ1cϕSVRfnaive ( ϕ1p ) dϕ1p ( 14 ) and SVRnull=∫ϕnullϕSVRfnull ( ϕ1p ) dϕ1p . ( 15 ) Using the relationship between fnaive ( ϕ1p ) and fnull ( ϕ1p ) ( Eq ( 7 ) ) in Eq ( 15 ) yielded SVRnull=∫ϕnullϕSVRfnaive ( ϕ1p ) dϕ1p∫ϕnull1fnaive ( ϕ1p ) dϕ1p=1NULLnaivePR∫ϕnullϕSVRfnaive ( ϕ1p ) dϕ1p ( 16 ) where NULLnaivePR is defined in Eq ( 11 ) . We next rearranged the integral in Eq ( 16 ) as ∫ϕnullϕSVRfnaive ( ϕ1p ) dϕ1p=∫ϕnullϕ1cfnaive ( ϕ1p ) dϕ1p+∫ϕ1cϕSVRfnaive ( ϕ1p ) dϕ1p=− ( 1−NULLnaivePR ) +SVRnaive ( 17 ) and obtained upon combining with Eq ( 16 ) and rearranging , SVRnaive=1−NULLnaivePR+NULLnaivePRSVRnull . ( 18 ) We fit the expression above to clinical data using the NLINFIT algorithm in MATLAB . We examined reports of all clinical trials with DAA-based treatments and considered those treatments for which SVR data on both treatment-naïve and treatment-experienced individuals was available . The resulting data is summarized in Table 1 and S1 Table . We also classified the patient populations into categories with and without liver cirrhosis . We performed statistical tests to ascertain the difference in the SVR rates between treatment-naïve and treatment-experienced individuals for specific treatments as well as when data for all the treatments considered were combined . We compared the predictions above of SVR rates with the data from clinical trials . Next , we considered SVR data on patients with and without liver cirrhosis separately . Predicting this data using our model was not possible because Δϕ and ϕnull in these populations were not known . We therefore fit Eq ( 18 ) to the two data sets separately using NULLnaivePR as an adjustable parameter . The distribution of baseline viral loads is not hugely different between the two populations , although mean viral loads were somewhat smaller in patients with liver cirrhosis [44] . Using Eq ( 11 ) and the best-fit estimates of NULLnaivePR , we obtained estimates of ϕnull in the two populations . Finally , we solved our model of viral dynamics ( Eqs ( 1 ) – ( 4 ) ) with ϕ1p=ϕnull for different values of ϕ1t=ϕ1p−Δϕ and identified the highest value of ϕ1t that yielded a null response . The resulting values of Δϕ=ϕ1p−ϕ1t provided estimates of the extent of increase in IFN-responsiveness due to standard PR treatment in cirrhotic and non-cirrhotic patients , respectively , which allowed recommendation of strategies to improve treatments in these subpopulations . Finally , we calculated the required duration of treatment to achieve SVR for different ϕ1p and γt . We chose parameters representative of daclatasvir as follows . The EC50 of daclatasvir against the wild-type [77] and the RAV [78] were 17 . 28 pM and 32346 . 26 pM , respectively . ( The molecular weight of daclatasvir is 738 . 89 g/mol . ) Using the pharmacokinetic parameters of daclatasvir [79] , the peak and trough plasma concentrations , Cmax = 1726 . 4 ng/ml and Cmin = 254 . 6 ng/ml , the dosing interval of 1 d , and the time to reach peak plasma concentration of 1 h , we calculated the average efficacy of daclatasvir against the wild-type and the RAV , following the procedure outlined earlier [77] , to be 0 . 99998 and 0 . 709 , respectively . With these values , we solved our model of viral dynamics ( Eqs ( 1 ) – ( 4 ) ) for different values of ϕ1p and the intrinsic fitness of the RAV , γ , and estimated the duration of treatment required to achieve SVR . For the common RAV Y93H , γ = 0 . 5751 [77] . Using this value of γ and the distribution of ϕ1p in treatment naïve individuals ( Eq ( 6 ) ) , we estimated the fraction of individuals treated who would achieve SVR within a defined treatment duration . | Treatment of hepatitis C virus ( HCV ) infection is seeing a paradigm shift with powerful drugs called direct-acting antiviral agents ( DAAs ) replacing earlier treatments involving interferon ( IFN ) . DAAs target specific HCV proteins . IFN stimulates our immune response against HCV . The two should thus work independently . Surprisingly , DAAs appear to work better in individuals who also tend to respond well to IFN . We hypothesized here that responsiveness to DAAs and IFN are causally linked . IFN can suppress viral replication , preventing the development of resistance to DAAs and improve DAA treatments . Using a new multiscale mathematical model and analysis of a vast body of clinical data , we found strong evidence supporting this hypothesis . Leveraging the causal relationship , we suggest new ways of optimizing DAA treatments , potentially improving their efficacy , tolerability , affordability and access . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"hepacivirus",
"pathogens",
"microbiology",
"viral",
"structure",
"liver",
"diseases",
"viruses",
"rna",
"viruses",
"pharmaceutics",
"gastroenterology",
"and",
"hepatology",
"pharmacology",
"viral",
"load",
"antimicrobial",
"resistance",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"hepatitis",
"c",
"virus",
"hepatitis",
"viruses",
"viral",
"replication",
"virions",
"cirrhosis",
"biochemistry",
"flaviviruses",
"virology",
"viral",
"pathogens",
"interferons",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"drug",
"therapy",
"organisms"
] | 2018 | Modelling how responsiveness to interferon improves interferon-free treatment of hepatitis C virus infection |
Dendritic cells ( DCs ) as professional antigen-presenting cells play an important role in the initiation and modulation of the adaptive immune response . However , their role in the innate immune response against bacterial infections is not completely defined . Here we have analyzed the role of DCs and their impact on the innate anti-bacterial host defense in an experimental infection model of Yersinia enterocolitica ( Ye ) . We used CD11c-diphtheria toxin ( DT ) mice to deplete DCs prior to severe infection with Ye . DC depletion significantly increased animal survival after Ye infection . The bacterial load in the spleen of DC-depleted mice was significantly lower than that of control mice throughout the infection . DC depletion was accompanied by an increase in the serum levels of CXCL1 , G-CSF , IL-1α , and CCL2 and an increase in the numbers of splenic phagocytes . Functionally , splenocytes from DC-depleted mice exhibited an increased bacterial killing capacity compared to splenocytes from control mice . Cellular studies further showed that this was due to an increased production of reactive oxygen species ( ROS ) by neutrophils . Adoptive transfer of neutrophils from DC-depleted mice into control mice prior to Ye infection reduced the bacterial load to the level of Ye-infected DC-depleted mice , suggesting that the increased number of phagocytes with additional ROS production account for the decreased bacterial load . Furthermore , after incubation with serum from DC-depleted mice splenocytes from control mice increased their bacterial killing capacity , most likely due to enhanced ROS production by neutrophils , indicating that serum factors from DC-depleted mice account for this effect . In summary , we could show that DC depletion triggers phagocyte accumulation in the spleen and enhances their anti-bacterial killing capacity upon bacterial infection .
Innate immunity as well as adaptive immunity is involved in the response of the host towards pathogens [1]–[3] . Dendritic cells ( DCs ) are professional antigen presenting cells playing a central role in immune response by linking the innate and adaptive immunity [4]–[6] . The activation of innate immune cells by microorganisms occurs via binding of pathogen-associated molecular patterns ( PAMPs ) to pattern-recognition receptors ( PRRs ) , e . g . Toll-like receptors ( TLRs ) [7] . Upon stimulation by TLR ligands , DCs mature and migrate from the site of infection to secondary lymphoid organs to induce pathogen-specific T-cell responses . Although the role of DCs in the initiation of the adaptive immune response is well established , their impact on immune cells of the innate immune response is less examined . Previous studies showed that the induction of sepsis in mice resulted in a profound loss of CD11c+ DCs from spleen and lymph nodes [8] , [9] . The administration of LPS or Escherichia coli in mice causes a pronounced reduction in DC numbers in the spleen induced by apoptosis [10] , [11] . It was also shown that patients suffering from sepsis displayed increased apoptosis of DCs in the spleen and that an early decrease in circulating DCs was correlated with increased disease severity and mortality [12] , [13] . Scumpia et al . showed that DCs were essential in the immune response to sepsis and suggested that strategies to maintain DC numbers or function may improve the outcome during polymicrobial sepsis [14] . We have recently shown that the Gram-negative bacterium Yersinia enterocolitica ( Ye ) affects the homeostasis of the CD4+ DCs and , to a lesser extent , the CD8α+ DC population in the spleen by the induction of cell proliferation and suppresses de novo DC generation [15] . While the role of DCs in adaptive host defense by instructing T cells is well established , their potential contribution to T cell independent innate host defense is poorly understood . In particular , interactions between DCs and phagocytes in the course of infection have not yet been addressed in depth . Therefore , the aim of the study was to address the importance of DCs for the innate immune response in vivo upon bacterial infection with Ye . This bacterium causes food borne acute and chronic gastrointestinal and systemic diseases in both humans and mice [16] . By means of its type III secretion system Ye is able to translocate its effector proteins ( Yops ) directly into the cytosol of host cells [17] , thereby preventing its uptake by the target cells . Phagocytosis and subsequent destruction of the pathogens are critical in the innate immune response . Professional phagocytes , such as neutrophils , macrophages , monocytes and DCs , are specialized to engulf large particles , including microorganisms . Monocytes arise from myeloid progenitors in the bone marrow and are defined as non-dividing circulating blood cells with a half-life of one day in mice [18] . Mouse blood monocytes express CD115 , CD11b and low levels of F4/80 and can be distinguished by the expression of Ly6C and CX3CR1 into Ly6ChiCX3CR1loCCR2+CD62L+ and Ly6CloCX3CR1hiCCR2−CD62L− monocytes [18] . Neutrophils are terminally differentiated effectors and the first cells to migrate toward sites of infection . Release of neutrophils from the bone marrow is mediated by the concerted action of G-CSF , CXCL1 , and CXCL2 [19] , [20] . At the site of infection , neutrophils engulf and kill bacteria through the production and secretion of proteases , reactive oxygen species and other proinflammatory mediators . Furthermore , neutrophils control the recruitment of other cells ( T cells , NK cells , macrophages , and immature DCs ) through the production of the chemokines CXCL1 , CCL3 , and CCL4 [21] . Early upon activation DCs also produce IL-8 thereby attracting neutrophils which leads to colocalization of neutrophils and immature DCs [22] . Mouse neutrophils express TLR2 , TLR4 and TLR9 mRNAs [23] , and can be activated by LPS leading to shedding of L-selectin and upregulation of CD11b [24] . In this study , we have used an inducible mouse model allowing depletion of CD11chi DCs by administration of diphtheria toxin ( DT ) to directly address their impact during initiation of innate immune response upon Ye infection in vivo [25] . We found that DC depletion per se increased the number of phagocytes and enhanced their anti-bacterial host defense in the spleen leading to increased survival of the mice upon Ye infection .
To address the impact of DCs on the survival of the mice upon a severe bacterial infection we administered DT i . p . one day prior to i . v . infection with 5×104 Ye to DC-depleted and control mice . DT was administered daily during the whole period of observation and survival was analyzed for up to 14 days . DC-depleted mice survived significantly longer than control mice with median survival of 12 . 5 days and 7 days , respectively ( p<0 . 005 ) , indicating that DC depletion was beneficial for the survival upon a lethal Ye infection ( Figure 1A ) . The survival rate correlated with the bacterial load in the spleen which displayed an increase in the colony forming units ( CFU ) over time ( Figure 1B ) . Overall , we observed significantly less bacterial load in the DC-depleted mice compared to control mice ( Figure 1B ) . Since the cellular composition of the spleens is changing with treatment and infection we analyzed the CFU/g spleen ( Figure S1B ) observing similar results . CD11c . DOG mice without DT treatment showed similar CFU in the spleen upon Ye infection as control mice ( Figure S1C ) , excluding intrinsic differences in the susceptibility of the mice which were used in this study . In addition , we performed immunofluorescence microscopy of cryosections from the spleen of Ye-infected mice staining both CD11c+ cells and Ye . In agreement with flow cytometry analysis ( Figure S1A ) the number of CD11c+ cells was found to be low in DC-depleted mice due to DT administration . Ye infection led to a decrease in the number of CD11c+ cells in control mice ( Figure 1C ) . This is consistent with previous findings from our group [15] . Moreover , massive abscess formation in the spleen of control mice was observed from 3 to 7 days post infection ( dpi ) , whereas in the spleen of DC-depleted mice only small abscesses were found at 3 dpi ( Figure 1C ) . Sepsis is characterized by increased levels of proinflammatory cytokines . One day post Ye infection the levels of the proinflammatory cytokines IL-6 , IFN-γ , and IL-12p40 were 2 to 10-fold increased in sera from control mice compared to infected DC-depleted mice or mice without infection , indicating that DCs promote the production of these proinflammatory mediators upon Ye infection ( Figure 1D ) . Altogether , these studies demonstrate that the depletion of DCs is beneficial for survival upon severe bacterial infection and is associated with lower bacterial load and lower production of proinflammatory cytokines . As DC-depleted mice displayed a significantly lower bacterial load in the spleen , already 1 dpi compared to control mice ( Figure 1B ) we hypothesized that this could reflect an altered splenocyte composition following DC depletion prior to infection . In fact , single DT treatment of uninfected mice led to a 3 to 4-fold increase in the frequency of inflammatory monocytes ( Gr-1+Ly6G−Ly6ChiCD11b+ ) and neutrophils ( Gr-1hiLy6G+Ly6C−/intCD11bhi , see Figure S2 for detailed gating strategy ) after 24 h in the spleen of DC-depleted mice compared to DT-treated control mice ( Figure 2A and [25] ) . Similarly , increased numbers of neutrophils and monocytes were observed in peripheral blood ( data not shown ) . However , we did not observe differences in the frequency of B cells , T cells or NK cells ( data not shown and [25] ) . To rule out the possibility that massive DC cell death and/or the phagocytosis of the DC debris might serve as a general pro-inflammatory signal and thereby causing the recruitment of neutrophils , we analyzed neutrophil numbers in the spleen from mixed bone marrow chimeras ( 80% CD11c . DOG/20% C57BL/6 ) . In these mice , a single DT treatment leads to the depletion of most DCs . However , 10 daily DT applications result in depletion of the pre-existing DCs and those that are continuously being generated from CD11c . DOG progenitors , while the C57BL/6 DC pool expands until reconstituting the whole compartment [26] . Thus , 10 days of DC depletion in these chimeric mice results in a normal DTR− DC compartment with massive DTR+ DC depletion . As expected , single DT treatment led to neutrophilia , similar to that in CD11c . DOG mice ( Figure S3 ) . Interestingly , 10 days of DT treatment in the chimeric mice revealed only a minor increase in the number of splenic neutrophils ( about 2 . 7-fold ) compared to the 27-fold increase observed in CD11c . DOG mice ( Figure S3 ) . Similar results were observed when quantifying the frequency of neutrophils ( Figure S3 ) . Furthermore , we found that the increased numbers of phagocytes in the spleen upon DC depletion are accompanied by significantly elevated serum levels of CCL2 , G-CSF , CXCL1 , Flt3L , and IL-1α from DC-depleted mice compared to control mice ( Figure 2B ) , all of which have been shown to be involved in leukocyte recruitment or maintenance [27]–[30] . These data point towards a regulation of neutrophil numbers by DCs via the repression of chemokines/growth factors rather than merely a side effect caused by DC death . A further increase in the frequency of monocytes and neutrophils in the spleen of both DC-depleted and control mice was observed upon Ye infection . The frequency of monocytes in the spleen of DC-depleted mice continuously increased up to 1 dpi and was 2 . 5-fold higher compared to control mice ( Figure 2C ) . The frequency of neutrophils continuously increased up to 6 h post infection . This was more pronounced in control mice , and reached similar frequency and numbers 6 h and 1 dpi in both groups of mice ( Figure 2C and Figure S4 ) . This is accompanied by significantly elevated serum levels of CCL2 , G-CSF , and CXCL1 in control mice 1 dpi compared to DC-depleted mice ( Figure 2D ) . These data suggest different recruitment kinetics of monocytes and neutrophils in response to Ye , similar as shown for intradermal E . coli infection [31] . Immunofluorescence microscopy confirmed the increase in the number of Gr-1+ cells in the red pulp upon DC depletion . Ye infection led to increased numbers and accumulation of Gr-1+ cells in the splenic red pulp of control mice and was associated with the formation of abscesses , whereas in DC-depleted mice the Ye-induced increase in Gr-1+ cells was more uniformly distributed and associated with the formation of microabscesses ( Figure 3 ) . In summary , our data indicate that DCs may regulate the numbers of splenic neutrophils and monocytes associated with increased chemokine production by a yet unknown mechanism . As DC depletion led to increased numbers of monocytes and neutrophils in the spleen , both of which are professional phagocytes , we hypothesized that these cells account for lower bacterial load observed already 30 min after Ye infection of DC-depleted mice ( Figure S5 ) . Therefore , phagocytosis of eGFP-expressing Ye by splenocytes was analyzed 30 min post intravenous administration . Flow cytometry analysis revealed two times less splenocytes associated with eGFP-Ye ( referred to as Ye+ cells ) ( Figure 4A , R1 , p<0 . 001 ) in the spleen from DC-depleted mice compared to control mice . Detailed flow cytometry analysis showed , however , striking differences in Ye+ cells in the various spleen cell subpopulations . In fact , Ye+ splenocytes from control mice comprised predominantly B cells ( 70% ) as well as DCs and neutrophils ( each 10% ) , whereas the Ye+ splenocytes from DC-depleted mice comprised predominantly neutrophils ( 32% ) and monocytes ( 17% ) ( Figure 4B ) . Calculation of the total numbers of Ye+ cells per spleen revealed 2 . 2×105 Ye+ neutrophils , 1 . 2×105 Ye+ monocytes , and 2 . 0×105 Ye+ B cells in DC-depleted mice compared to 1 . 0×105 Ye+ neutrophils , 0 . 2×105 Ye+ monocytes , and 12 . 0×105 Ye+ B cells in control mice ( Figure 4B ) , demonstrating that DC depletion increased the number of phagocytes associated with Ye in vivo , whereas the number of Ye+ B cells was dramatically reduced . Furthermore , immunofluorescence microscopy of cryosection from the spleen of DC-depleted mice 30 min post Ye infection revealed a low number of Ye and these were found next to clusters of Gr-1+ cells ( Figure 4C ) . In contrast , Ye colonies were obvious in the spleen from control mice and these were partially associated with a lower number of Gr-1+ cells as found in DC-depleted mice ( Figure 4C ) . In addition , analyzing the cell contact of neutrophils with DCs by immunofluorescence microscopy , we hardly observed colocalization of Ly6G+ cells with CD11c+cells in control mice ( Figure S6 ) , arguing against a direct cell contact-dependent regulation of neutrophils by DCs . To dissect whether Ye+ splenocytes reflect Ye associated with the membrane of the cells or Ye engulfed by the cells , we used multispectral imaging flow cytometry combining flow cytometry with microscopy at the single cell level . B cells were stained with CD19 and B220 antibodies , whereas monocytes and neutrophils were distinguished by CD11b and Ly6C surface staining ( Figure 5 and Figure S2 ) . Intracellular Ye were defined as described in Materials and Methods . By analyzing phagocytosis with this technique we found that 50–70% of neutrophils and monocytes harbor intracellular Ye . In contrast , only 10% of all Ye associated with B cells were intracellularly located . The frequencies of intracellular Ye in the various spleen cell subpopulations were similar in DC-depleted and control mice ( Figures 5A and B ) , indicating no differences in the phagocytosis rate of the splenocytes . In addition , colocalization of Ye with CD107a ( LAMP-1 ) protein expressed in late endosomes and lysosomes was analyzed as indicators of bacterial processing . Intracellular Ye in neutrophils and monocytes colocalized with the lysosomal marker CD107a ( Figures 5A and C: similarity score Ye/CD107a >1 ) but no difference was obvious in the phagocytes from DC-depleted and control mice , indicating similar bacterial processing rates in both groups of mice . Taken together , our data show that DC depletion did not affect the capacity of neutrophils and monocytes to engulf and process Ye in vivo . However , DC depletion led to a strong accumulation of neutrophils and monocytes in the spleen resulting in Ye being predominantly associated with these phagocytes . Despite similar bacterial phagocytosis and processing rates in DC-depleted and control mice , we hypothesized that the differences in the bacterial load were due to increased intracellular killing mechanism by the phagocytes from DC-depleted mice . Performing an in vitro killing assay , we observed as early as 10 min after incubation of splenocytes with Ye ( multiplicity of infection 1 ) , that the number of recovered intracellular bacteria was reduced by 86 . 6% in splenocytes from DC-depleted mice compared to only 64 . 3% in splenocytes from control mice ( Figure 6A ) . These data indicate that DC depletion resulted inmore efficient killing of Ye by splenocytes compared to the killing capacity by splenocytes from control mice . Additionally , the number of living intracellular Ye in sorted CD11b+Gr-1− cells and neutrophils was 4 to 14-fold higher one day and 27 to 49-fold higher three days post Ye infection in control mice compared to DC-depleted mice ( Figure 6B ) , indicating a better bacterial killing by the phagocytes from DC-depleted mice . To corroborate this hypothesis , ROS production was analyzed in both mice without and with Ye infection in vivo . We observed increased ROS levels in neutrophils from DC-depleted mice prior to infection ( Figure 6C left diagram ) as well as 2 h post Ye infection ( Figure 6C right diagram ) , indicating that the neutrophils were activated upon DC depletion and infection . We did not observe differences in ROS production by monocytes neither with nor without infection ( Figure 6C ) . These data provide evidence that DCs not only affect the number of neutrophils in the spleen but also increase their anti-bacterial killing capacity . To directly demonstrate that the increased number of neutrophils in combination with their increased ROS production in DC-depleted mice account for the initially decreased bacterial load in the spleen , neutrophils from DC-depleted mice were purified and adoptively transferred to control mice . These mice were then infected with Ye ( see Figure 6D ) . As a control , purified neutrophils from control mice were transferred into control mice prior to Ye infection . The bacterial load in the spleen of mice adoptively transferred with neutrophils from DC-depleted mice was significantly reduced compared to control mice without adoptive transfer and similar to that of DC-depleted mice 1 dpi ( Figure 6D ) . In contrast , adoptive transfer of neutrophils from control mice into control mice did not lead to a significant reduction of the bacterial load 1 dpi with Ye ( Figure 6D ) . In summary , our data demonstrate that DC depletion leads to increased number of phago cytes in >the spleen being highly effective in the clearance of bacteria . In order to analyze whether serum factors mediate the increased ROS production and killing capacity of neutrophils , we incubated splenocytes from control mice with serum from DC-depleted or control mice for 1 h . The analysis of ROS production by flow cytometry revealed a significant increase in ROS production by neutrophils incubated with serum from DC-depleted mice compared to neutrophils incubated with serum from control mice ( Figure 7A ) . ROS production in neutrophils from control mice incubated with serum from DC-depleted mice was comparable to neutrophils from DC-depleted mice incubated with serum from DC-depleted mice ( Figure 7A ) . Furthermore , splenocytes from control mice showed increased bacterial killing after incubation with serum from DC-depleted mice compared to incubation with control serum ( Figure 7B ) . These data indicate that indeed factors in the serum from DC-depleted mice cause elevated ROS production in neutrophils and enhance the bacterial killing capacity of splenocytes in vitro . To evaluate these findings in vivo , purified neutrophils from control mice ( CD45 . 1+ ) were adoptively transferred into either control mice or DC-depleted mice and analyzed for ROS production 2 h after transfer . We observed a higher frequency of transferred CD45 . 1+Ly6G+ neutrophils ( Figure 7C , R1 and Figure 7D ) in the spleen of DC-depleted compared to control mice . This indicates a better attraction of PMNs into the spleen upon DC-depletion . Furthermore , neutrophils tended to produce more ROS when transferred into DC-depleted mice compared to transfer into control mice , although differences in ROS production were statistically not significant ( Figure 7D ) . Within these transferred neutrophils two subpopulations could be distinguished by their expression of CD11b and Ly6G ( Figure 7C , R2: CD11bhiLy6Ghi and R3: CD11b+Ly6G+ ) . The frequency of CD11bhiLy6Ghi neutrophils was also significantly increased in the spleen of DC-depleted mice compared to control mice ( Figure 7C , R2 and Figure 7D ) . ROS production by these activated transferred CD11bhiLy6Ghi neutrophils was twice as high as ROS production by all transferred neutrophils and higher by CD11bhiLy6Ghi neutrophils transferred into DC-depleted mice than transferred into control mice . To elucidate whether the increased ROS production by neutrophils in DC-depleted mice and upon bacterial infection as well as their enhanced bacterial killing capacity is specific for Ye or a more general host defense mechanism against other bacteria as well , we analyzed the ROS production by neutrophils and the bacterial load in the spleen 2 h post infection with Salmonella typhimurium , Listeria monocytogenes and E . coli . ROS production by neutrophils was significantly increased in DC-depleted mice compared to control mice upon infection with S . typhimurium , but not with L . monocytogenes and E . coli ( Figure 8A ) . This indicates that different bacteria differently affect ROS production by neutrophils . Nevertheless , the bacterial load in the spleen of DC-depleted mice was significantly reduced compared to control mice upon infection with all three bacteria ( Figure 8B ) , indicating that the increased ROS production by neutrophils prior to infection leads to reduced bacterial load of several pathogens in the spleen . Taken together , our data provide evidence that the increased number of phagocytes combined with the enhanced killing capacity of neutrophils upon DC depletion is at least initially beneficial for the host by reducing the bacterial load upon infection .
The innate immune system is important for pathogen clearance . The role of DCs in the adaptive immune response is well established [4] . However , their function in the innate immune response against bacterial infections is not completely defined . In the present study we used a well established mouse infection model with the extracellular Gram-negative bacterium Ye in DC-depleted mice to define the impact of DCs on the innate immune defense against this pathogen . Ye infection of DC-depleted mice revealed a reduced bacterial load in the spleen compared to infected control mice . We found that DC depletion in these mice led to an increase in the number of neutrophils and monocytes in the spleen one day after DT treatment with a peak at day two after daily DT administration , as recently described [25] . In fact , prior to infection of mice with Ye we observed 1 . 2×106 more neutrophils and 7×105 more monocytes in the spleen of DC-depleted mice compared to control mice , demonstrating a quantitative difference in the number of phagocytes ( Table 1 ) . Whether the increased number of phagocytes in the spleen is due to the recruitment of preexisting or ad hoc differentiated phagocytes from the bone marrow remains to be addressed . Ye were more frequently associated with neutrophils ( 2 . 2-fold ) and monocytes ( 6-fold ) from DC-depleted mice , whereas most of the Ye in control mice were extracellular attached to B cells ( Figure 4 and 5 ) . Detailed cellular analysis revealed no qualitative difference in the phagocytosis and processing rates of Ye by neutrophils and monocytes in vivo ( Figure 5 ) . Calculation of the overall number of intracellular Ye in neutrophils and monocytes revealed 2 . 8-fold more intracellular Ye in DC-depleted mice compared to control mice . DC-depletion not only increased the number of neutrophils but also enhanced their production of antimicrobial substances ( ROS ) ( Figure 6 ) . Moreover , neutrophils from DC-depleted mice are more efficient in reducing the bacterial load than neutrophils from control mice , indicating that DC depletion enhances the innate anti-bacterial host defense by modulation of phagocyte homeostasis ( Table 1 ) . We cannot exclude that other effector mechanism than ROS account for the enhanced bacterial killing capacity of neutrophils . This issue could be assessed by analyzing bacterial killing capacity of neutrophils from CD11c . DOG mice on a gp91phox−/− background . The increased numbers of phagocytes upon DC depletion were associated with increased serum levels of G-CSF , CXCL1 , CCL2 , Flt3L , and IL-1α upon DC depletion . G-CSF was shown to induce proliferation of granulocytic precursors and release of mature neutrophils from the bone marrow by downregulation of CXCR4 on their cell surface [32] , [33] . CXCL1 was shown to act in cooperation with G-CSF stimulating neutrophil chemotaxis across the bone marrow endothelium [29] . CCL2 mediates the chemotaxis of CCR2+ monocytes and macrophages [30] . Systemic infection of mice with L . monocytogenes leads to recruitment of CCR2+ monocytes via CCL2 , into the spleen where they differentiate into TNF- and inducible NO synthase ( iNOS ) -producing DCs that are essential for control of the infection [34] . CCR2-mediated recruitment of monocytes was also shown to be essential for defense against Mycobacterium tuberculosis , Toxoplasma gondii , and Cryptococcus neoformans infection [35] . It is tempting to speculate that the recruited monocytes in the spleen upon DC depletion as well as upon Ye infection express CCR2 , due to the increased serum levels of CCL2 . The specific cellular mechanisms by which DC depletion increases the numbers of phagocytes and promotes enhanced neutrophil responses remain to be determined and are currently under investigation . So far we could show that this regulation is cell contact independent ( Figure S6 ) , but mediated by a factor or factors present in the serum upon DC-depletion ( Figure 7A and B ) . Finally , we could show that DC-depletion prior to infection reduced the bacterial load not only in the case of Ye infection , but also in the case of other bacteria as shown for S . typhimurium , L . monocytogenes and E . coli infection ( Figure 8 ) . Increased ROS production was observed upon infection with S . typhimurium , but not with L . monocytogenes and E . coli , suggesting either other defense mechanisms of activated neutrophils , or the increased number of neutrophils combined with their enhanced ROS production upon DC-depletion is sufficient to protect against these pathogens . Hochweller et al . recently described for the first time an increased number of neutrophils and monocytes in spleen following DC depletion [25] . Similarly , a previous study showed that bone marrow chimeras of CD11c . DTR and WT mice ( another mouse model for conditional DC depletion [36] ) develop myeloproliferative disorder ( MPD ) , indicated by massive increase in the number of CD11b+ cells after two weeks of DT treatment every second day [37] . Furthermore , constitutive DC depletion in mice also led to MPD at the age of three months [37] , suggesting a feedback loop regulating appropriate myelogenesis during homeostasis . In both models , elevated serum levels of Flt3L , a critical factor in the control of DC development [38] and maintenance in the periphery [28] , but no changes in M-CSF , GM-CSF , and TNF were observed [37] . Indeed , we also found significantly increased Flt3L in the sera of DC-depleted mice compared to control mice and increased myeloid progenitors in spleen responsive to Flt3L ( Figure 2 and [26] ) , which is likely due to less consumption of Flt3L by DCs in the periphery as mainly immediate DC progenitors and DCs express its receptor Flt3 [39] , [40] . Based on our data we favor the notion that , at least in our model , DCs affect the number of neutrophils and monocytes by modulating the production of growth factors ( G-CSF and Flt3L ) and chemokines ( CXCL1 and CCL2 ) by a yet unknown mechanism . DC depletion led to increased serum levels of IL-1α , that was recently shown to be produced in response to necrosis and stimulates CXCL1 production by non-immune cells , leading to the attraction of neutrophils [41] . We could exclude necrosis as a side effect of the increased number of phagocytes in the spleen upon DC-depletion using mixed bone marrow chimeras ( 80% CD11c . DOG/20% C57BL/6 , Figure S3 ) . These mice have normal DC numbers ( from C57BL/6 bone marrow cells ) after 10 days of DT treatment and still 80% of the DCs are depleted due to DT-treatment . If DC death would cause increase in phagocyte numbers , similar numbers of phagocytes should be seen in mixed bone marrow chimeras and CD11c . DOG mice after 10 days of DT treatment . Yet , this was not the case . Ye infection increased the number of neutrophils in the spleen , and this was more prominent in control mice compared to DC-depleted mice . Additionally , Ye infection increased the levels of G-CSF , CXCL1 , and CCL2 in the sera 7–10 times more in control mice compared to DC-depleted mice , indicating that DCs limit neutrophil numbers in the steady state to prevent tissue damage by these cells but are required for their recruitment upon infection . The latter conclusion is supported by recent findings from a bacterial pyelonephritis model that showed , that kidney DCs secrete CXCL2 upon a second instillation with uropathogenic E . coli leading to the recruitment of neutrophils and bacterial phagocytosis [42] . The simultaneous DC depletion ( CD11c . DTR model ) with E . coli instillation resulted in markedly delayed recruitment of neutrophils to the kidney , due to less CXCL2 secretion and bacterial clearance [42] . Scumpia et al . showed that DCs are essential in the immune response to sepsis as DC depletion ( CD11c . DTR model ) reduced the survival of mice in a cecal ligation and puncture ( CLP ) infection model . Adoptive transfer of BM-DCs improved the survival during this CLP induced polymicrobial sepsis [14] , but no changes - in bacterial load or in serum cytokine levels were observed and the underlying mechanism ( s ) remain unresolved . The differences to our study may be explained by the different mouse model as well as the more severe polymicrobial infection model . Rapid recruitment of neutrophils and abscess formation is required for bacterial clearance [43]–[45] . Recently it became evident that the bacterial load plays a pivotal role in neutrophil survival [46] . Upon Staphylococcus aureus infection the half-life of neutrophils in wound abscesses increased up to 3-fold depending on the inoculum [47] . The increased half-life is presumably mediated by anti-apoptotic signals and cytokines [47]–[49] . In our experimental setting the survival of neutrophils is not influenced by the infection ( data not shown ) . This study demonstrates for the first time that DC depletion not only increased neutrophil numbers in the spleen but also improved production of ROS and Ye killing capacity . In a burn-injured mouse model , pretreatment of the mice with IL-18 increased neutrophil counts and also enhanced neutrophil phagocytosis , ROS production and killing of methicillin-resistant S . aureus [50] . Upon DC depletion no changes in IL-18 serum levels were observed ( data not shown ) , indicating that other factors than IL-18 account for more effective neutrophils in our model . Treatment of burn-injured mice with Flt3L prior to Pseudomonas aeruginosa wound infection enhanced neutrophil chemotaxis , bacterial killing and survival [51] . Furthermore , adoptive transfer of neutrophils from Flt3L-treated mice reduced the bacterial load in the spleen , whereas neutrophils from DC-depleted ( CD11c . DTR model ) and Flt3L-treated mice did not , indicating that Flt3L modifies neutrophil responses via DCs in this model [51] . However , the cellular mechanism remains elusive . In our study , adoptive transfer of neutrophils from DC-depleted mice into control mice prior to Ye infection reduced the bacterial load in the spleen to the level of DC-depleted mice , whereas adoptive transfer of neutrophils from control mice did not , arguing against altered neutrophil response mediated via Flt3L modified DCs . Thus , neutrophils from DC-depleted mice with enhanced anti-bacterial activity account for this effect . Our results are supported by the finding that enhanced local recruitment of neutrophils in peritonitis-induced sepsis improves bacterial clearance and survival [52] . In conclusion , we provide evidence that DCs differently regulate splenic phagocyte numbers in the steady state or upon bacterial infection . Furthermore , the newly recruited neutrophils upon DC depletion display an improved bacterial killing capacity , thereby accounting for the decreased bacterial load and likely increased survival of these mice upon Ye infection . Beyond the anti-bacterial host defense , these studies point towards a complex interaction between DCs and phagocyte homeostasis by serum factors .
Ethics statement: Animal experiments were performed in strict accordance with the German regulations of the Society for Laboratory Animal Science ( GV-SOLAS ) and the European Health Law of the Federation of Laboratory Animal Science Associations ( FELASA ) . The protocol was approved by the Regierungspräsidium Tübingen ( Permit Numbers: IM5/08 , IZ2/11 ) . All efforts were made to minimize suffering . Female C57BL/6JolaHsd mice were purchased from Janvier ( St Berthevin Cedex , France ) and Harlan Winkelmann ( Borchen , Germany ) . CD11c . DOGxC57BL/6 mice [25] were bred under specific pathogen-free conditions in the animal facilities of the University of Tübingen . Mice used for experiments were between 6–9 weeks of age and were provided food and water ad libitum . Mice were infected with the indicated amount of Ye WA-314 ( serotype 0∶8 ) , WA-314 expressing GFP [53] , Salmonella enterica serovar typhimurium SL1344 , Escherichia coli JM83 or Listeria monocytogenes ATCC 43251 in 200 µl PBS into the tail vein . As a control , mice were injected only with 200 µl PBS . The bacterial load in the spleen was obtained after plating serial dilutions of the cell suspensions on Müller-Hinton or Luria Bertani agar plates . For systemic DC depletion BAC transgenic CD11c . DOG mice , that express the human diphtheria toxin receptor under control of the CD11c promoter , were injected intraperitoneally or intravenously with 8 ng/g bodyweight of diphtheria toxin ( Sigma ) in PBS one day prior to Ye infection and daily during infection . Spleens were cut into small pieces and digested for 30 min at 37°C in 2 ml RPMI 1640+2% FBS medium containing collagenase ( 1 mg/ml; type IV; Sigma-Aldrich ) and DNase I ( 100 µg/ml , Roche ) . To disrupt DC-T cell complexes , EDTA ( 0 . 1 ml , 0 . 1 M ( pH 7 . 2 ) ) was added and mixing continued for 5 min . Single cell suspensions were made by pipetting the digested organs . Undigested fibrous material was removed by filtration and erythrocytes were lysed with lysis buffer ( 150 mM NH4Cl , 10 mM KHCO3 , 2 mM NaEDTA ) . The total number of cells was determined by trypan blue exclusion . FACS buffer ( ( PBS containing 1% FBS ( Sigma-Aldrich ) and 0 . 09% NaN3 ( Sigma-Aldrich ) was used for all incubations and washing steps . Before staining , cells were incubated for 15 min at 4°C with hybridoma supernatant from 2 . 4G2 cell line producing anti-FcgRII/III mAb . Cells were stained with anti-CD11c-APC ( N418 Miltenyi Biotec ) , CD11c-PE ( N418 , eBiosciences ) , CD8α-PE-Cy7 ( 53-6 . 7 , eBiosciences ) , CD4-eFluor450 ( RM4-5 , eBiosciences ) , CD19-APC ( 6D5 , Miltenyi Biotec ) , MHC II-PerCP ( M5/114 . 15 . 2 , Biolegend ) , Ly6C-PE-Cy7 ( HK1 . 4 , Biolegend ) , Ly6G-FITC ( 1A8 , Biolegend ) , Ly6G-PE ( 1A8 , Miltenyi Biotec ) , Gr-1-eFluor450 ( RB6-8C5 , eBiosciences ) , CD11b-APC-eFluor780 ( M1/70 , eBiosciences ) , CD45 . 1-eFluor450 ( A20 , eBiosciences ) , CD62L-APC ( MEL14-H2 . 100 , Miltenyi Biotec ) for 20 min at 4°C . To exclude dead cells , 7-aminoactinomycin D ( 7-AAD; Sigma-Aldrich ) or aqua life dead ( Invitrogen ) was used . Samples were acquired for 6 to 8-colour analysis using a Canto-II flow cytometer ( BD Biosciences ) with DIVA software ( BD Biosciences ) and further analyzed using FlowJo 7 . 5 software ( TreeStar Inc ) . A total of 500 , 000–1 , 200 , 000 cells were acquired . For the analysis of viable intracellular Ye in splenic phagocytes after intravenous infection of the mice , CD19-expressing cells were depleted from splenic single cell suspensions by MACS technology using CD19 magnetic beads ( Miltenyi Biotec ) following the manufacturer's protocol . Fc block was performed and cells were stained with Gr-1-FITC ( RB6-8C5 , BD Biosciences ) , CD11b-PE ( M1/70 , BD Biosciences ) , and CD11c-APC ( N418 , Miltenyi Biotec ) in PBS . DCs , neutrophils , and CD11b+CD11c−Gr-1− cells were sorted on a FACS Aria cell sorter ( BD Biosciences ) and reanalyzed on a Canto-II flow cytometer . Cells were treated afterwards with gentamicin ( 100 µg/ml , Sigma Aldrich ) for 30 min at 37°C to kill extracellular bacteria . Cells were then lysed with PBS containing 0 . 1% tergitol TMN 10 ( Sigma-Aldrich ) and 0 . 1% bovine serum albumin ( Merck ) and the bacterial load in the spleen was obtained after plating serial dilutions of the suspensions on Müller-Hinton agar plates . Mice were treated with DT one day prior to infection with 5×108 WA-314 expressing eGFP . After 30 min the spleen was removed and the splenocytes were stained with Ly6C-Pacific blue ( HK1 . 4 , Biolegend ) and CD11b-APC ( M1/70 , Biolegend ) or CD45R ( B220 ) -Vioblue ( RA3-6B2 , Miltenyi Biotec ) and CD19-APC ( 6D5 , Beckmann Coulter ) , fixed with 1% paraformaldehyde , permeabilized with 0 . 1% saponin ( Sigma- Aldrich ) and 0 , 5% BSA ( Sigma-Aldrich ) in PBS and stained with CD107a-PE ( 1D4B , Biolegend ) . Images of up to 8 , 000 Ye-positive events were then acquired with multispectral imaging flow cytometry ( MIFC ) using an ImageStream equipped with a custom designed 40× objective ( 0 . 75 NA ) ( Amnis corp . , Seattle , USA ) [54] , [55] . Image data were analyzed with IDEAS 3 . 0 ( Amnis corp . ) , which allows an objective and unbiased analysis of thousands of images per sample on the single cell level . To quantify bacterial uptake regions of interest ( mask ) were defined for each cell . The first mask covered any fluorescence of the event , independent whether it originated from Ye or from the cells ( total event mask ) . Then a second mask that includes the cytoplasm and nuclei and excludes the plasma membrane was defined ( cytoplasm mask ) . To create this cytoplasm mask , we first created a filled mask based on the lineage markers ( e . g . CD19 or Ly6C ) that covers the entire cell and excludes the lineage negative Ye on the top of the cell . To make this mask more stringent , it was then eroded by one pixel , i . e . 500 nm . The resulting mask excludes the plasma membrane and specifies the cell interior only . Thereafter , the internalization score was calculated , which is a rescaled ratio of the Ye-GFP intensity in the cytoplasmic mask and in the total event mask . Thus , the higher the score the more Ye were internalized . We counted a cytoplasmic localization of Ye if the internalization score was >2 . Thereafter , the subcellular localization of intracellular Ye-GFP was evaluated by calculating the colocalization of Ye-GFP and CD107a using a rescaled Pearson's correlation coefficient , named similarity score [56] . Ye show a high degree of colocalization , if the similarity score is >1 . Splenocytes were incubated , where indicated , with serum diluted in RPMI +10% FBS ( 1∶5 ) for one hour at 37°C . 2×106 splenocytes were incubated with 2×106 Ye for 10 min at 37°C and afterwards serial dilutions were plated on Müller-Hinton agar plates . For kinetic studies ( Figure 6A ) cells were incubated as described , washed , and incubated further for the indicated time points in RPMI +10% FBS in the presence of gentamicin ( 100 µg/ml ) . Mice were treated with DT over night and were infected with 5×104 Ye . Mice were sacrificed after 2 h and the spleen was aseptically removed . Spleen cell suspensions were obtained and flow cytometry staining was performed as described above . 3×106 cells were incubated for 20 min at 37°C with 2′ , 7′-Dichlorofluorescin diacetate reagent ( DCFD , Sigma-Aldrich ) , washed and analyzed by flow cytometry . For the adoptive transfer of neutrophils CD11c . DOG mice or C57BL/6 mice were treated with DT over night ( Figure 6D ) . Splenocytes were obtained from these or from CD45 . 1 C57BL/6 mice and either B cells depleted using anti-CD19 beads or Ly6G+ cells enriched using anti-Ly6G beads and MACS technology as described above . Figure 6D: Flow cytometry staining was performed as follows: Gr-1-FITC , CD11b-APC-Alexa780 , and Ly6C-PE-Cy7 . Neutrophils were sorted on a FACS Aria cell sorter ( BD Biosciences ) . 1 . 2 to 1 . 8×106 neutrophils were adoptively transferred into each C57BL/6 mouse and infected with 5×104 Ye 30 min later . One day post infection CFU per spleen was analyzed by serial dilution . Figure 7C: 4×106 CD45 . 1+ neutrophils were adoptively transferred into each C57BL/6 or DC-depleted CD11c . DOG mouse . 2 h after transfer ROS production by CD45 . 1+Ly6G+ cells was analyzed as described above . Cytokines from sera were analyzed . IFN-γ and FLT3L were measured by ELISA ( eBiosciences and R&D , respectively ) . CXCL1 , CCL2 , G-CSF , IL-6 , and IL12-p40 were measured by multiplex bead technique ( Biorad ) according to the manufacturer's protocol . Tissues were embedded in Tissue-Tek OCT compound ( Sakura ) , frozen in liquid nitrogen and stored at −80°C . 5 µm cryostat sections were prepared and dried over night . Slides were fixed for 10 min with ice cold acetone , dried for 1 h and were then rehydrated for 15 min with PBS containing 0 . 25% bovine serum albumin . After blocking with Fc-block ( hybridoma supernatant from 2 . 4G2 cell line producing anti-FcγRII/III mAb ) in PBS containing 10% fetal bovine serum and 5% normal goat serum ( Sigma-Aldrich ) slides were incubated with polyclonal rabbit anti-yersinia antibodies ( WA-v; 5 µg/ml ) [57] in PBS containing 10% fetal bovine serum and 5% normal goat serum for 30 min at room temperature . Alexa Fluor 488-labeled goat-anti-rabbit IgG F ( ab' ) fragment ( 1 µg/ml , Invitrogen ) or DyLight 649-labeled goat-anti-rabbit IgG F ( ab' ) 2 fragment ( 5 µg/ml , Jackson ) were used as secondary antibodies . Slides were blocked with a biotin blocking kit ( Vector ) and then stained with biotin-conjugated anti-CD11c antibody ( HL3 , 5 µg/ml , BD Biosciences ) or biotin-conjugated anti-Gr-1 antibody ( RB6-8C5 , 5 µg/ml; eBiosciences ) in PBS-10% FBS , washed and then incubated with streptavidin-Alexa Fluor 546 ( 1 µg/ml in PBS-10% FBS , Invitrogen ) for 30 min at room temperature . For triple staining slides were further incubated with FITC-labeled rat-anti-mouse Ly6G antibody ( 1A8 , 5 µg/ml; Biolegend ) . Nuclei were stained with DAPI ( 10 µg/ml , Sigma-Aldrich ) . Slides were mounted in Mowiol ( Carl Roth ) . Labeled cells were visualized with a DMRE fluorescence microscope ( Leica ) or an Axiovert 200 M fluorescence microscope ( Zeiss ) . Image processing was performed with Adobe Photoshop ( Version 8 . 0 . 1 ) . Mixed BM chimera mice were generated as previously described [26] by transferring 2×106 Thy1 . 2-depleted donor BM cells into 10 Gy-irradiated recipient B6 mice . Donor BM consisted of a mixture of cells from CD11c . DOG CD45 . 1 mice ( DTR+ ) and eGFP mice ( DTR− ) at a ratio of 80∶20 . Experiments were started 8 to 10 weeks after reconstitution . Statistical analysis was performed using the GraphPad Prism 5 . 0 software ( GraphPad , San Diego , CA ) . Diagrams show mean values ± SD . Statistical analysis was performed using the unpaired two-tailed Student's t test . Statistical analysis of survival was performed by using the log-rank test . Data from cytokine production and adoptive transfer experiments were analyzed using one-way ANOVA with Bonferrfoni post test . When data were not normally distributed , a logarithmic transformation was applied prior to the analyses . The differences were considered as statistically significant if p<0 . 05 ( * ) , p<0 . 01 ( ** ) , p<0 . 005 ( *** ) or p<0 . 001 ( **** ) . | Dendritic cells ( DCs ) are professional antigen-presenting cells playing a crucial role in the initiation of T-cell responses to combat infection . DCs adapt their immune response according to the type of pathogen . For example , in response to intracellular bacteria , DCs produce IL-12 , thereby initiating Th1 polarization , whereas in response to extracellular parasites or extracellular bacteria , DCs instruct Th2 or Th17 polarization , respectively . Nevertheless , their role in innate immunity is less well understood . To address this , we studied the role of DCs upon infection with the Gram-negative enteropathogenic bacteria Yersinia enterocolitica ( Ye ) and used a mouse model to deplete DCs . We found that DCs have an unexpected role during severe infection as depletion of these cells resulted in better outcome of infection as well as less bacterial load . We also found that DC depletion increased the number of phagocytes with improved anti-bacterial capacity in the spleen . Our study provides new insights into the role of DCs in innate immune response upon bacterial infection and points towards a complex interaction between DCs and phagocyte homeostasis . DC alteration during infection might also be an interesting target for immunotherapy in the future to guide the outcome of infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"immune",
"cells",
"immunity",
"immunology",
"biology",
"microbiology",
"bacterial",
"pathogens",
"immune",
"system"
] | 2012 | Depletion of Dendritic Cells Enhances Innate Anti-Bacterial Host Defense through Modulation of Phagocyte Homeostasis |
Lentiviruses such as HIV-1 traverse nuclear pore complexes ( NPC ) and infect terminally differentiated non-dividing cells , but how they do this is unclear . The cytoplasmic NPC protein Nup358/RanBP2 was identified as an HIV-1 co-factor in previous studies . Here we report that HIV-1 capsid ( CA ) binds directly to the cyclophilin domain of Nup358/RanBP2 . Fusion of the Nup358/RanBP2 cyclophilin ( Cyp ) domain to the tripartite motif of TRIM5 created a novel inhibitor of HIV-1 replication , consistent with an interaction in vivo . In contrast to CypA binding to HIV-1 CA , Nup358 binding is insensitive to inhibition with cyclosporine , allowing contributions from CypA and Nup358 to be distinguished . Inhibition of CypA reduced dependence on Nup358 and the nuclear basket protein Nup153 , suggesting that CypA regulates the choice of the nuclear import machinery that is engaged by the virus . HIV-1 cyclophilin-binding mutants CA G89V and P90A favored integration in genomic regions with a higher density of transcription units and associated features than wild type virus . Integration preference of wild type virus in the presence of cyclosporine was similarly altered to regions of higher transcription density . In contrast , HIV-1 CA alterations in another patch on the capsid surface that render the virus less sensitive to Nup358 or TRN-SR2 depletion ( CA N74D , N57A ) resulted in integration in genomic regions sparse in transcription units . Both groups of CA mutants are impaired in replication in HeLa cells and human monocyte derived macrophages . Our findings link HIV-1 engagement of cyclophilins with both integration targeting and replication efficiency and provide insight into the conservation of viral cyclophilin recruitment .
The ability to infect terminally differentiated cells of the monocyte-macrophage lineage is a conserved property of lentiviruses , including HIV-1 [1] . This process requires pre-integration complexes ( PICs ) to traverse the nuclear pore , though the molecular mechanism remains unclear . The HIV-1 proteins matrix , Vpr and integrase , as well as a DNA triplex at the central polypurine tract , have been proposed to contribute , but contrary evidence has been presented for each [2]–[5] . Gammaretroviruses such as murine leukemia virus ( MLV ) are dependent on cell division for infectivity and infect non-dividing cells inefficiently [6] . Characterization of HIV-1/MLV chimeric viruses has suggested a role for the HIV-1 capsid ( CA ) in nuclear entry [7] . Furthermore , certain HIV-1 CA mutants are selectively defective in arrested cells but not in actively dividing cells again implicating a role for CA in HIV-1 nuclear entry [8]–[10] . The nuclear pore complex ( NPC ) , through which HIV replication intermediates must pass , consists of multiple copies of at least 30 different nuclear pore proteins ( Nups ) . Nup358 is a large 358 kDa protein that constitutes the cytoplasmic filaments and has a C-terminal cyclophilin ( Cyp ) domain . It was first named Nup358 [11] but has also been called RanBP2 [12] . We use its original name Nup358 throughout this study . Several roles have been proposed for Nup358 involving cell cycle control , nuclear export , and transportin/importin dependent nuclear import ( reviewed in [13] ) . In addition , Nup358 is a co-factor for HIV-1 replication , supporting nuclear entry of viral PICs and influencing target site preference for integration [14]–[17] . It has been unknown how the virus engages Nup358 and influences PIC traffic across the nuclear pore . Here we demonstrate that HIV-1 CA binds directly to the Nup358 Cyp domain ( Nup358Cyp ) with an affinity within three fold of its binding of the monomeric cytoplasmic cyclophilin , CypA , which is known to be important during HIV-1 infection . We also demonstrate that CypA is important for directing HIV-1 into a nuclear entry pathway involving Nup358 and subsequent engagement of the nuclear basket protein Nup153 , ensuring integration into preferred genomic loci . We report that altering CA interactions with Nup358 or CypA results in alterations in integration targeting preference , and reduced replication in macrophages . Our study provides the first evidence for direct interaction between HIV-1 CA and the NPC and suggests possible models for links between nuclear import , integration site selection and effective replication in primary human cells .
Several studies have shown that depletion of Nup358 reduces HIV-1 infectivity . We sought to define the HIV-1 determinant that confers its sensitivity to Nup358 depletion by studying infections with VSV-G pseudotyped viral vectors encoding GFP . Stable Nup358 depletion by transduction of HeLa cells with MLV or HIV-1 based shRNA expression vectors reduced HIV-1 GFP vector infectivity by 6- to 8-fold confirming Nup358's role as an HIV-1 cofactor [14] , [17] ( Figure 1A , B and Figure S1 ) . We validated effective shRNA targeting by western blotting , using a Nup358 specific antibody ( Figure 1B ) , as well as by co-transfecting the shRNA expression vector and a plasmid encoding GFP-tagged Nup358 into 293T cells ( Figure S2 ) . Studies on the role of Nup358 in HIV-1 replication have used M-group HIV-1 isolates [14]–[17] . In order to confirm the importance of Nup358 as a cofactor for other HIV-1 isolates we also tested the O-group HIV-1 virus MVP5180 [18] , [19] as a distantly related HIV-1 and found that this too was sensitive to Nup358 depletion , suggesting that Nup358 use is a conserved feature of HIV-1 biology ( Figure 1A and Figure S1C ) . We next tested whether the even more distantly related simian immunodeficiency virus from macaques ( SIVmac ) was sensitive to Nup358 depletion . In contrast to HIV-1 , infectivity of SIVmac , was not reduced by Nup358 RNAi , suggesting species-specificity of Nup358 use ( Figure 1A and Figure S1C ) . We next sought to identify the viral determinant for Nup358 RNAi sensitivity . Given that the HIV-1 capsid protein ( CA ) has been implicated in HIV-1 nuclear import [7] , [8] , [10] , we tested whether the different sensitivities to Nup358 depletion between HIV-1 and SIVmac could be accounted by their different CA proteins . We exchanged CA coding regions between HIV-1 and SIVmac and analyzed infectivities of chimeric viruses on Nup358 depleted cells . Replacement of SIVmac CA with CA from HIV-1 [20] rendered the chimeric SIVmac sensitive to Nup358 depletion , while replacement of HIV-1 CA with SIVmac CA [21] rendered HIV-1 largely insensitive to Nup358 depletion ( Figure 1C ) . For comparison , we examined the sensitivity of these viruses to transportin 3 ( TRN-SR2 ) depletion , and confirmed that TRN-SR2 specific shRNA reduced infectivity of both HIV-1 ( ∼8 to 10-fold ) and SIVmac ( ∼20-fold ) ( Figure 1A and Figure S1 ) . MLV GFP vector infectivity was not affected by depletion of these proteins as reported previously , consistent with MLV's inability to traverse the nuclear pore and infect non-dividing cells ( Figure 1A and Figure S1 ) . If Nup358 and TRN-SR2 facilitate nuclear entry of wild type HIV-1 , then their depletion should inhibit HIV-1 infection at the level of nuclear import . We confirmed that 2-LTR circle products of HIV-1 were modestly reduced in abundance in the Nup358 or TRN-SR2 depleted cells , whereas late reverse transcript production was unaffected ( Figure 1D ) [22] . However , we observed that the ten-fold reduction in infectivity was greater than the two to four-fold reduction in 2-LTR circles , possibly explained by an integration defect increasing the amount of 2-LTR circles . To measure integration we infected Nup358 or TRN-SR2 depleted cells with HIV-1 GFP vector , grew the cells for 2 weeks and measured the number of integrated proviruses by Taqman qPCR . We observed that the reduction of integrated proviruses in Nup358 depleted cells ( 5-fold ) was similar to the reduction of 2-LTR circles ( 4-fold ) ( Figure 1D and Figure S3 ) . In contrast , the reduction of proviruses in TRN-SR2 depleted cells was significantly greater ( 50-fold ) , than the reduction in 2-LTR circles ( 2 to 3-fold ) . This observations may suggest that Nup358 depletion blocks HIV-1 at a step prior to nuclear import but after reverse transcription , whereas TRN-SR2 depletion imposes two blocks , one at the stage of nuclear import ( reduction of 2-LTR circles ) and a second at integration . However , we suggest caution in interpretation of 2-LTR circle assay as a measure of nuclear entry given that 2-LTR circles are non productive for infection and their formation may have different co-factor requirements . Importantly , replication of wild type NL4 . 3GFP-IRES was also impaired in Nup358 or TRN-SR2 depleted HeLa cells expressing CD4 ( Figure 1E ) . Equivalent CD4 expression in these cells was confirmed by flow cytometry using fluorescent CD4 specific antibody ( Figure S4 ) . These data suggest that Nup358 and TRN-SR2 contribute to optimal viral nuclear entry , integration and eventually replication . Nup358 contains a cyclophilin domain ( Nup358Cyp ) at its extreme carboxyl-terminus . The HIV-1 N-terminal CA domain ( CANTD ) resides on the surface of the virion core and recruits CypA to viral cores [23] , [24] . The CA-dependent sensitivity of HIV-1 to Nup358 depletion led us to hypothesize that HIV-1 CANTD might also interact with Nup358Cyp in a similar manner to its interaction with CypA . To test this , we purified recombinant CypA and Nup358Cyp and measured binding to recombinant CANTD , using isothermal titration calorimetry ( ITC ) [25] . We found that the HIV-1 CANTD bound Nup358Cyp with a Kd of 16 µM , in a similar range to its Kd of 7 µM for CypA ( Figure 1F ) [25] . Surprisingly , the CypA inhibitor cyclosporine ( Cs ) did not prevent Nup358Cyp binding to HIV-1 CANTD whereas it did inhibit CypA binding . Whilst capsid interaction with both Nup358Cyp and CypA was entropically favourable , interaction with Nup358Cyp was more strongly entropically favourable than CypA . This does not markedly alter the affinity with respect to CypA as CANTD interaction with CypA is more enthalpically favourable than with Nup358Cyp . The different thermodynamic signatures between CypA and Nup358Cyp suggest that the two proteins do not form identical interactions . The entropic component of any interaction is the sum of changes in protein and solvent dynamics . Given that the ligand , CANTD , is the same in each experiment whilst Nup358Cyp and CypA comprise a single globular fold it is likely that the entropically favourable nature of both interactions is a consequence of releasing ordered water molecules upon complex formation . The larger entropic change associated with Nup358Cyp interaction may indicate a greater release of ordered water upon complexation , suggesting that the interface is larger than in CypA:capsid . SIVmac CANTD did not bind Nup358Cyp ( Figure 1F ) , and bound CypA with a very low affinity ( ∼800 µM ) ( Figure 1F ) , which becomes important below . The inability of Nup358Cyp to bind to SIVmac CANTD correlates with the insensitivity of SIVmac to Nup358 depletion in HeLa cells ( Figure 1A ) . To probe Nup358Cyp binding further we designed an HIV-1 inhibitor based on the simian restriction factor TRIMCyp . Owl monkey TRIMCyp blocks HIV-1 by binding incoming capsids via its CypA domain [26] . We replaced the TRIMCyp cyclophilin domain with human CypA or Nup358Cyp to make TRIMCypA and TRIMNup358 . We found that both TRIMNup358 and TRIMCypA blocked HIV-1 infectivity ( Figure 1G ) whereas SIVmac infection was not restricted , as expected from the lack of binding ( Figure S5 ) . Importantly , Cs treatment only rescued infectivity from TRIMCypA but not from TRIMNup358 , corroborating Cs sensitivity measured by ITC ( Figure 1F ) . We confirmed similar expression levels of chimeric proteins by western blot ( Figure 1G ) . These data are consistent with HIV-1 , but not SIVmac , CANTD efficiently binding Nup358Cyp in the context of TRIMNup358 in the cytoplasm of infected cells . This suggests that HIV-1 PICs containing CA or possibly entire capsid cores can interact directly with a component of the NPC , providing insight into how HIV-1 contacts the nuclear pore during the process of nuclear entry . Host proteins that interact with pathogens are often under positive selective pressure [27] . A higher rate of non-synonymous nucleotide substitutions ( dN ) than synonymous substitutions ( dS ) at a particular codon in an interspecies comparison provides evidence for such positive selection . We aligned Nup358Cyp DNA sequences from 12 different species ( Figure S6 ) and performed an analysis of codon-specific selective pressures using the program Random Effect Likelihood ( REL ) implemented on the online version of the HyPhy package [28] . Despite overall strong negative selection across the Cyp domain , we found Nup358Cyp codon 61 to be positively selected at a statistically significant level ( Bayes factor >50; Figure 2A ) . Indeed , residue 61 is extremely conserved as methionine across the whole vertebrate Cyp family except in Nup358Cyp . ( Figure 2B , C ) . In the case of Nup358Cyp lower vertebrates encode the ancestral methionine , whereas higher vertebrates encode valine , leucine or isoleucine at this position . Fixation of the positively selected site appears to have occurred after the divergence of fish and tetrapods , since Nup358Cyp sequences from fish ( e . g . Danio rerio ) retain methionine at position 61 ( Figure 2B , C ) . This observation suggests that Nup358 has been under selective pressure to evolve and that this has led to variation in the sequence at this position . The cyclophilin domain of Nup358 has been proposed to possess prolyl cis-trans isomerase activity , similarly to CypA [29] , [30] . Assuming that the Nup358Cyp active site is homologous to that of CypA then according to the CypA structure ( PDB:1FGL ) residue 61 is located directly at the bottom of the active site suggesting that it might impact on substrate specificity ( Figure 2D ) . To examine this further we made a TRIMNup358 mutant in which the valine at Nup358Cyp position 61 was changed to the ancestral residue methionine . TRIMNup358 V61M was no longer able to restrict HIV-1 , suggesting that binding to HIV-1 CA is influenced by this residue ( Figure 2E ) . Our results suggest that during evolution selective pressure , possibly from ancient pathogenic viruses , has driven the change of Nup358Cyp position 61 , altering substrate specificity in higher vertebrates . In turn HIV-1 is adapted to use this modified protein for nuclear entry in humans . If binding of HIV-1 CA to Nup358 is important for HIV-1 infectivity , then amino acid substitutions in CA that affect interaction with Nup358 should influence infectivity . Indeed , we found that whilst wild type HIV-1 infectivity is sensitive to both Nup358 as well as TRN-SR2 depletion , certain HIV-1 CA mutants were not , suggesting an inability to utilize these cofactors . We infected HeLa cells stably expressing Nup358 or TRN-SR2 shRNA with GFP-encoding VSV-G pseudotyped HIV-1 vectors bearing wild type or mutant CA . We found that the cyclophilin-binding mutants G89V and P90A were insensitive to Nup358 depletion but remained sensitive to TRN-SR2 depletion ( Figure 3A ) . We hypothesized that an inability to bind Nup358Cyp or CypA might underlie infectivity defects of other HIV-1 CA mutants . The HIV-1 CA mutant N57A is more severely defective in arrested cells than dividing cells ( Figure 3A ) [10] , suggesting that this residue may have a role in nuclear entry . Indeed , ITC demonstrates that N57A is impaired in binding Nup358Cyp ( Kd 55 µM ) but not CypA ( Kd 7 µM ) ( Figure 3B ) . As , N57A is less sensitive to both Nup358 and TRN-SR2 depletion ( Figure 3A ) , we hypothesize that its infectivity defect is caused by an inability to engage these proteins . We found that N57A was still restricted by TRIMNup358 ( Figure S5 ) , suggesting that increased avidity through Nup358Cyp dimerization in the context of TRIMNup358 may overcome the reduced affinity to monomeric Nup358Cyp . Importantly , N57A's insensitivity to Nup358 depletion suggests that it does not engage Nup358 during nuclear entry . Finally , we studied the HIV-1 CA mutant N74D , which is reported to be less sensitive to Nup358 or TRN-SR2 depletion ( Figure 3A ) [17] , [31] . Like N57A , N74D bound monomeric Nup358Cyp in ITC experiments with significantly lower affinity than wild type ( Kd 95 µM ) ( Figure 3B ) and like N57A , N74D was also restricted by TRIMNup358 ( Figure S5A ) . As for N57A , it seems contradictory that HIV-1 CA mutants that are less sensitive to Nup358 depletion are restricted by TRIMNup358 . We assume that binding characteristics of TRIMNup358 , and Nup358 itself , to CA are different particularly given that TRIM5 is reported to form cytoplasmic dimers [32] and higher-order multimers [33] , Thus a forced dimerization of Nup358Cyp by fusing it to TRIM5α , could increase binding of Nup358Cyp to CA by increasing avidity , thereby allowing restriction . In addition , it is possible that the decreased binding of N57A as well as N74D to TRIMNup358 is disguised by an increased sensitivity to restriction by this TRIM5 chimera . HIV-1 CA N57 is located at the base of helix 3 and N74 in helix 4 ( Figure 4C , D ) , suggesting that amino acid residues outside the Cyp-binding loop can impact on Nup358Cyp binding . We propose that Nup358 and TRN-SR2 define an import pathway used by wild type HIV-1 and that CA amino acid substitutions direct the virus to use Nup358 independent ( G89V , or P90A ) , or Nup358/TRN-SR2 independent ( N74D , or N57A ) import pathways . To test whether HIV-1 dependence on Nup358 is increased in non-dividing cells , we arrested HeLa cells with aphidicolin and measured infectivity of the CA mutant viruses . We found that only N57A and MLV infectivities were inhibited by aphidicolin treatment ( Figure 3A and Table S1 ) whereas mutants G89V , P90A and N74D were not affected . This suggests that G89V , P90A and N74D use Nup358/TRN-SR2 independent routes into the nucleus even in the absence of cell division . This hypothesis is supported by the observation that neither the wild type virus nor these mutants become additionally sensitive to Nup358 or TRN-SR2 RNAi in aphidicolin-arrested cells ( Figure 3A and Table S1 ) . On the other hand N57A is slightly increased in its sensitivity to aphidicolin particularly after Nup358 depletion . We conclude that these co-factors are required for HIV-1 infection of dividing and non-dividing cells . HIV integration is favored in chromosomal regions rich in genes and associated features such as CpG islands , DNAaseI hypersensitive sites , and high G/C content . We have shown that Nup358 or TRN-SR2 depletion reduces HIV-1 integration frequency near these features [16] . To test this for the CA mutants studied above , we sequenced 19 , 546 unique integration sites from HIV-1 and its mutants by 454/Roche pyrosequencing and compared their chromosomal distributions as described [16] , [34]–[37] . Although the HIV-1 CA mutants retained the preference for integration within transcription units , their integration site distributions diverged from wild type HIV-1 . The patterns clustered into two groups that map to two distinct areas on the CA surface ( Figure 4 ) . HIV-1 CA mutants N57A and N74D integrated into regions of chromatin associated with a significantly lower density of transcription units and associated features . For wild type HIV-1 this density was 15 transcription units/MB , whereas for CA mutants N57A or N74D the density was reduced to what is expected for random integration ( 7–9 transcription units/MB ) ( Figure 4A and Figure S7 ) . In contrast , the two Cyp-binding mutants , G89V and P90A , exhibited an opposite phenotype , with favored integration into regions of increased density of transcription units ( ∼20 transcription units/MB ) . The chimeric HIV-1 containing SIVmac CA , showed a further increased preference for regions dense in transcription units ( 25 transcription units/MB ) ( Figure 4A and Figure S7A ) . These latter three viruses similarly showed increased frequency of integration in areas rich in active genes , CpG islands , DNase sites , and high in GC content , features correlating with high gene density . Hierarchical clustering of the CA mutants based on these data separated the viruses into two groups: N57A and N74D , and the Cyp-binding mutants G89V , P90A and chimeric HIV-1 ( SIVCA ) ( Figure 4B ) . Wild type HIV-1 , which has an intermediate targeting phenotype , was an outlier within this second group . Thus amino acid substitutions in CA can alter integration targeting preference , resulting in either of two phenotypes . Because G89V and P90A influence targeting in the same direction , we infer that disruption of normal CypA interactions , and possibly Nup358 interactions , result in increased frequency of integration in regions with high densities of transcription units . The N74D and N57A substitutions are less sensitive to depletion of both Nup358 and TRN-SR2 , and N74D gains sensitivity to depletion of other nuclear pore proteins [17] . We thus infer that this pathway leads to favored integration in regions with lower densities of transcription units . We were surprised that HIV-1 CA mutants P90A and N74D , which are less sensitive to depletion of TRN-SR2 and/or Nup358 ( Figure 3A ) , and have different integration site preferences in unmodified cells ( Figure 4 ) , were as infectious as wild type virus in single round assays . This is true when the virus is pseudotyped with the VSV-G envelope ( Figure 3A ) or the natural HIV-1 gp160 envelope ( Figure S9 ) . To test whether these CA substitutions affect HIV-1 replication we compared replication of wild type HIV-1 NL4 . 3 ( Ba-L Env ) with CA mutants P90A and N74D in spreading infection in HeLa TZM-bl cells [38] . Interestingly , we found that replication of both HIV-1 CA mutants was impaired in these cells compared to wild type virus , suggesting that cofactors used for nuclear entry and/or integration site selection are important for optimal replication ( Figure 4E ) . We also found that HIV-1 NL4 . 3 ( Ba-L Env ) bearing CA alterations N74D or P90A replicated poorly in primary human MDM from four independent donors , whereas wild type virus replicated efficiently ( Figure 4F and Figure S7B ) . These data demonstrate that HIV-1 CA mutants P90A and N74D do not support optimal replication . One possible explanation is that this is due to differences in their integration site targeting as compared to the wild type virus , though other models are possible . Whether the defect in replication is due to a defect in viral gene expression remains unclear . However , it is clear that the mutant viruses that are unable to effectively utilize Nup358 or TRN-SR2 display a replication defect in a cell line and in primary human macrophages . The observation that HIV-1 CA mutants P90A and G89V , as well as chimeric HIV-1 ( SIVCA ) integrate into genome regions with higher densities of transcription units and associated features raised the possibility that integration targeting might be influenced by CypA binding to CA . Since cyclosporine ( Cs ) selectively inhibits CypA but not Nup358Cyp binding ( Figure 1F , G ) , we investigated whether Cs could retarget integration by HIV-1 . In fact , Cs treatment retargeted viral integration preferences in a way that phenocopied the CA G89V/P90A substitutions shifting integration preferences into regions of higher gene density ( Figure 5A , B ) . Thus preventing CypA-CA interactions with Cs has the same effect on integration targeting as amino acid substitutions in HIV-1 CA that block CypA binding , supporting the idea that integration targeting is truly affected by cyclophilin-CA interactions . Reduction of Nup358 by RNAi led to integration into low gene density/activity regions [16] but preventing CypA binding by CA amino acid substitutions ( G89V/P90A ) or Cs treatment shifted virus integration preferences into high gene density/activity regions . This suggested to us that Nup358 and CypA have different , possibly opposing effects on HIV-1 . Alternately , Cs treatment may somehow change the availability of Nup358 in the cell . To investigate this further we tested whether CypA inhibition in Nup358 depleted cells influences HIV-1 infectivity . Remarkably , Cs treatment specifically rescued HIV-1 infectivity reduced by Nup358 depletion to the level observed in control cells ( Figure 6A and Figure S8 ) . We note that the small inhibitory effect of Cs on HIV-1 infectivity is preserved and infectivity is rescued to the level of infectivity on control cells treated with Cs . Thus Cs inhibits HIV-1 GFP infectivity by 2–3 fold but concomitantly rescues infectivity from the effects of Nup358 depletion . Transient CypA depletion using shRNA expression had a similar effect as CypA inhibition with Cs , also rescuing infectivity reduced by Nup358 depletion ( Figure 6B ) . As expected , the CypA insensitive mutants G89V or P90A did not respond significantly to Cs treatment or CypA depletion by RNAi respectively ( Figure 6A , B ) . The infectivity of the HIV-1 CA mutant N74D was slightly reduced by Cs consistent with its reduced sensitivity to Nup358 depletion and supporting the notion that it is still able to recruit CypA as confirmed by ITC ( Figure 3B , 6A and Figure S8 ) . Cs also partially rescued HIV-1 infectivity in cells with strong TRN-SR2 depletion ( Figure 6A and Figure S8 ) , suggesting that TRN-SR2 participates in the Nup358 dependent import pathway into which the virus is directed by CypA . We were also able to show that the distantly related HIV-1 O-group virus MVP5180 was also specifically rescued upon Cs treatment/CypA depletion in Nup358 or TRN-SR2 depleted cells but was unaffected in control cells ( Figure 6A , B ) . This suggests that MVP5180 functionally interacts with CypA in a similar way to NL4 . 3 and this is concordant with the very similar co-crystal structures of M-group HIV-1 CANTD with CypA and O-group HIV-1 CANTD with CypA ( PDB ID: 1M9D ) [39] . Together these observations made using both NL4 . 3 and MVP5180 suggest that CypA acts upstream of Nup358 and that Nup358 is not required for HIV-1 infectivity in the absence of CypA activity . In other words we propose that CypA activity directs the virus to engage Nup358 . If CypA activity directs HIV-1 to interact with cytoplasmic Nup358 to traverse the NPC then reduced HIV-1 infectivity through depletion of nuclear pore proteins that act downstream of Nup358 should also be rescued by CypA inhibition . To test this we analyzed infectivity of HIV-1 NL4 . 3 and its CA mutants in HeLa cells depleted for Nup153 ( Figure 6C ) . Nup153 is a NPC component in the nuclear basket and has been highlighted in genome wide siRNA screens as co-factor for HIV-1 [14] , [40] , [41] . We found that HIV-1 NL4 . 3 infectivity was strongly reduced in Nup153 depleted cells by ∼10-fold , whereas MLV infection was not affected ( Figure 6C ) . However , the Cyp non-binding mutant HIV-1 CA G89V as well as mutants N74D and N57A , which are less dependent on Nup358/TRN-SR2 were only moderately affected ( ∼3-fold ) . When cells were treated with Cs during infection , infectivity reduced by Nup153 depletion was specifically rescued for wild type virus , whereas the HIV-1 CA mutants remained unaffected ( Figure 6C ) . The O-group HIV-1 MVP5180 was affected by Nup153 depletion similarly to NL4 . 3 and CypA inhibition rescued its infectivity similarly to what we observed in Nup358 depleted cells . These observations support the notion that inhibition of CypA recruitment leads HIV-1 to use different cellular cofactors , and perhaps a different pathway , for nuclear entry . Importantly , Cs treatment prevented spreading infection of wild type HIV-1 in human MDM [42] ( Figure 6D ) . Thus , productive infection in a biologically relevant cell type is dependent on the conserved use of cyclophilins . Our results suggest that inhibition of CypA may not only prevent the use of Nup358 but rather may direct HIV-1 into a Nup358/Nup153 independent nuclear entry pathway that may not be available or functional in MDM . In one possible model , for some cell lines such as HeLa cells , the use of alternate pathways and retargeting of integration preferences may not lead to large infectivity defects particularly when measuring infectivity using VSV-G pseudotyped HIV-1 vectors with GFP driven from a heterologous promoter . In replication assays using full length HIV-1 and primary targets of HIV-1 infection such as macrophages , Cs treatment ( Figure 6D ) or CA residue changes ( Figure 4F ) have their strongest inhibitory effects . Whilst our observations can be explained by various models , they support the notion that the cofactors that the virus has evolved to use , and has conserved the use of , such as Nup358 and TRN-SR2 , may be most important in the primary cells in which the virus naturally replicates . A model is presented in cartoon form ( Figure 7 ) .
Here we have presented data suggesting that HIV-1 uses a pathway that includes the cytoplasmic cyclophilin CypA and the nuclear pore associated cyclophilin Nup358 to access the nucleus and target preferred regions of the genome for integration . We find that a determinant for the use of this pathway is CA and we demonstrate the first direct interaction of HIV-1 CA with a component of the nuclear pore complex . Disrupting engagement of CypA/Nup358 by mutating CA or inhibiting CypA with Cs appears to cause HIV-1 to use a Nup358/Nup153 independent pathway . The role of CypA in this process remains obscure but our data suggests that it directs HIV-1 to utilize a nuclear entry pathway involving Nup358 and Nup153 . Indeed , roles for CypA in nuclear transport of cellular factors have been proposed before [43]–[45] . Our data illustrate that this HIV-1 nuclear import pathway is directly linked to integration site preference , which provides a candidate explanation for reduced replication in human MDM . Intriguingly , HIV-1 CA substitutions can influence the regions of the genome that the virus targets for integration . We have distinguished between the regions targeted for integration using criteria related to the density of transcription units , including GC content , DNAaseI hypersensitivity and gene expression . Infections with HIV-1 variants containing substitutions in CA that prevent CypA binding ( e . g . G89V ) , and inhibiting CypA binding with Cs , both lead to increased frequency of integration in regions with higher densities of transcription units ( Figure 4 and 5 ) , supporting the consistency of our observations . Furthermore , mutations that render HIV-1 less sensitive to both Nup358 and TRN-SR2 depletion ( CA N57A and N74D ) both shift integration preferences to regions with lower densities of transcription units ( Figure 4 ) . Thus mutations that prevent HIV-1 utilizing Nup358 and TRN-SR2 have the same effect as depletion of these proteins , as described in our previous study [16] . These observations suggest that nuclear entry pathways may lead to different areas of chromatin and provide probes to investigate this possibility . Several reports have suggested a role for HIV-1 CA in nuclear entry [7] , [8] , [10] , [17] , [40] , [46] . Using ITC experiments we demonstrate here that HIV-1 binds to the nuclear pore through interactions between CA and the C-terminal Cyp domain of Nup358 . This is the first direct evidence for an interaction between CA and the NPC and suggests that CA-containing PICs or whole capsid cores dock at the NPC prior to nuclear entry as previously inferred from microscopy studies [47] . Recruitment of cores through Nup358 may assist appropriate uncoating and interaction of PICs with the nuclear transport machinery including TRN-SR2 and Nup153 . Remarkably , Nup358Cyp shows evidence for positive selection and a positively selected residue affects restriction by TRIMNup358 suggesting that this residue impacts on HIV-1 CA binding . This is the first case of an HIV-1 co-factor displaying signs of positive selection . We speculate that ancient pathogens , possibly viruses , may have provided the necessary selective pressure for the change of residue 61 from methionine , which has been conserved in the entire cyclophilin family , to valine , isoleucine or leucine in Nup358Cyp . It will be interesting to examine whether other viruses that encounter the nucleus during their life cycle use Nup358 and whether this position influences their recruitment . Indeed , Nup358 has been suggested to be involved in HSV-1 capsid attachment to the nucleus , however the viral determinants for this process remain obscure [48] . We also demonstrate that HIV-1 CA sequence influences the sites in which HIV-1 integrates . Although there are many possible explanations for that , we hypothesize that this occurs through selection of the cofactors for nuclear import or the nuclear import pathway . In the future , it will be interesting to investigate whether TRN-SR2 functions to enhance cytoplasmic availability of HIV-1 co-factors required for nuclear import or integration site selection . Interaction with such co-factors may be disturbed by CA mutations leading to impaired nuclear import or integration . Surprisingly SIVmac , a primate lentivirus from rhesus macaques that was derived experimentally from SIV from sooty mangabeys [49] does not appear to utilize Nup358 during infection . SIVmac is however sensitive to TRN-SR2 depletion suggesting that it uses a related but somewhat different set of co-factors to enter the nucleus as compared to HIV-1 . SIVmac is known to integrate into genes in a similar way to HIV-1 but subtle differences between HIV-1 and SIVmac integration targeting may exist . The significance of these observations remains unclear and characterization of the pathways used by a variety of lentiviruses to enter the nucleus and target favored sites will undoubtedly be informative . Whilst our data don't rule out partial cytoplasmic uncoating we envisage the HIV-1 CA acting as a protective cage around the reverse transcription complex , shielding the viral macromolecules from pattern recognition by innate immune mediators present in the cytoplasm . Antagonistic Nup358 and CypA activities could be explained by a model in which CypA stabilizes or protects the core [50] , whilst Nup358 regulates uncoating at the nuclear pore [47] . In this regard Nup358 binding to the conical viral core could have different effects from monomeric CypA , as eight Nup358 proteins are attached to the NPC [51] . Multiple simultaneous Nup358-CA interactions might destabilize the HIV-1 core and liberated PICs could then interact with TRN-SR2 and the nuclear located Nup153 ensuring transport through the NPC to appropriate sites [52] . This model provides a rationale for conservation of Cyp binding and explains how CA might influence TRN-SR2 or Nup153 usage without direct interaction [22] , [40] , [46] , [52] . This model may also explain how the use of TRN-SR2 does not correlate with the ability of various integrase proteins to bind TRN-SR2 protein in vitro [46] . If lentiviruses regulate uncoating through interactions between CA and other host factors then their integrase proteins may be exposed to different karyopherins during this process . In this way the CA sequence and structure might be a stronger influence on the choice of karyopherins than integrase despite integrase being the ultimate target for karyopherin interaction . The Nup358 cyclophilin domain has been suggested to act as a chaperone by mediating prolyl cis-trans isomerization of cellular proteins [29] , [30] . It will be interesting to investigate whether Nup358 is enzymatically active on the HIV-1 capsid core and whether this causes uncoating at the nuclear pore . Currently available cyclosporins do not antagonize Nup358Cyp binding to HIV-1 CA but the fact that cyclophilins can be pharmacologically inhibited suggests the possibility of specifically inhibiting HIV-1 CA-Nup358Cyp interaction and possibly HIV-1 replication . Overall , our data demonstrate that rather than being lost during cytoplasmic uncoating , HIV-1 CA binds to the nuclear pore component Nup358 and directs the virus into a pathway that regulates its traffic between the cytoplasm and chromatin , playing a key role in the integration site targeting required for optimal continuation of the viral replication cycle .
VSV-G pseudotyped vectors derived from HIV-1 , SIVmac and MLV-B have been described as has their preparation by 293T transfection [53] . The HIV-1/SIVmac chimeric vectors have been described [20] , [21] as has the HIV-1 vector encoding MVP5180 Gag [19] . HIV-1 NL4 . 3GFP-IRES has been described [54] . Viral doses were measured by reverse transcriptase ( RT ) enzyme linked immunosorbant assay ( Roche ) . Viral vector infection assays using VSV-G pseudotyped viruses encoding GFP were analyzed by enumerating the number of green cells 48 hours post infection by flow cytometry . Viral vector infectivity experiments were performed in a 24-well plate format as described [53] . To measure late RT products , 2-LTR circles and integrated provirus , control or shRNA expressing cells were infected with VSV-G pseudotyped HIV-1 GFP encoding vector and then grown for indicated times . Total DNA was purified from 2 samples at each time point ( QiaAmp , Qiagen ) and 600 ng were subjected to Taqman quantitative PCR using late RT [55] , 2-LTR circle [56] or GFP [53] primers and probe to detect provirus as described . Infectivity was measured in parallel samples by flow cytometry 48 hours post infection . MDM were prepared from fresh blood from healthy volunteers as described [57] . Cells were infected with 400 pg RT/well in 24-well plates and subsequently fixed and stained using a CA specific antibody ( CA183 ) and a secondary antibody linked to beta galactosidase as described [57] . For measuring the effect of CypA inhibition on HIV-1 replication the assay was performed in the presence of 5 µM DMSO or Cs throughout the whole time course . TZM-bl infection assay was performed with 50 pg RT/20000 cells in 24-well plate dishes and RLU were measured at indicated time points . Methods for integration site sequencing and heat map and dendogram analysis have been described [16] . All RNA interference experiments were performed by expressing short hairpin RNA from either MLV vector pSIREN RetroQ ( Clontech ) ( for Nup358 , TRN-SR2 and Nup153 ) or pSUPER ( Oligoengine ) ( for CypA ) or if indicated from the HIV-1 vector pCSRQ , which was derived by subcloning the shRNA expression cassette from pSIREN RetroQ into pCSGW . The CypA shRNA target sequence has been described [58] . The Nup358 shRNA target sequence that was used throughout the study was 5-GCGAAGTGATGATATGTTT-3 . Nup153 shRNA target sequence was 5-CAATTCGTCTCAAGCATTA-3 . Both sequences were selected as 1 of the 4 target sequences from the Dharmacon siRNA smartpool for Nup358 or Nup153 , respectively . Both shRNAs had only minor toxic effects on the cells , unlike shRNAs derived from the other three target sequences of each smart pool ( Figure S1A , and data not shown ) . Additional Nup358 shRNA target sequences used in the experiment shown in Figure S1A were shRNA2 5-CAAACCACGTTATTACTAA-3 , shRNA3 5-CAGAACAACTTGCTATTAG-3 and shRNA4 5-GAAGGAATGTTCATCAGGA-3 . Specificity for each target sequence was confirmed by BLAT ( UCSC genome browser ) . For Nup358 , we confirmed effective targeting by co-transfecting the shRNA expression vector with a plasmid encoding GFP-tagged Nup358 ( Figure S2 ) , as well as by western blotting using a Nup358 specific antibody ( Figure 1B ) . TRN-SR2 target sequence and control have been described and were also validated by co-transfecting a plasmid encoding for TRN-SR2-IRES-eGFP with the expression vectors encoding shRNA or control ( SC ) ( Figure S2 ) [22] . The observations made for shRNA expressing HeLa cells were similar between populations of puromycin selected cells and clonal cells but the phenotype of cell clones was more stable , thus we used single cell clones for all experiments ( Figure S1A , B and data not shown ) . Nup358 , TRN-SR2 , Nup153 , CypA and beta-Actin were detected by western blot using a Nup358 antibody kindly given by Frauke Melchior , mouse TRN-SR2 antibody ab54353 ( Abcam ) , mouse Nup153 antibody ab24700 ( Abcam ) , rabbit CypA antibody SA296 ( Biomol ) and mouse beta-Actin antibody ab6276 ( Abcam ) and appropriate horseradish peroxidase linked secondary antibodies . TRIMCypA and TRIMNup358 were detected using anti-HA antibody 3F10 ( Roche ) . Cyclosporine ( Sandoz ) and aphidicolin ( Sigma ) were diluted in DMSO and used at 5–8 µM and 2 µg/ml , respectively . Isothermal titration calorimetry was performed as described [25] . Codon-specific selection analysis was performed using the Random Effect Likelihood ( REL ) algorithm as described [27] using the alignment in Figure S6 . | During infection HIV-1 enters the nucleus by crossing the nuclear membrane and incorporating itself into the host DNA by a process called integration . Here we show that the viral capsid protein gets tethered to a cyclophilin protein called Nup358 , a component of the nuclear membrane gateways that allow transport between the cytoplasm and the nucleus . Altering the capsid protein so that it cannot use Nup358 prevents viral replication in macrophages , a natural target cell type for HIV-1 . Intriguingly , these viral mutants are not less infectious in certain immortalised cell lines suggesting that in these cells nuclear entry is regulated differently . In this case similar to wild type virus , the mutant viruses integrate into host chromosomes but they integrate into different regions suggesting that the pathway into the nucleus dictates where the virus ends up in the host chromatin . We also show that another cyclophilin , the cytoplasmic protein cyclophilin A , influences the engagement of Nup358 as well as other proteins involved in HIV-1 nuclear entry . We hypothesise that HIV-1 has evolved to use cyclophilins so that it can access a particular pathway into the nucleus because alternative pathways lead to defects in integration targeting and viral replication in human macrophages . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"biochemistry",
"infectious",
"diseases",
"molecular",
"cell",
"biology",
"immunology",
"biology",
"microbiology",
"evolutionary",
"biology",
"proteomics"
] | 2011 | HIV-1 Capsid-Cyclophilin Interactions Determine Nuclear Import Pathway, Integration Targeting and Replication Efficiency |
While the induction of broadly neutralizing antibodies ( bNAbs ) is a major goal of HIV vaccination strategies , there is mounting evidence to suggest that antibodies with Fc effector function also contribute to protection against HIV infection . Here we investigated Fc effector functionality of HIV-specific IgG plasma antibodies over 3 years of infection in 23 individuals , 13 of whom developed bNAbs . Antibody-dependent cellular phagocytosis ( ADCP ) , complement deposition ( ADCD ) , cellular cytotoxicity ( ADCC ) and cellular trogocytosis ( ADCT ) were detected in almost all individuals with levels of activity increasing over time . At 6 months post-infection , individuals with bNAbs had significantly higher levels of ADCD and ADCT that correlated with antibody binding to C1q and FcγRIIa respectively . In addition , antibodies from individuals with bNAbs showed more IgG subclass diversity to multiple HIV antigens which also correlated with Fc polyfunctionality . Germinal center activity represented by CXCL13 levels and expression of activation-induced cytidine deaminase ( AID ) was found to be associated with neutralization breadth , Fc polyfunctionality and IgG subclass diversity . Overall , multivariate analysis by random forest classification was able to group bNAb individuals with 85% sensitivity and 80% specificity based on the properties of their antibody Fc early in HIV infection . Thus , the Fc effector function profile predicted the development of neutralization breadth in this cohort , suggesting that intrinsic immune factors within the germinal center provide a mechanistic link between the Fc and Fab of HIV-specific antibodies .
Antibodies form a link between the adaptive and innate immune systems and serve as a correlate of protection for many viral vaccines . They mediate diverse functions through the use of the Fab portion to bind specific antigens and the Fc portion that interacts with cellular receptors to effect a variety of additional non-neutralizing activities [1] . As yet , no HIV vaccine has been able to elicit broadly neutralizing antibodies ( bNAbs ) , however moderate efficacy in the RV144 vaccine trial correlated with Fc-mediated antibody-dependent cellular cytotoxicity ( ADCC ) in the absence of IgA response generating intense interest in understanding how these responses evolve [2–8] . Maturation of antibodies occurs in the germinal center where they undergo somatic hypermutation to generate high affinity neutralizing antibodies , as well as class-switch recombination to select constant regions that determine the scope of Fc effector functions . Both processes are dependent on the enzyme activation-induced cytidine deaminase ( AID ) as well as the cytokine milieu within the germinal center , suggesting that the Fab and Fc maturation processes of antigen-specific antibodies may be jointly regulated [9–11] . Although current vaccination strategies are unable to induce bNAbs , approximately 10–30% of individuals produce bNAbs during the course of HIV infection , with only 1–2% classified as elite neutralizers [12–16] . Several factors have been associated with the development of neutralization breadth , including duration of infection , high viral load , low CD4 count , genetic subtype and viral diversity [12 , 13 , 16–18] . The role of host immunological factors is less clear but includes the maintenance of a high level of T follicular helper cells , increased levels of CXCL13 early in infection , autoimmunity and ethnicity [16 , 19–22] . Isolation of bNAbs has revealed that they have unusual features , most notably a high level of somatic hypermutation , which is essential for mediating neutralizing activity [23 , 24] . Given the importance of bNAbs for HIV vaccine design , considerable effort has been invested in understanding how these antibodies evolve [25] . However , far less is known about the corresponding Fc effector response and IgG subclass usage in individuals who develop neutralization breadth . The antibody Fc portion mediates a wide range of functions that are dependent on their affinity for activating and inhibiting Fc receptors , lectins and complement proteins [26] . These functions include phagocytosis of pathogens by macrophages and monocytes , ADCC or direct lysis by natural killer cells , complement deposition and trogocytosis , an Fc dependent exchange of membrane proteins on the surface of an infected to an uninfected cell which may result in cell death [27 , 28] ( Richardson , et al . , submitted ) . Fc effector functions are modulated at the B cell level by biochemical properties such as glycosylation at a conserved site in the CH2 region and isotype or subclass selection determined by the cytokine milieu in the germinal centers [29–32] . Affinity and specificity for the antigen as determined by the Fab portion has also been shown to impact Fc effector functions [33–35] . Several studies have shown that Fc effector function may influence HIV disease progression , viral control and infant mortality [36–43] . Co-ordination between Fc effector functions has also been associated with HIV control [43 , 44] , which was also observed in the responses to the partially protective RV144 vaccine [3 , 5] . Furthermore , protective immunity in an adenovirus-26 based vaccine in primates was linked to Fc polyfunctionality [45] . However , direct evidence that non-neutralizing antibodies can protect via Fc-mediated functions is limited [46–48] although one study showed a reduced number of transmitted/founder viruses as a result of Fc activity [49] . More recently , non-neutralizing mAbs have been shown to clear HIV-infected cells and protect against infection in humanized mice , despite being less effective than bNAbs [50] . Perhaps more convincing is the accumulating evidence that Fc binding is required for bNAbs to optimally protect from infection , supress viral load or clear infected cells [51–55] . This study aimed to investigate Fc effector functions and kinetics in the context of a broad neutralizing response . Furthermore , we examine potential modulators of this response including binding to Fc receptors , IgG subclass diversity and the interplay between germinal center activity , Fc polyfunctionality and neutralization breadth . Our data show that individuals with bNAbs also have higher Fc polyfunctionality and increased subclass diversity , which was associated with greater AID expression in B cells . This study suggests that the Fc and Fab portions of HIV-specific antibodies are co-ordinately regulated and identifies an Fc effector profile that is associated with the development of bNAbs .
We aimed to determine whether individuals with HIV-1 specific bNAbs showed differences in Fc effector functions and what the kinetics of these responses were relative to the development of neutralization breadth . Plasma IgG was isolated from 13 bNAb and 10 no-bNAb individuals at 6 , 12 and 36 months post-infection matched for viral load ( S1 Table and S1A Fig ) , a major driver of immune activation and neutralization breadth [12 , 56] . These were tested in 4 different Fc effector assays measuring ADCC ( % granzyme B release ) , ADCP ( % fluorescent bead phagocytosed x MFI of phagocytosed beads ) , ADCD ( % C3b deposition x MFI of deposited C3b ) and ADCT ( % PKH-26 stained membrane transferred to CFSE+ macrophages ) . Three different HIV envelope glycoproteins were used to detect HIV-specific Fc responses including gp120 ConC , gp140 C . ZA . 1197MB and gp120 CAP45 . Transmitted/founder or acute viruses from 10 bNAb and 7 no-bNAb individuals for which sequences were available were used to calculate divergence from the sequence of the three antigens . We found that both groups showed equivalent levels of sequence divergence from each antigen , indicating no bias in terms of antigen binding based on autologous virus sequence ( S1B Fig ) . Furthermore , analysis of longitudinal gp160 sequences in 8 bNAb and 5 no-bNAb individuals with available data showed that viral diversity did not differ between the 2 groups , perhaps not unexpected given they were matched for viral load ( S1C Fig ) . IgG from almost all individuals , irrespective of the presence of bNAbs , had Fc effector function activity with ADCP and ADCD being the most readily detectable , particularly in response to gp120 ConC ( Fig 1A ) . At 6 months of infection , IgG from bNAb individuals showed significantly higher ADCD against gp120 ConC ( p = 0 . 030 ) , gp140 ( p = 0 . 042 ) and gp120 CAP45 ( p = 0 . 047 ) compared to no-bNAb individuals . This was also noted at 12 months for gp140 ( p = 0 . 030 ) ( S2 Fig ) . ADCT was higher in bNAb individuals at both 6 and 12 months post-infection against gp120 ConC ( p = 0 . 002; p = 0 . 030 ) and gp140 ( p = 0 . 004; p = 0 . 014 ) . There was no difference observed between the 2 groups for ADCC and ADCP against any of the antigens tested but all were significantly higher than the HIV-negative IgG control . A comparison of the cumulative activity of all 4 Fc effector functions against all 3 antigens over 36 months of infection between the two groups highlighted the differences seen at 6 months ( Figs 1B and S2 ) . Significant increases over time were noted primarily in the no-bNAb group for ADCD and ADCC and in the bNAb group for ADCT while high levels of ADCP were maintained in both groups ( S2 Fig ) . No-bNAb individuals had a steeper increase in function over time , as a result of a lower initial Fc effector response . Regardless of neutralization breadth , Fc effector function in both groups was comparable by 36 months . Adsorption of broad neutralizing activity from an individual with bNAbs ( CAP255 ) revealed a significant 3-fold reduction in ADCC , ADCP and 2-fold reduction in ADCD against gp120 ConC compared to the unadsorbed IgG ( S3 Fig ) . This suggests that the N332-specific antibodies that mediate broad neutralizing activity in CAP255 [12] , were also responsible for a significant proportion of Fc effector functionality . In order to account for the contributions of all four effector functions equitably , we calculated an Fc polyfunctionality Z-score using the gp120 ConC data from 6 months post-infection . Z-scores with values greater than zero indicated samples with good Fc polyfunctional activity . Of the 13 bNAb individuals , 8 showed positive Z-scores compared to only 1 of the no-bNAb individuals who had a weakly positive Z-score ( Fig 1C ) . The remaining 9 no-bNAb samples had negative Z-scores compared to 5 in the bNAb group . Thus IgG samples from individuals who later went on to develop bNAbs could be distinguished from no-bNAb individuals on the basis of their higher overall polyfunctional polyclonal response against gp120 ConC early in infection . Furthermore , neutralization breadth measured at 3 years post-infection correlated significantly with the Fc polyfunctionality Z-score calculated using data from 6 months of infection ( r = 0 . 45 , p = 0 . 030 , Fig 1D ) . There was also a correlation between these gp120 ConC Z-scores and those calculated using gp140 ( r = 0 . 69 , p<0 . 001 , S4A Fig ) , an antigen that also showed a significant correlation with neutralization breadth ( S4B Fig ) . Fc polyfunctionality Z-scores calculated using CAP45 . G3 gp120 did not distinguish bNAb individuals unlike the other 2 antigens ( Figs 1C and S4 ) . This is likely to be as a result of CAP45 . G3 gp120 having significantly lower titers of binding IgG antibodies than both ConC gp120 and C . ZA1197MB gp140 ( S5A Fig ) . In order to assess collaboration between different Fc effector functions , correlation coefficients between the functions were calculated . There was significant discordance in the no-bNAb group with negative correlations noted between ADCP and all other functions and concordance between ADCT and ADCD ( S4C Fig ) . In contrast , the bNAb group showed no significant discordance or concordance between Fc effector functions , further highlighting that individuals with and without bNAbs have distinct Fc effector function profiles early in HIV infection . Since antibody function is modulated by binding to different Fc receptors and complement proteins , we assessed the binding of gp120 ConC-specific IgG to 6 different Fc receptors and C1q ( the first subcomponent of the C1 complex of the classical pathway of complement activation ) using a multiplex-based assay ( as described in [57] ) . Polymorphisms of the FcγRIIa ( both R131 and H131 ) and FcγRIIIa ( both F158 and V158 ) receptors that affect IgG binding and impact HIV disease progression ( reviewed in [58] ) were also tested . Individuals with bNAbs had significantly higher levels of binding to all receptors except FcγRIIa-R131 where only a trend was evident ( Fig 2A ) . Since FcγRIIa is the primary receptor that mediates phagocytosis and this process is negatively regulated by FcγRIIb [59] , an increase in binding to both activating and inhibitory receptors among bNAb individuals was somewhat unexpected given that we observed no differences in phagocytosis between the two groups ( Fig 1A ) . We therefore determined the ratio of antibody binding to these 2 receptors which has previously been shown to correlate with phagocytic function [59 , 60] . Here we found no differences between the bNAb and no-bNAb groups irrespective of the FcγRIIa-H131/R131 genotype which was more consistent with the finding of similar ADCP levels in the two groups ( Fig 2B ) . ADCT correlated with binding to FcγRIIa and FcγRIIIa ( Fig 2C ) , two receptors that have been implicated in mediating this function [61 , 62] ( Richardson , et al . , submitted ) . In addition , we found several correlations for ADCD including FcγRIIa-R131 , FcγRIIb and FcγRIIIb , however C1q showed the strongest correlation consistent with this being the primary binding protein for complement deposition ( r = 0 . 65; p <0 . 001 ) . IgG comprises 4 subclasses each with distinct Fc regions that bind differentially to cellular Fc receptors and complement proteins . Using an antigen-specific IgG subclass multiplex assay , we measured the levels of IgG1-4 and total IgG against 12 HIV antigens including trimeric envelope , gp140 , monomeric gp120 , V2 , V3 , gp41 , membrane proximal external region ( MPER ) and p24 . First , to determine if there was any IgG subclass bias of antibodies able to mediate Fc effector functions , we analysed 6-month data with gp120 ConC as this antigen was used in all assays . Other than gp120-specific total IgG that was significantly associated with ADCD ( r = 0 . 74; p<0 . 001 ) and ADCC ( r = 0 . 46 , p = 0 . 010 ) , only ADCT was shown to correlate with HIV-specific IgG3 ( r = 0 . 48; p = 0 . 02 ) suggesting that multiple rather than single subclasses mediate Fc effector functions ( S5B Fig ) . Our earlier finding that bNAb individuals had greater binding to FcγRIIa-H131 and FcγRIIIa-V158 ( Fig 2A ) , receptors known to bind IgG2 and IgG4 [63] led us to explore whether individuals with bNAbs had a greater diversity of IgG subclasses specific to HIV . While there were no significant differences in HIV-specific IgG1 and IgG3 levels ( as a proportion of total antigen-specific IgG ) individuals with bNAbs showed significantly higher IgG2 and IgG4 binding to 7 of the 12 antigens tested , including BG505 SOSIP . 664 gp140 trimeric protein ( S6 Fig ) . This was also reflected when we represented the less frequent subclasses ( IgG2 , IgG3 and IgG4 ) as a proportion of IgG1 ( Fig 3A ) . Overall , we noted that individuals who developed bNAbs showed a greater relative abundance of IgG2 and IgG4 to IgG1 . In particular , IgG2 binding to MPER and p24 antigens and IgG4 binding to the A244 V2 antigen , gp120 ConC and gp41 were higher among bNAb individuals . Among the no-bNAb samples , levels of IgG2 and IgG4 were considerably lower with only a V2 antigen and p24 showing some reactivity . We calculated a subclass diversity score based on levels of IgG2 plus IgG4 relative to IgG1 and found this to be correlated with neutralization breadth ( r = 0 . 40 , p = 0 . 049 , Fig 3B ) and the Fc polyfunctionality Z-score ( r = 0 . 43 , p = 0 . 040 , Fig 3C ) . These data suggested that individuals who develop bNAbs in chronic infection undergo increased IgG class switching within the first 6 months of infection . Both class switch recombination and somatic hypermutation take place in germinal centers and are enabled by activation-induced cytidine deaminase ( or AID ) as well as interactions between T helper follicular ( Tfh ) cells with B cells . The cytokine CXCL13 , expressed by Tfh cells [64 , 65] , has been shown to correlate with neutralization breadth early in HIV infection and is defined as a marker of germinal center activity [19–21 , 56 , 66] . We measured CXCL13 levels in plasma by enzyme linked immunosorbent assay ( ELISA ) and found significantly higher levels in individuals with bNAbs at 6 months post infection ( p = 0 . 002 ) but not at later time points ( Fig 4A ) . HIV-negative samples were significantly lower than HIV-positive plasma ( p<0 . 001 ) similar to other studies showing that HIV infection is associated with increased expression of this cytokine [21] . The levels of CXCL13 at 6 months were also correlated with neutralization breadth ( r = 0 . 52 , p = 0 . 010 , Fig 4B ) and Fc polyfunction measured against gp120 ConC ( r = 0 . 60 , p = 0 . 003 , Fig 4C ) . CXCL13 was not correlated with IgG subclass diversity or total IgG levels . We next directly measured AID in B cells from 6 bNAb individuals and 6 no-bNAb individuals from 6 months post infection , from whom PBMC were available . In order to detect AID at a reliable level we stimulated PBMCs for 3 days with TLR9 and confirmed stimulation by the co-staining of live B cells ( defined as CD3/CD16/CD14- CD19+ as in S7A Fig ) with AID and with Ki67 ( a marker of proliferating cells ) . As expected , stimulation increased the percentage of B cells expressing AID but no significant difference was observed among the 2 groups or HIV-negative individuals ( S7B Fig ) . However , when AID expression was measured by the mean fluorescence intensity ( MFI ) of AID , we noted significantly higher expression of AID in stimulated B cells from individuals with bNAbs compared to no-bNAbs and HIV-negative individuals ( p = 0 . 010 , Kruskal-Wallis test with Tukey correction for multiple comparisons , Fig 4D ) . Even on unstimulated B cells , median AID expression levels tended to be higher in bNAb individuals . Interestingly , both neutralization breadth ( r = 0 . 58; p = 0 . 040 , Fig 4E ) and IgG subclass diversity ( r = 0 . 60 , p = 0 . 040 , Fig 4F ) correlated with AID expression suggesting germinal center activity not only plays a role in the development of bNAbs but is also an indicator of enhanced Fc effector function . In order to ascertain which of the 29 variables examined in this study were best able to separate bNAb from no-bNAb individuals , we used Spearman’s correlations that were adjusted for multiple comparisons by the Benjamini–Hochberg method ( S2 Table ) . Owing to the small sample size of the individuals tested for AID , this variable was omitted . Subclass diversity , Fc receptor and C1q binding antibodies targeting gp120 from both CAP45 and ConC , Fc polyfunctionality and CXCL13 were all significantly associated with the bNAb group . CD4 count was significantly negatively associated with the bNAb group and , as expected , we saw no association with viral load due to matching of the two groups for this variable . The 17 variables that showed significance , were used in further statistical analyses to refine groupings . When subjected to a principal components analysis , the sum of the first 2 components were able to explain 52 . 3% of the variability of the data set ( Fig 5A ) . Furthermore , when random forest classification was used to define bNAb and no-bNAb groups , it did so with 85% sensitivity 80% specificity , and 82 . 6% accuracy , only misidentifying 2 bNAb and 2 no-bNAb individuals shown in the confusion matrix in Fig 5B . The features important for this classification are represented by the Gini importance weighting , and identified increased binding to gp120 ConC FcγRIIIa-V158 , low CD4 count , gp120 ConC-specific total IgG and Fc polyfunction for gp120 ConC and gp140 as being the best features to classify bNAb individuals ( Fig 5C ) . We then tested the strength of the model by permutation testing , allowing for 100 , 000 random data shuffles , revealing that there was a 0 . 38% probability that the same model could be obtained at random ( Fig 5D ) . Collectively , these analyses indicate that Fc properties can be used to reliably discriminate between individuals who develop bNAbs from those who do not .
In this study , we found that HIV-infected individuals who develop bNAbs also have a distinct Fc effector function profile and increased subclass diversity associated with markers of enhanced germinal center activity . Specifically , these individuals showed more potent trogocytosis and complement deposition as well as IgG2 and IgG4 responses to multiple HIV antigens early in infection compared to individuals who did not develop bNAbs . Interestingly , both CXCL13 levels in plasma and AID expression levels in activated B cells correlated with neutralization breadth and Fc polyfunctionality . Our data suggest that markers of germinal center activity that link Fab and Fc could be exploited for vaccine design to harness the full potential of HIV-specific antibodies . To conduct this study we made use of longitudinal samples collected from HIV-1 infected participants in the CAPRISA acute infection cohort in which the development of bNAbs has been well described [12] . We selected all 13 participants who developed bNAbs by 3 years of infection as well as 10 matched no-bNAb controls . Less than 20% of individuals in the CAPRISA cohort developed bNAbs , similar to many other studies , and so we selected all 13 available bNAb individuals and matched them to those with no-bNAbs that had similar viral levels . While this is not representative of the general population , our aim was to compare those 2 distinct phenotypes in order to maximize our ability to detect differences associated with bNAb development . We found that , unlike neutralization , Fc effector function was detected early in infection in the majority of individuals . Furthermore , levels of ADCD , ADCT , ADCC increased over time while ADCP remained high throughout 3 years of infection . Other studies that have examined the kinetics of ADCC [36 , 67–69] and ADCP [70–72] from acute to chronic HIV infection have shown inconsistent results with some showing higher activity early in infection while others suggest , similar to our study , that these activities increase over time . These differences are likely due to the use of a variety of assays and antigens and supports the efforts being made to standardise Fc effector assays [43] . Interestingly , we found differences in Fc effector functions between bNAb and no-bNAb individuals at 6 months post-infection prior to the development of neutralization breadth . Other factors previously shown to be associated with neutralization breadth , such as high viral load and low CD4 count , in this and other cohorts , were also seen early in infection [12 , 14 , 18 , 20] . However , at 3 years post-infection we found no differences in Fc effector function between the two groups further highlighting that events early in HIV infection are signatures , but not necessarily drivers , for the later development of neutralization breadth . Positive correlations between different Fc effector functions have previously been reported among elite controllers and vaccinees in the RV144 trial suggesting a co-ordinated immune response in these groups [5 , 44] . Our study showed that while bNAb individuals did not show significant concordance between Fc functions there was also no significant discordance between the various functions . This differentiated them from no-bNAb individuals who showed significant discordance , which has also been seen in chronic HIV infection [43] . Only one other study has examined Fc effector function in the context of neutralization breadth and found that ADCC did not differ between the groups , similar to our findings [69] . Here we extend this to examine additional Fc effector functions and find that bNAb individuals could be distinguished from no-bNAb individuals by having higher levels of ADCD and ADCT at 6 months of infection . Recent data suggested that ADCD may have been a correlate of reduced risk in RV144 as a result of V2-specific antibodies efficiently activating complement [73] . In our study , ADCD was strongly correlated with total HIV-specific IgG levels which may have facilitated increased IgG complex formation in individuals that develop bNAbs [74] . Furthermore , binding of C3 components to CR2 ( complement receptor 2 ) on the surface of follicular dendritic cells ( FDC ) in the germinal center facilitates the presentation of antigen to B cells during the processes of affinity maturation and isotype switching [75 , 76] . Thus our finding of higher levels of C3b deposition by IgG present in bNAb individuals could contribute to increased antigen presentation through binding to CR2 on FDCs . In contrast , increased ADCT was not due to higher IgG levels although it was the only Fc effector function that showed a correlation with a single IgG subclass , namely IgG3 . Since IgG3 preferentially binds to higher affinity polymorphic variants FcγRIIa-H131 and FcγRIIIa-V158 this may explain the associations we observed between binding to these receptors and ADCT . While the role of Fc-mediated trogocytosis in HIV infection has not yet been explored , one potential mechanism that has been reported in studies of cancer is that repeated “nibbling” of membrane proteins on tumorigenic cells by effector cells can result in cell killing [27] . Others have shown ADCT facilitates the deposition of “snatched” antigen-specific IgG on the surface of effector cells , potentially increasing antigen presentation for T cell help [77–79] . Whether either of these mechanisms contributed to the development of neutralization breadth will require an analysis of the antigens captured by ADCT and if these are able to enhance HIV-specific immune responses when deposited on antigen-naïve cells . The relative abundance of IgG2 and IgG4 was higher in bNAb individuals across multiple HIV antigens and this correlated with neutralization breadth . In support of this , we found that antibodies from bNAb individuals had higher binding to FcγRIIa-H131 and FcγRIIIa-V158 as compared to the respective lower affinity polymorphic variants R131 and F158 . H131 and V158 are known to bind more efficiently to IgG2 and IgG4 respectively although these levels are still significantly lower than IgG1 and IgG3 [63 , 80] . This may reflect increased relative abundance of these subclasses in bNAb individuals which may also explain why we saw no differences in ADCC . Both IgG2 and IgG4 , located downstream of IgG3 and IgG1 on chromosome 14 [81 , 82] , have been shown to dampen down the inflammatory response by competing with IgG1 [83 , 84] . Furthermore , IgG4 binds FcγRIIb , the only inhibitory Fc receptor , with a higher affinity than all other subclasses [80] . Thus the use of multiple IgG subclasses in bNAb individuals may help to balance the highly pro-inflammatory activities of IgG1 and IgG3 early in infection , increasing the diversity of the antibody response [85] and potentially promoting events that are required for later development of neutralization breadth . Although HIV-specific IgG2 and IgG4 showed greater relative abundance , it should be noted that both were still present at very low levels compared to IgG1 and IgG3 . Despite this and the fact that IgG2 and IgG4 have reduced Fc effector functionality , subclass diversity in this study still correlated with overall Fc polyfunction . An excess of IgG2 and IgG4 is thought to have compromised the effectiveness of gp120 vaccines possibly due to competition with IgG1 and IgG3 [5 , 86 , 87] . Thus , in current vaccination strategies we may lack the balance of subclass diversity required to support bNAb development . Indeed , others have found higher isotype diversity , specifically IgG2 levels , among viremic controllers suggesting that a balance of IgG subclasses is a key factor in immune function [82 , 85 , 88] . Alternatively , subclass diversity may be a secondary consequence of elevated AID activity in individuals with bNAbs . If so , then somatic hypermutation required for bNAb development may come at the expense of IgG1/3 dominance because of increased downstream class switching . Irrespective of the cause , subclass diversity is an important consideration for pathogenic outcomes and may have implications in vaccination strategies [82] . Similar to others , we found that levels of CXCL13 were higher in bNAb individuals early in infection suggesting that this cytokine provides important signals for bNAb development [20 , 21 , 56 , 66] . However , we are the first to examine the relationship between CXCL13 and Fc effector function and find that it correlated more strongly with Fc effector function than with neutralization breadth . Since CXCL13 is necessary for migration of B cells to germinal centers where clonal selection , somatic hypermutation and class-switching occurs , these data suggest that the Fc region of bNAbs are subjected to similar processes . AID plays a more direct role in facilitating somatic hypermutation and class switching explaining the correlation we observed with neutralization breadth as well as subclass diversity of the Fc . Cohen and colleagues have previously shown that AID transcripts were higher in bNAb individuals by ex vivo transcriptional profiling [20] . Here we show using the different approach of flow cytometry , that expression of the AID enzyme in activated B cells from bNAb individuals is higher than no-bNAb individuals . Collectively , these data indicate that immune factors associated with germinal centers drive diversification of both Fab and Fc functions . The Fab and Fc regions are encoded by different genes and class switch recombination produces new effector function in the context of an existing antibody specificity suggesting independent evolution of these two portions . However , several groups have shown that the isotype or subclass of the constant region can subtly affect Fab function such as neutralization and avidity across various diseases [89–92] . In addition , class-switch recombination and somatic hypermutation both require AID to target transcription and DNA cleavage respectively [10] . Our study provides evidence for similar modulation of the Fc and the Fab driven by AID in the context of an HIV-specific polyclonal response . In our model , we show that HIV-specific binding to Fc receptors , Fc polyfunctionality and total HIV-specific IgG levels were among the top five features that classify bNAb individuals . Among these , low CD4 count and high HIV-specific IgG levels have been previously associated with the development of neutralization breadth [12 , 13] . Our data support findings from other cohorts showing correlations between neutralization breadth with immunologic analytes , such as CXCL13 [21 , 66] . However , we are the first to describe that enhanced complement deposition and trogocytic activity are associated with the development of neutralizing antibodies , which will need to be confirmed in other cohorts . Furthermore , we showed , through adsorption experiments , that bNAbs in one individual were also responsible for Fc effector functions . Whether these functions are mediated by the same or separate antibody molecules will require further study , preferably via the isolation of HIV-specific monoclonal antibodies with their native Fc regions . Nevertheless , our finding that the Fab and Fc functions may be jointly regulated in individuals with bNAbs could be important in the context of vaccination where the aim is to induce a polyclonal response with both neutralizing and Fc effector functionality . Overall , this study illustrates that the functions and characteristics of the antibody Fc were able to reliably classify bNAb individuals . Furthermore , early Fc function and subclass diversity predicted which HIV-infected individuals went on to develop bNAbs . Unlike broad neutralization , Fc-mediated activities were present in all individuals and suggests that favourable Fc effector function could be elicited as a prelude to a bNAb response which is encouraging for vaccination strategies . Additional studies are required to understand how neutralization and Fc effector functions can be favourably tuned to produce a protective polyclonal response . Our data suggests that common immune factors underlie the development of both neutralization breadth and Fc effector function . Moreover , as Fc effector function differences occur well before the development of neutralization breadth , these properties could be used to identify individuals with the necessary germinal center activity to respond to a vaccine aimed at the generation of bNAbs .
CAPRISA 002 and 004 acute HIV infection cohorts were approved by the Biomedical Research Ethics Committee of the University of KwaZulu-Natal ( M160791 ) and this specific study was approved by the Human Research Ethics Committee of the University of the Witwatersrand ( M150313 ) . All participants were adults and provided written informed consent to have their stored samples used for future studies . All healthy subjects in this study were adults and provided written informed consent to obtain both PBMCs ( peripheral blood mononuclear cells ) and plasma and have their samples stored for future use . Serum from participants in the CAPRISA 002 and 004 acute HIV infection cohorts [93 , 94] were previously screened for neutralization breadth using a 44 virus multi-clade panel [12] . Samples from individuals able to neutralize at least 40% of a 44 virus panel at 3 years post-infection were defined as having developed bNAbs . Individuals with no-bNAbs were also matched based on viral load at 6 months and were not significantly different at all time points tested ( S1A Fig ) . By these criteria , we included plasma samples from 13 bNAb and 10 no-bNAb individuals at 6 months , 1 year and 3 years post-infection ( S1 Table ) . IgG was isolated from plasma using Protein G according to the manufacturer’s instructions in order to eliminate the cofounding impact of cytokines or plasma proteins on cell-based assays , quantified by a nanodrop spectrophotometer ( Pierce Biotechnology , Rockford , IL ) and confirmed by IgG ELISA . Pooled IgG from HIV-positive samples from the NIH AIDS Reagent programme ( HIVIG ) was used in all assays to normalise for plate to plate variation while samples from 5 HIV-negative individuals from the same cohort were used as negative controls . Plasmids encoding histidine-tagged recombinant gp120 from ConC and gp120 from CAP45 . G3 envelope sequences were transfected using polyethylenimine 25 kDa ( Polysciences Inc , Warrington , PA ) into HEK293T cells obtained from Dr George Shaw ( University of Alabama , Birmingham , AL ) . Cells were cultured at 37°C , 5% CO2 in DMEM containing 10% heat-inactivated fetal bovine serum ( Gibco , Gaithersburg , MD ) with 50 μg/ml gentamicin ( Sigma-Aldrich , St Louis , MO ) and disrupted at confluency by treatment with 0 . 25% trypsin in 1 mM EDTA ( Sigma-Aldrich , St Louis , MO ) . Recombinant proteins were expressed and purified as previously described [95] . BG505 SOSIP . 664 gp140 trimer was produced in HEK293F suspension cells ( Invitrogen ) and purified by size exclusion chromatography ( SEC ) [96] . Prior to use trimer was subject to quality control by ELISA binding of CAP256-VRC26 . 25 and PGT151 and the lack of binding to F105 . The C . ZA . 1197MB strains of gp41 , p24 and gp140 was purchased from Immune Tech ( Lexington , New York ) , CAP88 . B5 V3 peptide , CAP248 MPER peptide and MPR . 03 were purchased from Peptide 2 . 0 ( Chantilly , Virginia ) . V1V2 scaffolded proteins were expressed in HEK293S cells ( ATCC CRL-3022 N-acetylglucosaminyltransferase I deleted ) , grown in a shaking incubator at 37°C , 5% CO2 , 70% humidity at 125rpm . Cultures were harvested after seven days and purified by sequential Ni-NTA and SEC . Measurement of divergence of autologous viral sequences from antigens was done using available transmitted/founder or acute gp120 viral sequences from 10/13 bNAb and 7/10 no-bNAb individuals which were aligned using MUSCLE ( MUltiple Sequence Comparison by Log- Expectation https://www . ebi . ac . uk/Tools/msa/muscle/ ) . A script adapted from SONAR ( Software for the Ontogenic aNalysis of Antibody Repertoires ) [97] was adapted to calculate the percentage nucleotide differences of autologous viruses compared to the antigen sequences ( ConC gp120 , C . ZA . 1197MB gp140 and CAP45 . G3 gp120 ) . Diversity in gp160 was estimated using longitudinal sequences from 8 bNAb and 5 no-bNAb individuals . This included sequences from 6 ( between 5 and 28 per individual ) , 12 ( between 8 and 30 per individual ) and 36 months ( between 5 and 19 per individual ) post-infection . CAP256 was superinfected and sequences from both the primary and superinfecting viruses were included . Mean pairwise genetic distances between sequences were calculated using MEGA v7 [98] . The THP-1 phagocytosis assay was performed as previously described [99] using 1μM neutravidin beads ( Molecular Probes Inc , Eugene , OR ) coated with gp120 ConC , gp120 CAP45 . G3 or gp140 C . ZA . 1197MB . Polyclonal IgG samples were titrated and tested at a final concentration of 100μg/ml . Phagocytic scores were calculated as the geometric mean fluorescent intensity ( MFI ) of the beads multiplied by the percentage bead uptake . This , including all other flow cytometry work was completed on a FACSAria II ( BD biosciences , Franklin Lakes , New Jersey ) . THP-1 cells were obtained from the NIH AIDS Reagent Program and cultured at 37°C , 5% CO2 in RPMI containing 10% heat-inactivated fetal bovine serum ( Gibco , Gaithersburg , MD ) with 1% Penicillin Streptomycin ( Gibco , Gaithersburg , MD ) and not allowed to exceed 4 x 105 cells/ml . ADCC activity was detected by the previously described ADCC-GranToxiLux ( GTL ) assay using antigen-coated cells [100] . This assay was chosen as it is high-throughput and has been previously validated . Whole PBMCs from a healthy donor were used as effector cells . The FcγRIIIa receptor was genotyped as being homozygous for valine at position 158 by the TaqMan SNP genotyping assay ( rs396991 ) ( Applied Biosystems , Foster City , CA ) to ensure high levels of lysis . Target CEM-NKR . CCR5 ( CEM-natural killer resistant T lymphoblast cell line transduced with CCR5 ) cells were coated with gp120 ConC and gp140 C . ZA . 1197MB at 2 . 5μg/ml and 10μg/ml respectively . Optimal coating concentration was determined by titration of the antigen and measuring residual levels of unbound CD4 with anti-CD4 FITC ( SK3 clone , BD Biosciences ) . MAb A32 was used as a positive control with Palivizumab ( MedImmune , LLC; Gaithersburg , MD ) used as negative control . The results , analysed in FlowJo ( FlowJo LLC , Ashland , Oregon ) are expressed as % Granzyme B ( GzB ) activity , defined as the percentage of cells positive for proteolytically active GzB out of the total viable target cell population . The final results are expressed after subtracting the background represented by the % GzB activity observed in wells containing effector and target cell populations in the absence of IgG . CEM-NKR . CCR5 cells were obtained for the NIH AIDS Reagent programme and were cultured at 37°C , 5% CO2 in RPMI containing 10% heat-inactivated fetal bovine serum ( Gibco , Gaithersburg , MD ) with 1% Penicillin Streptomycin ( Gibco , Gaithersburg , MD ) . ADCD was determined by the deposition of the complement component C3b on the surface of CEM-NKR . CCR5 cells [3] . Target cells were pulsed with 6μg gp120 ConC , 14μg gp120 CAP45 . G3 or 6μg gp140 C . ZA . 1197MB in 100μl of R10 media ( 10% FBS 1% Pen/Strep RPMI , Gibco , Gaithersburg , MD ) determined by titration as described above for 1 hour at room temperature and incubated with 100μg/ml of IgG preparation . HIV-negative plasma was used as a source of complement and diluted 1 in 10 in 0 . 1% gelatin/ veronal buffer ( Sigma-Aldrich , St Louis , MO ) and 150μl added and incubated for 20 minutes at 37°C . The cells were then washed in 15mM EDTA in PBS and C3b was detected by flow cytometry using an anti-human/mouse complement component C3/C3b/iC3b mAb ( Cedarlane , Burlington , Canada ) . Unpulsed cells were used as background controls and HIV-negative plasma was heat-inactivated at 56°C to remove complement as a negative control . The ADCD score was defined as geometric MFI multiplied by % cells positive for C3b deposition . CEM-NKR . CCR5 cells were pulsed with gp120 ConC ( 2 . 5μg/ml ) , gp120 CAP45 . G3 ( 25μg/ml ) or gp140 C . ZA . 1197MB ( 10μg/ml ) for 75 minutes at room temperature . Optimal coating concentrations were determined as described above . Cells were stained with PKH26 dye ( Paul Karl Horan 26 dye ) as per instructions from the manufacturer ( Sigma-Aldrich , St Louis , MO ) and resuspended at 2 million cells/ml . IgG at a final concentration of 100μg/ml were added to the cells and incubated for 30 minutes at 37°C . THP-1 cells were stained with intracellular CFSE ( carboxyfluorescein succinimidyl ester ) and 150μl at 6 . 7 x 105 cells/ml was added to the plate and incubated for a further hour at 37°C . Cells were then washed with 15mM EDTA in PBS . Flow cytometry was used to distinguish PKH26+ CFSE+ THP-1 cells ( i . e . the uninfected monocytes that have received membrane fragments from the coated cells ) and are represented as a proportion of total THP-1 cells . Doublets were excluded from the analysis by singlet gating ( Richardson , et al . , submitted ) . The assay was gated on stained CEM and THP-1 cells incubated in the absence of IgG to ensure that we did not measure antibody-independent trogocytosis . Uncoated PKH26 stained CEM cells were also incubated with THP-1 cells in the presence of HIV-specific IgG in order to ensure that the responses seen were HIV-specific . HIV-negative IgG as well as Palivizumab were used as negative controls while HIVIG was used to standardise between runs . Adsorption of broadly neutralizing antibodies that target the N332 glycan from isolated plasma IgG was performed as previously described [101] . For this we used ST09 ( 1gut-mV3 scaffold ) , a V3 scaffold that specifically binds N332-directed bNAbs [102] ( a gift from Dr Peter Kwong , Vaccine Research Center , NIH , USA ) covalently coupled to tosyl-activated magnetic beads . Following adsorption , the depletion of anti-ST09 ( 1gut-mV3 scaffold ) activity was measured by ELISA and reduction of neutralization was confirmed by neutralization assay as described elsewhere [103] . A customised multiplex assay was used as previously described [104] . Briefly , Multiplex microplex carboxylated beads ( Luminex , Madison , WI ) were coupled to 12 HIV specific antigens . Fifty μl of a 100 microspheres/μl bead preparation was incubated with purified IgG overnight ( 100μg/μl ) at 4°C . Levels of bulk IgG and IgG1-IgG4 were detected by PE-conjugated detection agents ( Southern Biotech , Birmingham , AL ) by a Bio-Plex200 . The mean of PBS only samples added to 3 times the standard deviation was subtracted from all samples . Similarly , for the Fc binding array , HIV-specific antigen coated microspheres ( 10 beads/μl ) were added to 5μg/ml of purified IgG in black clear bottom 384-well black plates in replicate and incubated for 2 hours . Antigen binding to FcγRIIa ( H131/R131 ) , FcγRIIb , FcγRIIIa ( F158/V158 ) , FcγRIIIb ( NA2 ) and C1q was detected by incubating the tetrameric PE-conjugated reagents ( as described elsewhere [57] ) for an hour . HIVIG was used as a positive control and to track plate-to-plate variation . CXCL13 was detected in plasma using the Human CXCL13 Quantikine ELISA Kit ( R&D Systems ) as described by the manufacturer . PBMCs from selected individuals were thawed in the presence of Benzonase Nuclease ( Novagen , Madison , WI ) and rested overnight at 37°C . Half a million cells in 200μl of RPMI-10 media was stimulated with 0 . 5uM TLR9 agonist ODN-2006 ( Invivogen , San Diego , CA ) for 3 days at 37°C . A total of 1–3 million cells for both the stimulated and unstimulated controls were pooled after 3 days for flow cytometric staining . Surface staining was performed in PBS/0 . 1% BSA including CD16 , CD14 , CD3 ( APC-CY7 ) , CD19 ( PE-CY7 ) , IgD ( FITC ) , CD38 ( PE-CY5 ) ( BD Pharmingen , Franklin Lakes , NJ ) and was followed by intracellular staining using the BD Cytofix/Cytoperm kit as per the instructions from the manufacturer for AID ( AF647 , BD ) and Ki67 ( PerCP . Cy5 , eBiosciences , San Diego , CA ) . B cells were defined as lymphocytes , single cells , live cells ( LIVE/DEAD Fixable Dead Cell Stain ) , CD3-/CD16-/CD14- ( APC-CY7 ) , CD19+ ( PE-Cy7 ) . A HIV-negative control donor PBMC sample was run in each set of experiments to ensure consistency between runs . Fc polyfunctionality Z-scores were calculated by standardising each Fc effector function ( where the mean of the function is subtracted from the individual value and divided by the standard deviation of the mean ) and then adding all the Z-scores for each function per individual . All comparisons between groups were done with non-parametric tests including Mann-Whitney U tests ( for two groups ) and Kruskal-Wallis tests with Tukey’s correction for multiple comparison and all confidence intervals were at 95% . All correlations reported are non-parametric Spearman’s correlations and all statistical analysis was done with two-sided testing with an alpha level of 0 . 05 . Univariate analysis was performed using GraphPad 6 ( GraphPad Software , Inc , La Jolla , CA ) . For the multivariate analysis , the mean added to 3 standard deviations of the mean of HIV-negative samples for HIV-specific experiments was subtracted from HIV-positive data at 6 months post-infection and was then standardised by Z-score prior to analysis . In addition , P-values of Spearman’s correlations with bNAb and no-bNAb groups was corrected for multiple comparisons by adjusting alpha levels using the false discovery rate ( where the FDR was 5% ) and Benjamini–Hochberg procedure [105] . Significant features were used in a principal components analysis ( PCA ) using JMP13 from SAS ( Cary , NC ) . No missing data was imputed . Significant features only were again used to classify individuals into groups in order to decrease overfitting and was done by random forest classification using the R package ‘randomForest’ with the number of trees set at 500 and cross-validation indicated by out-of-bag estimate . Feature weighting or importance in the classification was represented by the mean decrease of Gini importance which measures the average gain of purity by splits of a given variable . If the variable is useful it tends to give a relatively large decrease in mean Gini-gain [106] . The resulting confusion matrix was constructed by using the ‘ggplot2’ package . The model was validated by permutation testing performed in R with 100000 data shuffles . | Some HIV-infected individuals develop antibodies that are capable of neutralizing the majority of HIV strains , a highly desirable function mediated by the antibody Fab portion . While antibodies elicited by current vaccines have failed to recreate this activity , the partial protection seen in the RV144 vaccine trial has been attributed to antibody Fc-mediated effector functions such as cell killing . In this study , we found that HIV-infected individuals who show a diversified and potent Fc response early in infection were more likely to develop broadly neutralizing antibodies later on . Examination of B cell functions associated with good germinal center activity , provided evidence for a common mechanistic link between the regulation of the Fc and Fab mediated activities in these individuals . Our finding of an Fc effector function profile that arises early and predicts neutralization breadth could be used in the evaluation of vaccine candidates designed to generate neutralizing antibodies . Common immune determinants associated with both Fab and Fc function could furthermore be exploited for vaccine design to harness the full potential of HIV-specific antibodies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"hiv",
"infections",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"immunology",
"cell",
"processes",
"microbiology",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"vaccines",
"rna",
"viruses",
"infectious",
"disease",
"control",
"antibodies",
"viral",
"load",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"hiv",
"microbial",
"pathogens",
"phagocytosis",
"biochemistry",
"antibody-producing",
"cells",
"signal",
"transduction",
"cell",
"biology",
"b",
"cells",
"virology",
"physiology",
"viral",
"pathogens",
"fc",
"receptors",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"immune",
"receptors",
"viral",
"diseases",
"lentivirus",
"organisms"
] | 2018 | HIV-specific Fc effector function early in infection predicts the development of broadly neutralizing antibodies |
Identifying a protein's functional sites is an important step towards characterizing its molecular function . Numerous structure- and sequence-based methods have been developed for this problem . Here we introduce ConCavity , a small molecule binding site prediction algorithm that integrates evolutionary sequence conservation estimates with structure-based methods for identifying protein surface cavities . In large-scale testing on a diverse set of single- and multi-chain protein structures , we show that ConCavity substantially outperforms existing methods for identifying both 3D ligand binding pockets and individual ligand binding residues . As part of our testing , we perform one of the first direct comparisons of conservation-based and structure-based methods . We find that the two approaches provide largely complementary information , which can be combined to improve upon either approach alone . We also demonstrate that ConCavity has state-of-the-art performance in predicting catalytic sites and drug binding pockets . Overall , the algorithms and analysis presented here significantly improve our ability to identify ligand binding sites and further advance our understanding of the relationship between evolutionary sequence conservation and structural and functional attributes of proteins . Data , source code , and prediction visualizations are available on the ConCavity web site ( http://compbio . cs . princeton . edu/concavity/ ) .
Sequence-based functional site prediction has been dominated by the search for residue positions that show evidence of evolutionary constraint . Amino acid conservation in the columns of a multiple sequence alignment of homologs is the most common source of such estimates ( see [22] for a review ) . Recent approaches that compare alignment column amino acid distributions to a background amino acid distribution outperform many existing conservation measures [2] , [27] . However , the success of conservation-based prediction varies based on the type of functional residue sought; sequence conservation has been shown to be strongly correlated with ligand binding and catalytic sites , but less so with residues in protein-protein interfaces ( PPIs ) [2] . A variety of techniques have been used to incorporate phylogenetic information into sequence-based functional site prediction , e . g . , traversing phylogenetic trees [28] , [29] , statistical rate inference [26] , analysis of functional subfamilies [9] , [12] , and phylogenetic motifs [30] . Recently , evolutionary conservation has been combined with other properties predicted from sequence , e . g . , secondary structure and relative solvent accessibility , to identify functional sites [31] . Structure-based methods for functional site prediction seek to identify protein surface regions favorable for interactions . Ligand binding pockets and residues have been a major focus of these methods [1] , [13]–[21] . Ligsite [16] and Surfnet [14] identify pockets by seeking points near the protein surface that are surrounded in most directions by the protein . CASTp [17] , [19] applies alpha shape theory from computational geometry to detect and measure cavities . In contrast to these geometric approaches , other methods use models of energetics to identify potential binding sites [23] , [25] , [32]–[34] . Recent algorithms have focused on van der Waals energetics to create grid potential maps around the surface of the protein . PocketFinder [23] uses an aliphatic carbon as the probe , and Q-SiteFinder [25] uses a methyl group . Our work builds upon geometry and energetics based approaches to ligand binding pocket prediction , but it should be noted that there are other structure-based approaches that do no fit in these categories ( e . g . , Theoretical Microscopic Titration Curves ( THEMATICS ) [35] , binding site similarity [36] , phage display libraries [37] , and residue interaction graphs [38] ) . In contrast to sequence-based predictions , structure-based methods often can make predictions both at the level of residues and regions in space that are likely to contain ligands . Several previous binding site prediction algorithms have considered both sequence and structure . ConSurf [39] provides a visualization of sequence conservation values on the surface of a protein structure , and the recent PatchFinder [40] method automates the prediction of functional surface patches from ConSurf . Spatially clustered residues with high Evolutionary Trace values were found to overlap with functional sites [41] , and Panchenko et al . [42] found that averaging sequence conservation across spatially clustered positions provides improvement in functional site identification in certain settings . Several groups have attempted to identify and separate structural and functional constraints on residues [43] , [44] . Wang et al . [45] perform logistic regression on three sequence-based properties and predict functional sites by estimating the effect on structural stability of mutations at each position . Though these approaches make use of protein structures , they do not explicitly consider the surface geometry of the protein in prediction . Geometric , chemical , and evolutionary criteria have been used together to define motifs that represent known binding sites for use in protein function prediction [46] . Machine learning algorithms have been applied to features based on sequence and structure [47] , [48] to predict catalytic sites [5] , [49]–[51] and recently to predict drug targets [52] and a limited set of ligand and ion binding sites [53]–[55] . Sequence conservation has been found to be a dominant predictor in these contexts . Most similar to ConCavity are two recent approaches to ligand binding site identification that have used evolutionary conservation in a post-processing step to rerank [1] or refine [56] geometry based pocket predictions . In contrast , ConCavity integrates conservation directly into the search for pockets . This allows it to identify pockets that are not found when considering structure alone , and enables straightforward analysis of the relationship between sequence conservation , structural patterns , and functional importance .
For simplicity of exposition , we begin by comparing ConCavity's performance to a representative structural method and a representative conservation method . We use Ligsite+ as the representative structure-based method , and refer to it as “Structure” . Ligsite+ is our implementation ( as indicated by superscript “+” ) of a popular geometry based surface pocket identification algorithm . We demonstrate in the Methods section that Ligsite+ provides a fair representation of these methods . We choose Jensen-Shannon divergence ( JSD ) to represent conservation methods and refer to it as “Conservation . ” JSD has been previously shown to provide state-of-the-art performance in identifying catalytic sites and ligand binding sites [2] . We have developed three versions of ConCavity that integrate evolutionary conservation into different surface pocket prediction algorithms ( Ligsite [16] , Surfnet [14] , or PocketFinder [23] ) . When the underlying algorithm is relevant , we refer to these versions as ConCavityL , ConCavityS , and ConCavityP . However , for simplicity , we will use ConCavityL as representative of these approaches and call it “ConCavity . ” ConCavity and Structure produce predictions of ligand binding pockets and residues . The pocket predictions are given as non-zero values on a regular 3D grid that surrounds the protein; the score associated with each grid point represents an estimated likelihood that it overlaps a bound ligand atom . Similarly , each residue in the protein sequence is assigned a score that represents its likelihood of contacting a bound ligand . Conservation only makes residue-level predictions , because it does not consider protein structure . All methods are evaluated on 332 proteins from the non-redundant LigASite 7 . 0 dataset [24] . To evaluate pocket identification performance , we predict ligand locations on the the holo version of the dataset , in order to use the bound ligands' locations as positives . When evaluating residue predictions , we predict ligand binding residues on the apo structures , and the residues annotated as ligand binding ( as derived from the holo structures ) are used as positives . We quantify the overall performance of each method's predictions in two ways . First , for both pocket and residue prediction , we generate precision-recall ( PR ) curves that reflect the ability of each method's grid and residue scores to identify ligand atoms and ligand binding residues , respectively . ( Just as residues are assigned a range of ligand binding scores , grid points in predicted pockets get a range of scores , since there may be more evidence that a ligand is bound in one part of a pocket than another . ) Second , for each set of predicted pockets ( corresponding to groups of non-zero values in the 3D grid ) , we consider how well they overlap known ligands via the Jaccard coefficient . The Jaccard coefficient captures the tradeoff between precision and recall by taking the ratio of the intersection of the predicted pocket and the actual ligand over their union . The Jaccard coefficient ranges between zero and one , and a high value implies that the prediction covers the ligand well and has a similar volume . We assess the significance of the difference in performance of methods on the dataset with respect to a given statistic via the Wilcoxon rank-sum test . Figure 1 compares ConCavity with its constituent structure and conservation based components . Figure 1A shows that , within predicted pockets , grid points with higher scores are more likely to overlap the ligand , and that the significant improvement of ConCavity over Structure ( p<2 . 2e−16 ) exists across the range of score thresholds . Figure 1B demonstrates that the superior performance of ConCavity holds when predicting ligand binding residues as well ( p = 6 . 80e−13 ) . ConCavity's ability to identify ligand binding residues is striking: across this diverse dataset , the first residue prediction of ConCavity will be in contact with a ligand in nearly 80% of proteins . ConCavity also maintains high precision across the full recall range: precision of 65% at 50% recall and better than 30% when all ligand-binding residues have been identified . As mentioned above , this large improvement exists when predicting ligand locations as well; however , the PR curves illustrate that fully identifying a ligand's position is more difficult for each of the methods than finding all contacting residues . The ligand overlap statistics presented in Table 1 also demonstrate the superior performance of ConCavity . In nearly 95% of structures , ConCavity's predictions overlap with a bound ligand . Structure's predictions overlap ligands in nearly 92% of the proteins considered . The differences between the methods become more stark when we examine the magnitude of these overlaps . Both ConCavity and Structure predict pockets with total volume ( Prediction Vol . ) similar to that of all relevant ligands ( Ligand Vol . ) , but ConCavity's pockets overlap a larger fraction of the ligand volume . Thus ConCavity has a significantly higher Jaccard coefficient ( p<2 . 2e−16 ) . This suggests that the integration of sequence conservation with structural pocket identification results in more accurate pockets than when using structural features alone . Figure 1B also provides a direct comparison of ligand binding site prediction methods based on sequence conservation with those based on structural features . Structure outperforms Conservation , a state-of-the-art method for estimating sequence conservation . Protein residues can be evolutionarily conserved for a number of reasons , so it is not surprising that Conservation identifies many non-ligand-binding residues , and thus , does not perform as well as Structure . Figures 2 and 3 present pocket and residue predictions of Conservation , Structure , and ConCavity on three example proteins . In general , different types of positions are predicted by Conservation and Structure . If we consider the number of known ligand binding residues for each protein in the dataset , and take this number of top predictions for the Structure and Conservation methods , the overlap is only 26% . The residues predicted by sequence conservation are spread throughout the protein ( Figure 2 ) ; ligand-binding residues are often very conserved , but many other positions are highly conserved as well due to other functional constraints . In contrast , the structure-based predictions are strongly clustered around surface pockets ( Figure 3 , left column ) ; many of these residues near pockets are not evolutionarily conserved . However , these features provide largely complementary information about importance for ligand binding . Over the entire dataset , 68% of residues predicted by both Conservation and Structure are in contact with ligands , while only 16% and 43% of those predicted by only conservation or structure respectively are ligand binding . ConCavity takes advantage of this complementarity to achieve its dramatic improvement; it gives high scores to positions that show evidence of both being in a well-formed pocket and being evolutionarily conserved . The examples of Figures 2 and 3 illustrate this and highlight several common patterns in ConCavity's improved predictions . For 3CWK , a cellular retinoic acid-binding protein , Structure and ConCavity's residue predictions center on the main ligand binding pocket ( Figure 3A ) , while Conservation gives high scores to some positions in the binding site , but also to some unrelated residues ( Figure 2A ) . Looking at the ligand location predictions ( green meshes in Figure 3A ) , Structure and ConCavity both find the pocket , but the signal from conservation enables ConCavity to more accurately trace the ligand's location . This illustrates how the pattern of functional conservation observed at the protein surface influences the shape of the predicted pocket . Ligands often do not completely fill surface pockets; if the contacting residues are conserved , our approach can suggest a more accurate shape . The results for 2CWH ( Figure 3B ) and 1G6C ( Figure 3C ) demonstrate that ConCavity can predict dramatically different sets of pockets than are obtained when considering structure alone . In 2CWH , both methods identify the ligands , but Structure over-predicts the bottom left binding pocket and predicts an additional pocket that does not have a ligand bound . ConCavity traces the ligands more closely and does not predict any additional pockets . Structure performs quite poorly on the tetramer 1G6C: it predicts several pockets that do not bind ligands; it fails to completely identify several ligands; and it misses one ligand entirely . In stark contrast , ConCavity's four predicted pockets each accurately trace a ligand . The incorporation of conservation resulted in the accurate prediction of a pocket in a region where no pocket was predicted using structure alone . Images of predictions for all methods on all proteins in the dataset are available in the Text S1 file , and ConCavity's predictions for all structures in the Protein Quaternary Structure ( PQS ) database are available online . We now compare the performance of ConCavity to several existing ligand binding site identification methods with publicly available web servers . LigsiteCS [1] is an updated version of geometry-based Ligsite , and LigsiteCSC [1] is a similar structural method that considers evolutionary conservation information . Q-SiteFinder [25] estimates van der Waals interactions between the protein and a probe in a fashion similar to PocketFinder . CASTp [19] is a geometry-based algorithm for finding pockets based on analysis of the protein's alpha shape . Each of the servers produces a list of predicted pockets represented by sets of residues; however , none of them provide a full 3D representation of a predicted pocket . As a result , we assess their ability to predict ligand binding residues . See the Methods section for more information on the generation and processing of the servers' predictions . In brief , the residues predicted by each server are ranked according to the highest ranking pocket to which they are assigned , i . e . , all residues from the first predicted pocket are given a higher score than those from the second and so on . We re-implemented the conservation component of LigsiteCSC , because the conservation-based re-ranking option on the web server did not work for many of the proteins in our dataset . We used JSD as the conservation scoring method . Figure 4 presents the ligand binding residue PR-curves for each of these methods . ConCavity significantly outperforms LigsiteCS , LigsiteCSC+ , Q-SiteFinder , and CASTp ( p<2 . 2e−16 for each ) . Surprisingly , Conservation is competitive with these structure-based approaches . Several of the servers did not produce predictions for a small subset of the proteins in the database , e . g . , the Q-SiteFinder server does not accept proteins with more than 10 , 000 atoms . Figure 4 is based on 234 proteins from the LigASite dataset for which were able to obtain and evaluate predictions for all methods . Thus the curve for ConCavity is slightly different than those found in the other figures , but its performance does not change significantly . LigsiteCSC+ is the previous method most similar to ConCavity; it uses sequence conservation to rerank the pockets predicted by LigsiteCS . LigsiteCSC+ provides slight improvement over LigsiteCS , but the improvement is dwarfed by that of ConCavity over Structure ( Figure 1 ) . This illustrates the benefit of incorporating conservation information directly into the search for pockets in contrast to using conservation information to post-process predicted pockets . The poor performance of these previous methods at identifying ligand binding residues is due in part to the fact that they do not distinguish among the residues near a predicted binding pocket . The entire pocket is a useful starting place for analysis , but many residues in a binding pocket will not actually contact the ligand . Knowledge of the specific ligand binding residues is of most interest to researchers . The predictions of our methods reflect this---residues within the same pocket can receive different ligand binding scores . The inability of previous methods to differentiate residues in a pocket from one another is one reason why we elect to use our own implementations of previous structure-based methods as representatives of these approaches in all other comparisons . See the Methods section for more details . We tested an additional approach for combining sequence conservation with structural information that was suggested by the observation that clusters of conserved residues in 3D often overlap with binding sites [41] , [42] . Briefly , the method performs a 3D Gaussian blur of the conservation scores of each residue , and assigns each residue the maximum overlapping value . Thus residues nearby in space to other conserved residues get high scores . This approach improved on considering conservation alone , but was not competitive with ConCavity ( Text S1 ) . We also considered the clusters of conserved residues generated by the Evolutionary Trace ( ET ) Viewer [57] . The clusters defined at 25% protein coverage were ranked by size , and residues within the clusters were ranked by their raw ET score . This approach did not perform as well as the above clustering algorithm ( data not shown ) , and was limited to single chain proteins , because ET returns predictions for only one chain of multi-chain proteins . In the previous sections , we used ConCavityL , which integrates evolutionary sequence conservation estimates from the Jensen-Shannon divergence ( JSD ) into Ligsite+ , to represent the performance of the ConCavity approach . However , our strategy for combining sequence conservation with structural predictions is general; it can be used with a variety of grid-based surface pocket identification algorithms and conservation estimation methods . Figure 5 gives PR-curves that demonstrate that ConCavity provides excellent performance whether the structural approaches are based on geometric properties ( Ligsite+ , Surfnet+ ) or energetics ( PocketFinder+ ) . The significant improvement holds for predicting both ligand locations in space ( p<2 . 2e−16 for each pair ) ( Figure 5A ) and ligand binding residues ( p = 6 . 802e−13 for Ligsite+ , p<2 . 2e−16 for PocketFinder+ , p<2 . 2e−16 for Surfnet+ ) ( Figure 5B ) . The three ConCavity versions perform similarly despite the variation in performance between Ligiste+ , Surfnet+ , and Pocketfinder+ . In the following sections we will include performance statistics for all three methods when space and clarity allow . When not presented here , results for all methods are available in the supplementary file Text S1 . We have also found that ConCavity achieves similar performance when a different state-of-the-art method [26] is used to score evolutionary sequence conservation ( Text S1 ) . Proteins consisting of multiple subunits generally have more pockets than single-chain proteins due to the gaps that often form between chains . To investigate the effect of structural complexity on performance , we partitioned the dataset according to the number of chains present in the structure predicted by the Protein Quaternary Structure ( PQS ) server [58] and performed our previous evaluations on the partitioned sets . Figure 6 gives these statistics for ConCavity , Structure , and Conservation . To enable side-by-side comparison , we report the area under the PR curves ( PR-AUC ) rather than giving the full curves . As the number of chains in the structure increases , there is a substantial decrease in the performance of Structure . The pattern is seen both when predicting ligand binding residues ( Figure 6A ) and pockets ( Figure 6B , C ) . This effect is so large that , for proteins with five or more chains , Conservation outperforms Structure . The number of chains in the protein has little effect on Conservation's performance . The performance of Random on proteins with a small number of chains is slightly worse than on proteins with many chains ( e . g . , Residue PR-AUC for 1 chain: 0 . 097 , 2 chains: 0 . 110 , 3 chains: 0 . 127 , 4 chains: 0 . 119 , 5+ chains: 0 . 142 ) , so the drop in Structure's performance is not the result of the proportion of positives in each set . These observations emphasize the importance of including multi-chain proteins in the evaluation . The homo-tetramer 1G6C in Figure 3C provides an illustrative example of the failure of Structure on multi-chain proteins . There is a large gap between the chains in the center of the structure , and several additional pockets are formed at the interface of pairs of contacting chains . As seen in the figure , the large central cavity does not bind a ligand; however , it is the largest pocket predicted by Structure . This is frequently observed among the predictions . While some pockets between protein chains are involved in ligand binding , many of them are not . As the number of chains increases , so does the number of such potentially misleading pockets . By incorporating sequence conservation information , ConCavity accurately identifies ligand binding pockets in multi-chain proteins . The conservation profile on the surface of 1G6C provides a clear example of this; the pockets that exhibit sequence conservation are those that bind ligands ( Figure 2C ) . 1G6C is not an exception . ConCavity provides significant performance improvement for each partition of the dataset in all three evaluations , and greatly reduces the effect of the large number of non-ligand-binding pockets in multi-chain proteins on performance . ConCavity also provides improvement over Structure on the set of one chain proteins . This is notable because these proteins do not have between-chain gaps , so the improvement comes from tracing ligands and selecting among intra-chain pockets more accurately than using structural information alone ( as in Figure 3A ) . The binding of a ligand induces conformational changes to a protein [59] . As a result , the 3D structure of the binding site can differ between structures of the same protein with a ligand bound ( holo ) and not bound ( apo ) . In the holo structures , the relevant side-chains are in conformations that contact the ligand , and this often defines the binding pocket more clearly than in apo structures . To investigate the effect of the additional information provided in holo structures on performance , we evaluated the methods on both sets ( Table 2 ) . As expected , all methods performed better on the holo ( bound ) structures than the corresponding apo ( unbound ) structures . However , all previous conclusions hold whether considering apo structures or holo structures; the ranking of the methods is consistent , and the improvement provided by considering conservation is similarly large . PR curves for this comparison are given in the supplementary file Text S1 . We will continue to report residue prediction results computed using the apo structures when possible in order to accurately assess the performance of the algorithms in the situation faced by ligand binding site prediction methods in the real world . The LigASite apo dataset contains protein molecules that carry out a range of different functions . Enzymes are by far the most common; they make up 254 of the 332 proteins in the dataset . The remaining 78 non-enzyme ligand binding proteins are involved in a wide variety of functions , e . g . , transport , signaling , nucleic acid binding , and immune system response . Table 3 compares the performance of the ligand binding site prediction methods on enzymes and non-enzymes . There is more variation within each method's performance on non-enzyme proteins , and all methods perform significantly better on the enzymes ( e . g . , p = 3 . 336e−4 for ConCavityL ) . Active sites in enzymes are usually found in large clefts on the protein surface and consistently exhibit evolutionary sequence conservation [60] , [61] , so even though enzymes bind a wide array of substrates , these common features may simplify prediction when compared to the variety of binding mechanisms found in other proteins . Despite the drop in performance on non-enzyme proteins , the main conclusions from the earlier sections still hold . However , the improvement provided by ConCavity is not as great on the non-enzymes . This could be the result of the more complex patterns of conservation found in non-enzyme proteins , and the comparatively poor performance of Conservation in this setting . It is also possible that Ligsite+'s approach is particularly well suited to identifying binding sites in non-enzymes . Overall , these results highlight the importance of using a diverse dataset to evaluate functional site predictions . Knowledge of small molecule binding sites is of considerable use in drug discovery and design . Many of the techniques used to screen potential targets , e . g . , docking and virtual screening , are computationally intensive and feasible only when focused on a specific region of the protein surface . Structure based surface cavity identification algorithms can guide analysis in such situations [52] . To test ConCavity's ability to identify drug binding sites , we evaluated it on a set of 98 protein-drug complexes [62] . The superior performance provided by ConCavity over Structure on the diverse set of proteins considered above suggests that ConCavity would likely be useful in the drug screening pipeline . Table 4 compares the ligand overlap PR-AUC and Jaccard coefficient for the three versions of ConCavity and their structure-based analogs . Each ConCavity method significantly improves on the methods that only consider structural features ( e . g . , p = 1 . 25e−6 on overlap PR-AUC and p = 2 . 06e−6 on Jaccard for ConCavityL ) . While the improvement is not quite as large on this dataset as that seen on the more diverse LigASite dataset , it is still significant . It is possible that this is due to the fact that drug compounds are not the proteins' natural ligands; the evolutionary conservation of the residues in binding pockets may reflect the pressures related to binding the actual ligands rather than the drugs . While ConCavity signficantly outperforms previous approaches , its performance is not flawless . In Figure 7 , we give three example structures that illustrate patterns observed when ConCavity performs poorly . Handling these cases is likely to be important for further improvements in ligand binding site prediction . The first pattern common among these difficult cases is evolutionary sequence conservation information leading predictions away from actual ligand binding sites . Figure 7A provides an example in which the ligand binding site is less conserved than other parts of the protein . The ActR protein from Streptomyces coelicolor ( PDB: 3B6A ) contains both a small molecule ligand-binding and a DNA-binding domain [63] . The ligand-binding domain is in the bottom , less-conserved half of the structure . The DNA-binding domain is found in the more conserved top half of the given structure . The greater conservation of this domain causes ConCavity to focus on the DNA-binding site over the ligand binding site . In other cases , conservation information is uninformative due to a lack of homologous sequences . Conservation estimates based on low quality sequence alignments may harm performance for some structures , but we have found that they still provide a net performance gain overall ( Text S1 ) . Figure 7 also provides two examples of another difficult case: ligands bound outside of clearly defined , concave surface pockets . In Figure 7B , ConCavity identifies the center of the ring-shaped structure of the pentameric B-subunit of a shiga-like toxin ( PDB: 1CQF ) as the binding site . This protein binds to glycolipids , like the globotriaosylceramide ( Gb3 ) shown , via a relatively flat interface that surrounds the center of the ring [64] . The center cavity ( ConCavity's prediction ) is filled by a portion of the A-subunit of the toxin ( not included in the structure ) which after binding breaks off and enters the host cell . Figure 7C shows the structure of a dimeric noncatalytic carbohydrate binding module ( CBM29 ) from Piromyces equi complexed with mannohexaose ( PDB: 1GWL ) . The carbohydrate ligands bind in long flat clefts on the protein surface [65] . Even though these sites exhibit significant evolutionary conservation , their geometry prevents them from being predicted . Instead , a less conserved pocket formed between the chains is highlighted by ConCavity . Overall , cases such as these are rare; ConCavity's predictions fail to overlap a ligand in only 5% of structures . In addition , some of these “incorrect” predictions are actually functionally relevant binding sites for other types of interactions as illustrated in Figure 7 . Ligand-binding sites are not the only type of functional site of interest to biologists . A large amount of attention has been given to the problem of identifying catalytic sites . As noted above , the majority of enzyme active sites are found in large clefts on the protein surface , so even though the structural methods considered in this paper were not intended to identify catalytic sites , they could perform well at this task . Table 5 gives the results of an evaluation of the methods' ability to predict catalytic sites ( defined by the Catalytic Site Atlas [66] ) in the LigASite apo dataset . Compared to ligand binding site prediction , the relative performance of the methods is different in this context . The ConCavity approach still significantly outperforms the others ( p<2 . 2e−16 for Structure , p = 8 . 223e−4 for Conservation ) . Most surprisingly , Conservation significantly outperforms methods based on structure alone ( p = 9 . 863e−3 Ligsite+ , p = 4 . 694e−6 Pocketfinder+ , p = 1 . 171e−6 Surfnet+ ) . All the methods have lower PR-AUC when predicting catalytic sites than predicting ligand-binding residues ( e . g . , ConCavityL has PR-AUC of 0 . 315 versus 0 . 608 ) ; this is due in large part to the considerably smaller number of catalytic residues than ligand-binding residues per protein sequences . These results imply that being very evolutionarily conserved is more indicative of a role in catalysis than being found in a surface pocket . Though catalytic sites are usually found in pockets near bound ligands , there are many fewer catalytic sites per protein than ligand-binding residues . As a result simply searching for residues in pockets identifies many non-catalytic residues . This is consistent with earlier machine learning studies that found conservation to be a dominant predictive feature [5] , [49] , [50] , and it suggests that new structural patterns should be sought to improve the identification of catalytic sites . Several previous methods have combined sequence conservation and structural properties in machine learning frameworks to predict catalytic sites [5] , [50] , [51] . Direct comparison with these methods is difficult because most datasets and algorithms are not readily available . Tong et al . [51] compared the precision and recall of several machine learning methods on different datasets in an attempt to develop a qualitative understanding of their relative performance . While it is not prudent to draw conclusions based on cross-dataset comparisons , we note for completeness that ConCavity's catalytic site predictions the diverse LigASite dataset achieve higher precision ( 23 . 8% ) at full recall than the maximum precision ( over all recall levels ) reported for methods in their comparisons .
Evolutionary sequence conservation and protein 3D structures have commonly been used to identify functionally important sites; here , we integrate these two approaches in ConCavity , a new algorithm for ligand binding site prediction . By evaluating a range of conservation and structure-based prediction strategies on a large , diverse dataset of ligand binding sites , we establish that structural approaches generally outperform sequence conservation , and that by combining the two , ConCavity outperforms conservation-alone and structure-alone on about 95% and 70% of structures respectively . Overall , ConCavity's first predicted residue contacts a ligand in nearly 80% of the apo structures examined , and it maintains high precision across all recall levels . These results hold for the three variants of ConCavity we considered , each of which uses a different underlying structure-based component . In addition , ConCavity's integrated approach provides significant improvement over conservation and structure-based approaches on the common task of identifying drug binding sites . Combining sequence conservation-based methods with structure information is especially powerful in the case of multi-meric proteins . Our analysis has shown that the performance of structural approaches for identifying ligand binding sites dramatically decreases as the number of chains in the structure increases; conservation alone outperforms structure-based approaches on proteins with five or more chains . It is difficult to determine from structural attributes alone if a pocket formed at a chain interface binds a ligand or not . However , ligand binding pockets usually exhibit high evolutionary sequence conservation . ConCavity , which takes advantage of this complementary information , performs very well on multi-chain proteins; the presence of many non-ligand binding pockets between chains has little effect on its performance . While ConCavity outperforms previous approaches , we have found two main causes of poor results: misleading evolutionary sequence conservation information and ligands that bind partially or entirely outside of well-defined concave surface pockets . Ligand binding sites may lack strong conservation for a number of reasons: the underlying sequence alignment may be of low quality , there may be other more conserved functional regions in the protein , and some sites are hypervariable for functional reasons [67] . The alignment quality issue will become less relevant as sequence data coverage and conservation estimation methods improve . The second two cases may require the integration of additional features to better distinguish different types of functional sites . Similarly , finding biologically relevant ligands that bind outside of concave surface pockets will likely require the development of additional structural descriptors . Missing or incomplete ligands also affect the apparent performance of the methods , but such issues are unavoidable due to the nature of the structural data . In implementing and evaluating previous 3D grid-based ligand binding site prediction approaches , we have found that the methods used both to aggregate grid values into coherent pockets as well as to map these pockets onto surface residues can have a large effect on performance . In order to focus on the improvement provided by considering evolutionary sequence conservation , the results for previous structure-based methods presented above use our new algorithms for these steps . We describe the details of our approaches in the Methods section . On a high level , the new methodologies we propose provide significant improvement by predicting a flexible number of well-formed pockets for each structure and by assigning each residue a likelihood of binding a ligand based on its local environment rather than on the rank of the entire pocket . We have used morphological properties of ligands to guide pocket creation , but the most appropriate algorithms for these steps depend strongly on the nature of the prediction task . These steps have received considerably less attention than computing grid values; our results suggest that they should be given careful consideration in the future . We have focused on the prediction of ligand binding sites , but the direct synthesis of conservation and structure information is likely to be beneficial for predicting other types of functionally important sites . Our application of ConCavity to catalytic site prediction illustrates the promise and challenges of such an approach . Catalytic sites are usually found in surface pockets , but considering structural evidence alone performs quite poorly---worse than sequence conservation . Combining structure with evolutionary conservation provides a modest gain in performance over conservation alone . Protein-protein interface residues are another appealing target for prediction; much can be learned about a protein by characterizing its interactions with other proteins . However , protein-protein interaction sites provide additional challenges; they are usually large , flat , and often poorly conserved [68] . ConCavity is not appropriate for this task . Other types of functional sites also lack simple attributes that correlate strongly with functional importance . Analysis of these sites' geometries , physical properties , and functional roles will produce more accurate predictors , and may also lead to new insights about the general mechanisms by which proteins accomplish their molecular functions . In summary , this article significantly advances the state-of-the-art in ligand binding site identification by improving the philosophy , methodology , and evaluation of prediction methods . It also increases our understanding of the relationship between evolutionary sequence conservation , structural attributes of proteins , and functional importance . By making our source code and predictions available online , we hope to establish a platform from which the prediction of functional sites and the integration of sequence and structure data can be investigated further .
This section describes the components of the ConCavity algorithm for predicting ligand binding residues from protein 3D structures and evolutionary sequence conservation . ConCavity proceeds in three conceptual steps: grid creation , pocket extraction , and residue mapping ( Figure 8 ) . First , the structural and evolutionary properties of a given protein are used to create a regular 3D grid surrounding the protein in which the score associated with each grid point represents an estimated likelihood that it overlaps a bound ligand atom ( Figure 8A ) . Second , groups of contiguous , high-scoring grid points are clustered to extract pockets that adhere to given shape and size constraints ( Figure 8B ) . Finally , every protein residue is scored with an estimate of how likely it is to bind to a ligand based on its proximity to extracted pockets ( Figure 8C ) . Grid-based strategies have been employed by several previous systems for ligand binding site prediction ( e . g . , [14] , [16] , [23] ) . However , our adaptations to the three steps significantly affect the quality of predictions . First , we demonstrate how to integrate evolutionary information directly into the grid creation step for three different grid-based pocket prediction algorithms . Second , we introduce a method that employs mathematical morphology operators to extract well-shaped pockets from a grid . Third , we provide a robust method for mapping grid-based ligand binding predictions to protein residues based on Gaussian blurring . The details of these three methods and an evaluation of their impacts on ligand-binding predictions are described in the following subsections . We have compared ConCavity to several methods for ligand binding site prediction . Many of these methods lack publicly accessible implementations , and those that are available output different representations of predicted pockets and residues . In this section , we describe of how we generate predictions for all previous methods considered in our evaluations . In some cases we have completely reimplemented strategies and in others we have post-processed the output of existing implementations . Table 8 provides a summary of these details . As mentioned earlier , a “+” appended to the method name indicates that it is ( at least in part ) our implementation , e . g . , Ligsite+ . The prediction methods described in this paper take protein 3D structures and/or multiple sequence alignments as input . Protein structures were downloaded from the Protein Quaternary Structure ( PQS ) server [58] . Predicted quaternary structures were used ( rather than the tertiary structures provided in PDB files ) so as to consider pockets and protein-ligand contacts for proteins in their biologically active states . All alignments come from the Homology-derived Secondary Structure of Proteins ( HSSP ) database [70] . All images of 3D structures were rendered with PyMol [71] . Ligand binding sites as defined by the non-redundant version of the LigASite dataset ( v7 . 0 ) [24] were used to evaluate method predictions . This set consists of 337 proteins with apo ( unbound ) structures , each having less than 25% sequence identity with any other protein in the set . Five of the 337 structures were left out of the evaluation: 1P5T , 1YJG , and 3DL3 lacked holo ligand information in the database , and 2PCY and 3EZM , because their corresponding holo structures are not in PQS or HSSP . Each apo structure has at least one associated holo ( bound ) structure in which biologically relevant ligands are identified in order to define ligand binding residues and map them to the apo structure . If multiple holo structures are available for the protein , the sets of contacting residues are combined to define the binding residues for the apo structure . We select the structures for our LigASite holo evaluation set by taking the holo structure with the most ligand contacting residues for each apo structure . The average number of holo structures for each apo structure is 2 . 58 , and the maximum for any single structure is 32 . The average chain length is 276 residues with a minimum of 59 and a maximum of 1023 . The average number of positives---sites contacting a biologically relevant ligand---per chain is 25 residues ( about 11% of the chain ) . The apo dataset includes many proteins with multiple chains; the average number of chains per protein is 2 . 22 . The chain distribution is: 1 chain: 143 , 2 chains: 112 , 3 chains: 18 , 4 chains: 35 , 5 or more chains: 24 . The drug dataset comes from a set of 100 non-redundant 3D structures selected by [62] . This set contains a diverse set of high-quality structures ( resolution <3 Å ) with drug or drug-like molecules ( molecular weight between 200 and 600 , and 1−12 rotatable bonds ) bound . Structure 1LY7 has been removed from the PDB , and 1R09 could not be parsed . We consider the 98 remaining structures . The catalytic site annotations were taken from version 2 . 2 . 9 of the Catalytic Site Atlas [66] . There are 153 proteins in the LigASite apo dataset with entries in the Catalytic Site Atlas . These proteins have an average of 3 . 2 catalytic sites per chain ( just over 1% of all residues in the chain ) . Predictions of ligand binding pockets are represented by non-zero values in a regular 3D grid around the protein . These represent regions in space thought to contain ligands . These predictions are evaluated in two ways: on the pocket level by computing their overlap with known ligands , and on the grid level by analyzing how well the grid scores rank grid points that overlap ligand atoms . We use a grid with rasterized van der Waals spheres for ligand atoms from the PQS structure as the “positive” set of grid points . From this , we calculate the intersection and union of the actual ligand atoms and the predictions . We compare methods using the over-prediction factor ( Prediction Volume/Ligand Volume ) , precision ( Intersection Volume/Prediction Volume ) , recall ( Intersection Volume/Ligand Volume ) , and Jaccard coefficient ( Intersection Volume/Union Volume ) . We also create precision-recall ( PR ) curves , which compare precision ( TP/ ( TP + FP ) ) on the y-axis with recall ( TP/ ( TP + FN ) ) on the x-axis , to evaluate the ability of each method to predict whether a ligand atom is present at a grid point . We consider grid points that overlap a ligand atom as positives . To construct the PR curve , we calculate the precision and recall at each cutoff of the grid values in the pocket prediction grid . To summarize the performance of each method , we construct a composite PR curve [72] by averaging the precision at each recall level for each structure in the dataset . As a reference point , we include the performance of a random classifier averaged over all the structures as well . The expected performance of a random method is the number of positives over the number of all grid points . The method and code of Davis and Goadrich [73] is used to calculate the area under the PR curve ( PR-AUC ) . The significance of the difference between methods is assessed using the Wilcoxon signed-rank test over paired performance statistics for all structures in the dataset . The significance of the difference in performance of a single method on different datasets is calculated with the Wilcoxon rank-sum test . For the residue-based evaluation , we consider how well each method's residue scores identify ligand binding residues . Positives are those residues in contact with a ligand as defined by LigASite database . PR curves were made by calculating , for each chain , the precision and recall at each position on the ranked list of residue scores . Composite PR curves were computed as described for the grid point evaluation , but curves were first averaged over the chains in a structure and then over structures . PR curves were constructed similarly for the catalytic site analysis , but positives were defined as those residues listed in the Catalytic Site Atlas . | Protein molecules are ubiquitous in the cell; they perform thousands of functions crucial for life . Proteins accomplish nearly all of these functions by interacting with other molecules . These interactions are mediated by specific amino acid positions in the proteins . Knowledge of these “functional sites” is crucial for understanding the molecular mechanisms by which proteins carry out their functions; however , functional sites have not been identified in the vast majority of proteins . Here , we present ConCavity , a computational method that predicts small molecule binding sites in proteins by combining analysis of evolutionary sequence conservation and protein 3D structure . ConCavity provides significant improvement over previous approaches , especially on large , multi-chain proteins . In contrast to earlier methods which only predict entire binding sites , ConCavity makes specific predictions of positions in space that are likely to overlap ligand atoms and of residues that are likely to contact bound ligands . These predictions can be used to aid computational function prediction , to guide experimental protein analysis , and to focus computationally intensive techniques used in drug discovery . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"computational",
"biology/macromolecular",
"structure",
"analysis",
"computer",
"science",
"molecular",
"biology/bioinformatics",
"computational",
"biology/macromolecular",
"sequence",
"analysis",
"computational",
"biology"
] | 2009 | Predicting Protein Ligand Binding Sites by Combining Evolutionary Sequence Conservation and 3D Structure |
Over the last 15 years , visceral leishmaniasis ( VL ) has emerged as a public health concern in Tbilisi , the capital of Georgia . Seroepidemiological surveys were conducted to determine the prevalence and incidence of infection in children and dogs within the main focus of VL , and to identify risk factors associated with human infection . Of 4 , 250 children investigated , 7 . 3% were positive by direct agglutination test in a baseline survey; an apparent incidence rate of 6 . 0% was estimated by one year follow-up . None of the seropositive children progressed to VL during the survey . Increased seropositivity at one year was predicted by presence at baseline of clustered flying insects ( OR = 1 . 49; P = 0 . 001 ) , perceived satisfactory sanitation ( OR = 1 . 65; P<0 . 001 ) , stray dogs ( OR = 1 . 33; P = 0 . 023 ) , and by persistent fever during the 6 months prior to baseline survey ( OR = 14 . 2; P<0 . 001 ) . Overall , 18 . 2% ( 107/588 ) of domestic and 15 . 3% ( 110/718 ) of stray dogs were seropositive by the rk39 dipstick test . Clinical VL signs were found in 1 . 3% of domestic and 2 . 9% of stray , seropositive dogs . Parasites isolated from human and dog samples were identified by PCR and phylogenetic analysis of the Leishmania 70 kDa heat-shock protein ( HSP70 ) gene as Leishmania infantum . There is an active focus of L . infantum transmission in Tbilisi with a high prevalence of human and canine infections .
Visceral leishmaniasis ( VL ) caused by parasites of the Leishmania donovani complex ( L . donovani , L . infantum/L . chagasi ) is a severe disease with a fatal outcome if left untreated . It is clinically characterized by low-grade fever , enlarged spleen and liver , and weight loss . Anthroponotic VL caused by L . donovani is endemic in East African countries , Northeast India , Nepal , and Bangladesh [1] , [2] . In Latin America VL is zoonotic and the causative agent is L . chagasi [3] . In the Mediterranean basin zoonotic VL is caused by L . infantum , with incidence ranging from 0 . 02/100 , 000 to 8 . 53/100 , 000 [4] . Dogs are recognized as the primary reservoirs of zoonotic VL [5] and the prevalence of canine VL in Mediterranean countries varies from 1 . 1% to 48 . 4% [4] , [6] . Zoonotic VL is widely distributed in countries of the former Soviet Union . Human cases are registered in Middle Asia and the Caucasus , including the Republic of Georgia [7] . Historically , leishmaniasis in Georgia has been sporadic and confined mainly to the eastern part of the country [8] . Current active foci are located in Tbilisi , the capital of Georgia , and in the region Shida Kartli ( Figure 1A ) [9] , [10] . Since 1990 , the number of VL cases recorded annually has increased substantially , from 10–12 cases in the early 90's , to 171 cases in 2008 . Out of 1535 patients registered in Georgia during 1995–2008 , 917 ( 60% ) were from Tbilisi , with 17 fatal cases ( official statistical records , National Center for Disease Control ( NCDC ) , Tbilisi , Georgia ) . Due to the lack of VL surveillance system in Georgia , the prevalence and incidence of human and canine infection with Leishmania has remained unknown . Consequently , an appropriate control strategy has not been formulated . In this paper we present the results of a 3-year prospective study carried out during 2006–2008 in Tbilisi with the aims to determine the prevalence and incidence of Leishmania infection in children , as well as its prevalence in domestic and stray dogs , and to confirm the identity of the Leishmania parasite responsible for the disease .
Tbilisi is located at 500–800 m above sea level and is divided into northern and southern parts by the Mtkvari River . The southern part is mostly hilly with canyons and ravines; the northern part is terraced . Of the 10 urban districts , most of the VL cases ( ∼70% ) registered during 1997–2004 originated from Krtsanisi , Mtatsminda and Vake located in the southern part of the city along the foot of the Mtatsminda Mountain . These 3 districts , representing active foci of VL , were selected as study sites ( Figure 1B ) . The populations of the districts were: Krtsanisi – 27 , 047 ( 5 , 155 children 1–14 yrs of age ) ; Mtatsminda – 43 , 133 ( 8 , 914 ) ; Vake – 27 , 053 ( 6 , 855 ) . Bone marrow aspirates were smeared onto microscopic slides , dried , fixed with methanol , stained with Giemsa , and examined microscopically under a 100× oil immersion objective for presence of Leishmania amastigotes . Standard procedures for human serodiagnosis by DAT [11] , [12] were performed using the freeze-dried Leishmania antigen ( Royal Tropical Institute , Amsterdam , Netherlands ) to measure the antibody titer in finger-prick blood . Briefly , blood samples were eluted from filter papers overnight in serum diluent and serially diluted in twofold dilutions from 1∶100 to 1∶51 , 200 in 96-well plates . The DAT antigen was added to each dilution and the results were read after 18 h of incubation at room temperature . To determine the cut off point for the DAT , blood samples from 100 children of the same age group living in the non-endemic Adjara region , in the city of Batumi ( western Georgia ) were tested at dilutions ranging from 1∶200 to 1∶25 , 600 . None of the Batumi samples was positive at or above 1∶6 , 400 dilutions , thus samples giving titers of ≥1∶6 , 400 were considered seropositive . This cutoff is higher than is typically used for serodiagnosis of active VL [11] , [12] , therefore the specificity of the DAT in our survey should be at least as high as seen in the previous studies ( 119/124 = 96% in a mostly endemic population [11] , 434/435 = 99 . 8% in a population with various other diseases [12] ) . Using the 1∶6 , 400 cutoff for detection of active cases in an endemic population , Oskam et al . [13] observed perfect specificity of the DAT ( 130/130 = 100% ) . Samples that produced borderline agglutination at ≥1∶6 , 400 were scored as suspected positive , although the prevalence and incidence rates were calculated from only ‘confirmed’ seropositives . The suspected positives were included in the denominators of these calculations . Blood samples from patients with confirmed diagnosis of VL provided by the Research Institute of Medical Parasitology and Tropical Medicine ( RIMPTM , Tbilisi , Georgia ) were used as positive controls , and blood samples from DAT negative healthy individuals without signs or symptoms of leishmaniasis were used as negative controls . The Kalazar Detect rapid test was performed according to the manufacturer's protocol ( InBios International , Inc . Seattle , WA , USA ) using 20 µl for each canine serum sample . The test was considered positive when a control line and test line appeared in the test area within 10 minutes , negative if only a single control line appeared , and invalid if none of the lines appeared . PCR analysis was performed to identify Leishmania parasites isolated from dogs and humans . DNA was extracted using the DNeasy Blood & Tissue Mini Kit ( QIAGEN ) from bone marrow cultures isolated from serologically positive domestic dogs and a culture established from the bone marrow of a patient with clinically confirmed VL diagnosis at the RIMPTM . The primers Uni21 ( 5′ GGG GTT GGT GTA AAA TAG GCC 3′ ) and Lmj4 ( 5′ CTA GTT TCC CGC TCC GAG 3′ ) were used for initial PCR analysis [14] . The PCR was performed in a 25 µl reaction mixture , containing 12 . 5 µl of 2× FastStart PCR Master ( Roche ) , 1 µl of each primer ( 10 µM ) , and 5 µl of template DNA ( ∼20 ng ) . Conditions for the reaction were as follows: initial denaturation at 94°C for 5 min , amplification for 35 cycles at 94°C–1 min , 60°C–1 min , 72°C–1 . 5 min , elongation at 72°C–10 min , and holding at 4°C . The PCR products were separated in 1 . 4% agarose gel , stained with ethidium bromide and visualized by ultraviolet light . L . infantum ( MHOM/ES/00/UCM-1 ) and L . major MHOM/IL/80/Friedlin reference strains were included for comparison . Further Leishmania identification was achieved by amplification of the 70 kDa heat-shock protein ( HSP70 ) gene from dog and human samples using primers HSP70sen ( 5′-GACGGTGCCTGCCTACTTCAA-3′ ) and HSP70ant ( 5′-CCGCCCATGCTCTGGTACATC-3′ ) [15] . The PCR was completed in a volume of 50 µl contained 5 µl of 10× PCR Buffer II , 1 µl of each primer ( 10 µM ) , 50 ng of template DNA , 1 µl of Taq DNA polymerase ( AccuPrime™ Taq DNA Polymerase System; Invitrogen ) and 16 µl of H2O . The samples were incubated at 94°C for 2 minutes followed by 35 cycles of 94°C for 30 seconds , 60°C for 30 seconds and 68°C for 1 . 5 minutes . The amplified 1422 bp products were separated by electrophoresis through 1 . 2% agarose gel and visualized using SYBR Safe DNA gel stain ( Invitrogen ) . The excised bands were extracted with Ultrafree-DA DNA Extraction ( Millipore ) and cleaned with three washes of ultrapure H2O through an YM-30 Microcon filter ( Millipore ) . The cleaned products were ligated into pCR 4-TOPO ( Invitrogen ) and the plasmid transformed into One Shot TOP10 competent Escherichia coli ( Invitrogen ) . Colonies were screened by PCR with HSP70sen and HSP70ant primers . Positive colonies were grown overnight and the plasmids purified with the PureLink Quick Plasmid Miniprep Kit ( Invitrogen ) . The plasmids were sequenced bidirectionally with M13R ( 5′-CAGGAAACAGCTATGACC-3′ ) and M13F ( 5′-TGTAAAACGACGGCCAGT-3′ ) . Nucleotide sequences of the HSP70 gene of Leishmania parasites from dog and human samples were aligned against published sequences for L . infantum , L . donovani , L . tropica , L . major and L . aethiopica strains [16] using Clustal X , version 1 . 83 [17] . Phylogenetic analysis was conducted on the alignments using MEGA , version 3 . 1 , generating trees by neighbor-joining and testing phylogeny with 1 , 000 replications to calculate node support [18] . Prevalence and incidence rates with confidence intervals were estimated using logistic models with generalized estimating equations ( GEE ) with a working independence assumption [19] to account for the within-family correlation . The responses for the logistic models were either the baseline DAT tests ( for prevalence ) or the one-year DAT tests only from those who were negative at baseline ( for incidence ) . These estimates make no adjustment for the fact that the DAT is an imperfect diagnostic tool . However , because we expect the specificity to be near 100% ( see DAT section above ) , and because imperfect sensitivity will only cause the apparent prevalence to be higher than the true prevalence [20] , our prevalence estimates should be interpreted as estimating the lower bounds on the true prevalence . Our incidence estimates should also be interpreted cautiously as apparent incidences , since the adjustments needed to estimate true incidence [20] require sensitivity estimates . While the DAT sensitivity studies in the literature [21] typically refer to clinical disease for which the DAT results can be readily validated by parasite detection , our interest was subclinical infection for which there is so far no proven way to confirm the presence of parasites in the blood or other tissues . We predicted seropositivity after one year using baseline disease status , region , and answers to survey questions at baseline using GEE cumulative logit models on the ordered responses: negative , suspected positive , and confirmed positive . We eliminated several survey questions that essentially duplicated other questions , and with the remaining questions we built one overall GEE model . Additionally , we built separate GEE models for each survey question using only that question plus baseline disease status and region . Odds ratios ( ORs ) are ratios of either the odds of confirmed seropositivity or the odds of suspected seropositivity [22] . ORs for yes/no questions estimate the ratio of the odds of disease for a subject who answered “yes” over the odds for a different hypothetical subject who answered “no” but is matched on all other variables in the model . Analyses were done in SAS version 9 . 1 with a p-value of <0 . 05 considered statistically significant .
We obtained baseline DAT responses on 99 . 6% of the subjects for a total of 4 , 250 children from 2 , 968 households; 1 , 459 children ( 34% ) were 1–4 years and 2 , 791 ( 66% ) were 5–14 years; 2 , 096 ( 49% ) were girls and 2 , 150 ( 51% ) were boys . The overall baseline prevalence was 7 . 3% with an incidence rate , estimated by a one year follow-up study , of 6 . 0% . The baseline seropositivity was highest in Mtatsminda ( 10 . 6% ) followed by Vake ( 6 . 3% ) and Krtsanisi ( 0 . 5% ) , while the incidence rate was highest in Vake ( 7 . 9% ) followed by Mtatsminda ( 5 . 2% ) and Krtasanisi ( 4 . 5% ) ( Table 1 ) . None of the seropositive children who were asymptomatic at baseline developed clinical signs of VL during the follow-up period , although 7% ( 296 children ) of all investigated children were seropositive during both the baseline survey and one year follow-up . Fourteen of 310 children ( 4 . 5% ) who were positive at baseline became negative in the follow-up survey . In 85 households with 2 or more children , all children were positive either in both baseline and follow up surveys , or seroconverted during the follow up period ( Figure 2 ) . Hepatosplenomegaly , jointly with other clinical signs of active disease ( weight loss , fatigue , anorexia , fever of unknown etiology ) , was detected in 2 children during clinical examination in the baseline survey . These children were excluded from the study and sent to the RIMPTM where the diagnosis of VL was parasitologically confirmed . Twenty two children with a previous history of VL were identified during the baseline study . Of these children , 11 ( 50% ) were positive by DAT both in the baseline and follow-up surveys . We predicted seropositivity at follow-up from only baseline disease status , region , and each baseline variable alone ( Table 2 , unadjusted OR ) or by a complete model with all baseline variables included ( Table 2 , adjusted OR ) . We found that by both unadjusted and adjusted methods the odds of seropositivity at follow-up were increased by clustered flying insects at sunset/sunrise , perceived satisfactory sanitary conditions , and stray dogs; odds were decreased by the presence of nearby woodlands . The leading risk factor positively associated with seropositivity , with an OR of 14 . 2 ( unadjusted ) or 13 . 6 ( adjusted ) ( p<0 . 001 for both methods ) , was fever lasting more than 2 weeks for which antibiotic therapy was not effective , and occurring during the 6 months prior to the day of the baseline interview ( Table 2 ) . No association of seropositivity by either method was found in relation to age or gender , or the use of nets on doors and windows . The use of repellents and the nearby existence of facilities for domestic animals were significant only after adjusting for the other variables . A history of lack of appetite at baseline was predictive of disease one year later but this effect disappears after adjusting for the other variables . Importantly , there was no association of seropositivity with having a pet dog ( s ) at home or in the yard by either method ( Table 2 ) . Overall , 107 of 588 domestic dogs ( 18 . 2% ) and 110 of 718 stray dogs ( 15 . 3% ) were found seropositive by the rK39 dipstick test ( Table 3 ) . Among the domestic dogs surveyed , the highest seroprevalence rate was in Vake – 31 . 2% ( 81 of 260 dogs ) , followed by Krtsanisi – 10 . 1% ( 14 of 138 ) , and Mtatsminda – 6 . 3% ( 12 of 190 ) . In the stray population , however , the highest seroprevalence rate was observed in Krtsanisi – 19 . 9% ( 36 of 181 ) , followed by 14 . 5% ( 48 of 331 ) in Vake , and 12 . 6% ( 26 of 206 ) in Mtatsminda ( Table 3 ) . Clinical signs of canine VL were found only in 1 . 9% ( 2 of 107 ) and 2 . 7% ( 3 of 110 ) of seropositive domestic and stray dogs , respectively . However , the presence of Leishmania amastigotes was confirmed by microscopy in 49 of 75 ( 65% ) bone marrow aspirates taken from seropositive domestic dogs . No statistically significant difference was observed between seropositivity and gender or age of investigated dogs . A difference in seroprevalence according to breed was observed among the 33 breeds and mongrels included in this study . A higher seropositivity was observed in Doberman Pinchers – 3 of 3 , hounds – 4 of 5 , pit bulls – 7 of 16 , European shepherds – 2 of 6 , German shepherds – 23 of 70 , Drathaars – 1 of 3; compared to Caucasus shepherds – 1 of 31 , Rottweilers – 1 of 15 , poodles – 3 of 31 , and mongrels – 31 of 232 . Amplification of DNA extracted from bone marrow of five seropositive dogs and one infected child with parasitologically confirmed VL produced single bands of approximately 800 bp identical in size to the L . infantum reference strain ( Figure 3A ) . No amplification was observed in the negative control without template DNA . Phylogenetic analysis of the 70 kDa heat-shock protein ( HSP70 ) nucleotide sequences from DNA extracted from the dog and human bone marrow confirmed the identity of the parasite as L . infantum ( Figure 3B ) .
We report results of the first large seroepidemiological study of an important , emerging focus of human and canine VL in Tbilisi . Overall , 310 ( 7 . 3% ) of 4 , 250 investigated children were found seropositive by DAT at the baseline survey , with 235 out of 3 , 896 seronegative children ( 6 . 0% ) converting to positive within a one year follow-up . The choice of DAT as the serological test was justified by its prior application in a large number of field-based epidemiologic surveys of VL [23]–[25] . Selection of the cut-off titer of ≥1∶6 , 400 was based on results obtained from an extensive survey of children living in a non-endemic region of Georgia . For comparison , DAT titers of ≥1∶3 , 200 have been used by different investigators as a cut-off for the assessment of L . donovani/L . infantum seropositivity [23] , [25]–[28] , although Harith et al . [23] suggested a single serum dilution of 1∶6 , 400 for use in mass screenings . Moreover , they compared filter paper eluates with the corresponding serum samples and found no significant difference in DAT titers [23] . The true prevalence is likely higher than 7 . 3% because of the imperfect sensitivity of the DAT at the 1∶6 , 400 cut-off . With imperfect sensitivity , some of those that have VL will not be detected . By using probability laws , and assuming the sensitivity and specificity are known , we can get an estimate of the true prevalence [20] ( see Text S1 ) : True Prevalence = ( Apparent Prevalence+Specificity−1 ) / ( Sensitivity+Specificity−1 ) . Using data on clinical VL from Oskam et al . ( 1999 ) , we estimated the specificity as 100% ( 130/130 ) and the sensitivity as 71 . 7% ( 147/205 ) , so that the corrected estimate of true prevalence using the above formula is 10 . 2% . These calculations depend on the estimates of sensitivity and specificity , and if the specificity is less than 100% then the True Prevalence estimate would lower than 10 . 2% , while if the sensitivity is lower ( which we might expect for detecting subclinical disease ) then the estimated true prevalence would be higher than 10 . 2% . Different prevalence rates of asymptomatic seropositive humans have been found in Greece ( 0 . 5% ) [6] , Turkey ( 2 . 6% ) [29] , France ( 10%–28% ) [30] , Israel ( 3 . 0% ) [31] , Iran ( 1 . 6% ) [32] , and Azerbaijan ( 8 . 0% ) ( neighboring country of Georgia ) [33] . Our data indicate the existence of a high frequency of asymptomatic human carriers in Tbilisi that conforms with results of studies reporting circulation of L . infantum in peripheral blood of asymptomatic healthy individuals [34]–[36] . All of the 310 children who were seropositive at baseline remained asymptomatic during the one year follow-up , and only 14 ( 4 . 5% ) reverted to seronegative . Reversion of seropositive asymptomatic individuals to seronegative is well described [37] , for example , in Kenya , where 75% of seropositive asymptomatic individuals became seronegative within 12–36 months without developing clinical signs of VL [38] , or in Spain , where only 50% of the L . infantum asymptomatic seropositive blood donors remained positive after one year [39] . The question of how such changes in serological titers reflect changes in parasitemia remains unresolved . Interestingly , we found clusters of seropositive children in 85 households with more than one child where all children living in the same household were either positive for anti-Leishmania antibody in both baseline and follow-up surveys , or converted to positive in the follow-up survey . None of these children developed clinical signs of VL during our study . Of these households , 33 ( 39% ) were located in Vake district and half of them ( 17 households ) on the same street; 48 households ( 56% ) were in Mtatsminda district , and only 4 households ( 5% ) in Krtsanisi district . Existence of these clusters , along with the absence of an association of seropositivity with domestic dogs , is consistent with the possibility that asymptomatic human cases serve as infection reservoirs . One important objective of our survey was identification of the potential risk factors associated with seropositivity . We found that children who had persistent fever within the 6 months prior to the interview had a much higher probability ( OR = 14 . 2 unadjusted; p<0 . 001 ) of becoming seropositive than children without this symptom . Some investigators have classified seropositive patients without any symptoms of VL as asymptomatic and seropositive patients having one or a combination of mild symptoms as subclinical . Subclinical patients can progress to the classic form of VL or resolve their symptoms over different periods of time [32] , [37] , [40] , [41] . The seropositive children in our study with no other symptoms but fever can be considered as having a subclinical form of VL , and represent a significant and under reported morbid condition associated with L . infantum infection in this population . Children living in the households that were perceived to have unsatisfactory sanitary conditions were at significantly lower risk of infection ( p<0 . 001 , both adjusted and unadjusted ) , as this perception may be associated with more attention to sanitary issues . Children living where clusters of blood-sucking insects were reported had higher risk of infection ( p = 0 . 001 unadjusted , p = 0 . 004 adjusted ) . The results of the analysis indicated no significant association of seropositivity with sex or age of the investigated population , which conforms to the data from similar studies [6] , [42] , [43] . The presence of domestic livestock in the yard or the use of screens on the doors and windows or repellents against insects also did not have a statistically significant association with seropositive children . Decreased association with seropositivity in the presence of woodlands ( p<0 . 001 , unadjusted; p = 0 . 024 , adjusted ) suggests that woodlands act as a barrier to sand fly-human contact , likely due to the presence of wild animals as a preferred blood meal source . Although Gavgani et al . [44] suggested that domestic dogs represent a significant risk factor for human VL , our findings did not show any association of seropositive children with domestic dogs , similar to other studies [35] , [43] . By contrast , our analysis showed significant correlation with the nearby presence of stray dogs ( , p = 0 . 024 , unadjusted; p<0 . 001 , adjusted ) . The increased risk of infection is associated with a high overall seroprevalence rate in our survey detected by the rK39 dipstick test for stray dogs ( 15 . 3% ) . This conforms well with the range of average seroprevalence rates of canine leishmaniasis reported from different Mediterranean countries: 20% in Portugal , 8 . 5% in Spain , 4–20% in France , 2–15% in Italy , 25% in Greece , 20% in Cyprus , 15 . 7% in Turkey [4] . It is curious that there was no association with seropositivity and the presence of domestic dogs despite the fact that the seroprevalence rate in domestic dogs was even higher ( 18 . 2% ) . A high prevalence of infected asymptomatic dogs within L . infantum endemic areas of Spain , Italy , Brazil , Iran has been a consistent finding [5] , [45]–[47]; lower rates were reported from Greece , Azerbaijan and Turkey [6] , [33] , [48] . While some authors reported that sand flies were not able to acquire infections from asymptomatic dogs [5] , others proved that transmission to the vector from asymptomatic , seropositive dogs was possible [49]–[51] . Importantly , our own studies revealed that 49 of 75 bone marrow aspirates taken from seropositive domestic dogs were positive for amastigotes by microscopy , confirming their potential as infection reservoirs . As for the risk factors associated with dogs becoming seropositive , there was no association with sex or age of dogs , or as mentioned , with domestic versus stray dogs . In contrast to our results , Gavgani et al . [44] showed significantly higher seroprevalence of L . infantum in stray ( 26 . 6% ) in comparison with domestic ( 12 . 7% ) dogs . Several studies showed no specific correlation between seropositivity and sex of dogs [47] , [52] , [53] , although age-related differences supporting higher rates in older dogs have been reported [6] , [47] . The influence of breed of dogs on seroprevalence is also inconclusive [6] , [52] , [53] , though in general , higher rates of canine VL were reported in German shepherds , Dobermans , boxers and hounds , which favorably support the results of our survey . Numerous PCR-based methods have been used for detection and identification of Leishmania parasites , including amplification of the kinetoplast DNA ( kDNA ) [14] , [54]–[58] . Using the Uni21 and Lmj4 primer pair [14] , the 800 bp amplicon was obtained from bone marrow DNA of five serologically positive dogs and one child with parasitologically confirmed VL . Though Leishmania-specific , this PCR product size is similar in L . infantum , L . donovani and L . tropica . Therefore , the identity of the Leishmania species from our study area was confirmed as L . infantum by sequence analysis of the 70 kDa heat-shock protein ( HSP70 ) , demonstrating that this species is responsible for human and canine VL in Tbilisi . Taken together , our results confirm the presence of an extremely active focus of VL in Tbilisi with a high prevalence of human and canine infections . Importantly , recent data , demonstrating the spread of human VL cases to other parts of Tbilisi and to non-endemic territories of Georgia , indicate an urgent need to formulate and implement effective control measures . | Visceral leishmaniasis ( VL ) has emerged as a public health problem in Tbilisi , the capital of Georgia . Dogs are the main infection reservoirs for transmission by sand flies of Leishmania infantum to humans , many of whom may become infected without developing disease . Since majority of cases are in children we were interested to know the rate of infection in children and in dogs living within the area where cases of VL have been found , and what factors may affect the risk of infection . Using a test that detects the presence of antibodies in blood as a marker of infection , 7 . 3% of 4 , 250 children examined were positive at the baseline survey , and 6% became positive after one year . Overall , 18 . 2% of domestic and 15 . 3% of stray dogs were seropositive . The infected children were more apt to live in areas where clustered flying insects and stray dogs were observed , and were far more likely to have experienced a persistent fever in the 6 months before the survey . We conclude that there is very active transmission of L . infantum to both humans and dogs in Tbilisi , and that children remain at high risk of developing clinical disease and sub-clinical infection . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"epidemiology",
"infectious",
"disease",
"epidemiology",
"leishmaniasis",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control"
] | 2011 | Epidemiologic Aspects of an Emerging Focus of Visceral Leishmaniasis in Tbilisi, Georgia |
Epstein-Barr virus ( EBV ) is a ubiquitous human γ-herpesvirus that can give rise to cancers of both B-cell and epithelial cell origin . In EBV-induced cancers of epithelial origin , including nasopharyngeal carcinomas ( NPCs ) and gastric carcinomas , the latent EBV genome expresses very high levels of a cluster of 22 viral pre-miRNAs , called the miR-BARTs , and these have previously been shown to confer a degree of resistance to pro-apoptotic drugs . Here , we present an analysis of the ability of individual miR-BART pre-miRNAs to confer an anti-apoptotic phenotype and report that five of the 22 miR-BARTs demonstrate this ability . We next used photoactivatable ribonucleoside-enhanced crosslinking and immunoprecipitation ( PAR-CLIP ) to globally identify the mRNA targets bound by these miR-BARTs in latently infected epithelial cells . This led to the identification of ten mRNAs encoding pro-apoptotic mRNA targets , all of which could be confirmed as valid targets for the five anti-apoptotic miR-BARTs by indicator assays and by demonstrating that ectopic expression of physiological levels of the relevant miR-BART in the epithelial cell line AGS resulted in a significant repression of the target mRNA as well as the encoded protein product . Using RNA interference , we further demonstrated that knockdown of at least seven of these cellular miR-BART target transcripts phenocopies the anti-apoptotic activity seen upon expression of the relevant EBV miR-BART miRNA . Together , these observations validate previously published reports arguing that the miR-BARTs can exert an anti-apoptotic effect in EBV-infected epithelial cells and provide a mechanistic explanation for this activity . Moreover , these results identify and validate a substantial number of novel mRNA targets for the anti-apoptotic miR-BARTs .
MicroRNAs ( miRNAs ) are 22 ± 2 nucleotide ( nt ) non-coding RNAs that are expressed by all multicellular eukaryotes as well as by several viruses [1–3] . MiRNAs are generally initially transcribed by RNA polymerase II in the form of a long primary miRNA ( pri-miRNA ) precursor that is sequentially processed by the RNase III enzymes Drosha , in the nucleus , to generate the pre-miRNA intermediate and Dicer , in the cytoplasm , to yield the mature miRNA [1 , 4] . Upon loading into the RNA-induced silencing complex ( RISC ) , the miRNA serves as a guide RNA to direct RISC to partially complementary target sites [5] . Particularly important in this regard is the miRNA seed sequence , extending from position 2 to 8 on the miRNA , which is exposed during mRNA binding by RISC and plays a key role in target mRNA recognition [5 , 6] . Because seed sequence complementarity to an mRNA target is generally not only necessary but frequently also sufficient for effective RISC recruitment , it is predicted that each miRNA functionally interacts with >100 mRNA targets . RISC binding in turn results in the translational inhibition and partial destabilization of the target mRNA [5] . The accurate identification of these mRNA targets , and more importantly , the discovery of mRNA targets that are phenotypically relevant , remains the most difficult challenge in understanding miRNA function . This is particularly difficult in the case of virally encoded miRNAs as these are subject to rapid evolution and , unlike cellular miRNA target sites , which have co-evolved with host cell miRNAs , cellular mRNA targets for viral miRNAs are generally not evolutionarily conserved . Efforts to identify important mRNA targets for viral miRNAs have therefore generally followed one of two approaches , which have been respectively referred to as the “bottom up” and “top down” approach [2] . In the “top down” approaches , the investigator first identifies a phenotype exerted by a miRNA then seeks to determine which mRNA target ( s ) is responsible for this phenotype . Conversely , in the “bottom up” approach , the investigator first uses computational methods or experimental techniques , such as microarray analysis or a cross-linking/immunoprecipitation approach , to globally identify mRNA targets for a given viral miRNA then seeks to confirm that the phenotypic effect predicted upon downregulation of a given mRNA target is actually observed . These approaches are not , of course , mutually exclusive as tools for the global identification of mRNA targets for a given viral miRNA can provide critical information for efforts to identify the mRNA target ( s ) that explain a miRNA phenotype . Epstein-Barr virus ( EBV ) encodes two miRNA clusters that are differentially expressed during latent EBV infection [7–10] . In latency III , as seen for example in lymphoblastoid cell lines ( LCLs ) of primary B-cell origin , EBV expresses a high level of the three viral pre-miRNAs encoded in the miR-BHRF1 cluster and moderate levels of the 22 pre-miRNAs encoded in the miR-BART cluster [7 , 10 , 11] . Consistent with this expression pattern , mutational inactivation of the miR-BHRF1 cluster severely impairs B-cell transformation by EBV , with the resultant LCLs showing a slow growth phenotype , while loss of all 22 miR-BARTs has at most a modest effect on B-cell transformation [12–15] . Conversely , in EBV-transformed epithelial cells that are in latency II , including nasopharyngeal carcinoma ( NPC ) cells and EBV-induced gastric carcinomas , the miR-BHRF1 cluster is not expressed while the miR-BARTs are transcribed at substantial levels [7 , 9 , 10 , 16 , 17] . Whether the miR-BART miRNAs are required for the transformation of primary human epithelial cells by EBV remains unclear , due to the lack of good in vitro systems to study this process . However , analysis using the gastric carcinoma cell line AGS strongly suggests that this is likely to be the case . AGS cells are normally EBV-negative but can be readily infected by EBV to establish a latent infection marked by high level expression of the miR-BARTs , as well as the viral EBNA1 protein and the EBER non-coding RNAs , but only very low levels of the other viral latent proteins , including LMP1 and EBNA2 [18 , 19] . Strikingly , EBV+ AGS cells show enhanced anchorage independent cell growth and the ectopic expression of the miR-BART miRNAs in AGS cells also inhibits apoptosis [18 , 20 , 21] . This latter result is consistent with a number of reports that have provided evidence for the downregulation of pro-apoptotic cellular genes by individual miR-BART miRNAs [14 , 20–24] . However , at present a full understanding of how the EBV miR-BART miRNAs inhibit apoptosis to promote the viability of EBV-infected epithelial cells remains elusive . Here , we report a systematic effort to identify pro-apoptotic mRNA targets for the EBV miR-BART miRNAs . We demonstrate that at least five of the 22 miR-BART pre-miRNAs have anti-apoptotic activity and we identify seven pro-apoptotic cellular mRNA targets , six of them novel , that contribute significantly to the observed anti-apoptotic phenotype . Together , these data represent a substantial increase in our understanding of the role of the miR-BART miRNAs in promoting EBV infection and latency .
We initially performed deep sequencing of the small RNA population ( ~18 to ~24 nt in size ) in C666 cells , as previously described [11 , 37] . This resulted in a total of 26 . 6x106 reads , of which 25 . 8x106 ( ~97% ) mapped to either the human or EBV genome ( See S2 Table for the complete data set ) . Of these , 23 . 9x106 ( ~93% ) represent known mature miRNAs or miRNA passenger strands , with 6 . 7x106 ( ~28% ) mapping to the EBV miR-BART locus and the remaining 17 . 2 x106 reads ( ~72% ) representing known human miRNAs ( Fig 4A ) . Among the miR-BARTs , we recovered reads from all 22 known miR-BART miRNA and miRNA passenger strands , but only those miRNA passenger strands representing ≥10% of the total reads derived from a given pre-miR-BART intermediate are shown in Fig 4B . The most highly expressed miR-BARTs detected in C666 cells were miR-BART2-5p , miR-BART9-3p and miR-BART19-3p . Next we performed PAR-CLIP to globally identify the mRNA targets for the miR-BARTs in C666 cells using an antibody that immunoprecipitates all four human Argonaut ( Ago ) proteins . The PAR-CLIP library gave 16 . 7x106 reads , of which 6 . 6x106 could be mapped to a unique sequence present in either the human or EBV genome ( see S3 Table for the complete data set ) . Computational definition of binding site clusters and assignment to expressed miRNAs [38] revealed that the majority of both the cellular miRNA and miR-BART clusters mapped to 3’UTRs , although significant numbers of clusters also were observed in mRNA coding sequences ( CDS ) or in intronic regions ( Fig 4C ) . Of the total number of 3’UTR miRNA binding clusters that were detected , 792 were computationally assigned to one of the EBV miR-BART miRNAs or to a miR-BART passenger strand , based on seed homology , including a moderate number of potential mRNA targets for the anti-apoptotic miRNAs miR-BART3 , 6 , 8 , 16 and 22 . Inspection of these mRNA targets revealed several with known pro-apoptotic activity including FEM1B and CASZ1a ( miR-BART3 ) , OCT1 ( miR-BART6 ) , ARID2 ( miR-BART8 ) , CREBBP and SH2B3 ( miR-BART16 ) and finally PPP3R1 , PAK2 and TP53INP1 ( miR-BART22 ) ( see S4 Table for a summary the known functions of these gene products and relevant citations ) . As noted above , all of these mRNAs contained 3’UTR targets , identified by PAR-CLIP , that bear full seed homology to the indicated miR-BART miRNA ( Fig 5A ) . We therefore used PCR to clone the 3’UTRs of each of these human mRNAs ( see S1 Table for sequence coordinates ) and inserted these 3’ to the FLuc gene , as described in Fig 3 . As may be observed ( Fig 5B ) , every 3’ UTR tested conferred substantial inhibition of FLuc activity in cells co-expressing the cognate miR-BART miRNA that was statistically significant ( p<0 . 05 ) . All these 3’UTRs contained a single predicted miR-BART target site that was captured by PAR-CLIP except for CASZ1a , which also contained a captured site with seed homology to miR-BART18 . However , this potential target was not responsive to co-expressed miR-BART18 in co-transfected cells ( Fig 5B ) . In order to test whether the suppression of Fluc activity by a given miR-BART is indeed due to the seed homology present in the clusters captured by PAR-CLIP , we introduced transversion mutations at the 3’UTR nucleotides pairing to miRNA seed positions 2 , 4 and 6 in eight of the nine captured clusters in these 3’UTRs . In the ninth 3’UTR , derived from SH2B3 , the predicted target site for miR-BART16 was removed by deletion . The 3’UTR of DICE1 was not mutated , as the single miR-BART3 binding site present in this 3’UTR has been previously validated ( 23 ) . By assaying 3’UTR-containing FLuc reporters containing these mutations , in parallel with the wild-type 3’UTR-based FLuc reporters used in Fig 5B , we observed that loss of seed homology in the captured PAR-CLIP clusters resulted in the complete loss of miR-BART mediated inhibition of FLuc activity , consistent with our hypothesis that the clusters captured by PAR-CLIP are bound by the predicted miR-BARTs ( S3A Fig ) . However , in the case of CASZ1a , mutation of the single predicted miR-BART3-5p target site only led to a partial recovery of FLuc activity , indicating an additional miR-BART3 target site ( s ) is present in the CASZ1a 3’UTR . Indeed , we computationally identified a potential site with seed homology to miR-BART3-3p ( S3B Fig ) that was not detected by PAR-CLIP but that could account for this residual inhibition . Although miR-BART3-3p is in principle a passenger strand , with miR-BART3-5p representing the dominant arm , both strands are actually recovered at similar levels and have a comparable number of binding clusters ( Fig 4B ) . In conclusion , this mutational analysis confirmed that all 9 candidate 3’UTR target sites are indeed binding sites for the predicted miR-BART . If the pro-apoptotic genes listed in Fig 5 and S4 Table are indeed authentic targets of the miR-BART miRNA listed in the same figure , then expression of physiological levels of that miR-BART in AGS cells should result in a reduction in the expression of that gene [1] . We initially performed qRT-PCR analysis of control AGS cells and of the AGS cells described in Fig 1 that express close to physiological levels of one of the five anti-apoptotic miR-BARTs ( Fig 2 ) looking at the nine cellular mRNAs listed in Fig 5 as well as the mRNA encoding DICE1 , a previously reported [23] potentially pro-apoptotic target for miR-BART3 confirmed by indicator assay in Fig 3 . As noted in Fig 6A , this analysis is more complex in the case of CASZ1 , which is expressed in two spliced variants , encoding CASZ1a and a shorter protein called CASZ1b [39] , as only the 3’UTR found in the longer mRNA splicing isoform encoding CASZ1a is predicted to be a target for miR-BART3 . As shown in Fig 6B , we observed a significant ( p<0 . 05 ) reduction in the level of expression of all the predicted mRNA targets with the exception of the TP53INP1/miR-BART22 combination , where the observed reduction in mRNA expression fell slightly short of significance . Importantly , while expression of the CASZ1a mRNA was significantly reduced in the presence of pre-miR-BART3 , expression of the CASZ1b mRNA was not , as predicted ( Fig 6A ) . While miRNAs can clearly reduce the steady-state expression level of target mRNAs , evidence suggests that a major , and possibly the primary , effect of RISC binding to the 3’UTR of a target mRNA is to reduce the translation of that mRNA [40 , 41] . To address whether the anti-apoptotic miR-BARTs indeed reduce the expression of the proteins encoded by the 10 pro-apoptotic genes listed in Figs 5 and 6 , we performed Western analyses of all ten proteins in the transduced AGS cells expressing the individual anti-apoptotic miR-BART miRNAs . Representative Western blots are shown in Fig 7A and a compilation of data derived from four independent experiments for each of the 10 proteins is shown in Fig 7B . As may be observed , these data uncover statistically significant ( p<0 . 05 ) decreases in expression for all 10 cellular proteins under analysis , with the only non-repressed protein being CASZ1b which , as shown in Fig 6A , is encoded by a spliced mRNA isoform that is not expected to bind miR-BART3 . In contrast , expression of the CASZ1a protein was , as expected , repressed upon miR-BART3 expression ( Fig 7 ) . The work described so far has identified several putatively pro-apoptotic cellular mRNA targets for the five anti-apoptotic EBV miR-BART miRNAs miR-BART3 , 6 , 8 , 16 and 22 and shown that these are , in fact , downregulated at both the mRNA and protein level in AGS cells expressing physiological levels of the miR-BART miRNA in question . However , these data do not address whether this downregulation is , in fact , causatively related to the observed reduction in apoptosis . To test this hypothesis , we constructed two lentiviral vectors expressing artificial miRNAs ( amiRNAs ) specific for each of the 10 candidate mRNA targets , a total of 20 vectors [42] . These were used to transduce AGS cells that were then selected for blasticidin resistance and tested for knockdown of the encoded protein target by Western blot . As shown in S4 Fig , 16 distinct amiRNAs demonstrated some degree of knockdown ranging from >10-fold to as little as ~30% , for the targets CASZ1 , OCT1 , SH2B3 , ARID2 , PAK2 , TP53INP1 and CREBBP as well as DICE1 . Unfortunately , we did not observe significant knockdown of PPP3R1 or FEM1B with either amiRNA tested and these two potential targets were therefore not further addressed . However , we were able to test the other 16 amiRNA- expressing AGS cell lines for their ability to resist the induction of apoptosis seen upon incubation in 5 μM etoposide . As shown in Fig 8 , we observed a statistically significant ( p<0 . 05 ) reduction in apoptosis for both amiRNAs specific for CASZ1 , DICE1 , SH2B3 , PAK2 and TP53INP1 . We also observed a significant reduction in apoptosis in one of the two cell lines expressing an amiRNA specific for OCT1 and CREBBP , with the other cell line showing a trend towards lower apoptosis that did not achieve significance due to a high standard deviation between assay replicates . Finally , neither amiRNA specific for ARID2 resulted in reduced apoptosis , suggesting that this protein is perhaps not , in fact , functionally pro-apoptotic in AGS cells . In conclusion , our data demonstrate that amiRNAs specific for seven distinct cellular genes identified as targets for anti-apoptotic EBV miR-BART miRNAs are able to phenocopy the anti-apoptotic activity of these viral miRNAs .
The primary goal of this study was to determine if any of the miR-BART miRNAs expressed at high levels in EBV transformed epithelial cells have an anti-apoptotic phenotype and , if so , to identify and validate the cellular mRNA targets that mediate this phenotype . This work was initially prompted by published reports arguing that the expression of clusters of miR-BART miRNAs in the gastric carcinoma cell line AGS inhibits the apoptosis caused by exposure to the topoisomerase II inhibitor etoposide [20] and reports , based largely on computational approaches , that identified several individual pro-apoptotic cellular genes as potential targets for specific miR-BART miRNAs [14 , 20–24] . Because the phenotypes exerted by miRNAs can be influenced by their expression level [28] , we initially decided to construct stable cell lines , derived from human AGS cells , that individually expressed a close to physiological level of each of the miR-BART miRNAs using lentiviral vector transduction . As shown in Fig 1 , we were indeed able to achieve a level of expression in AGS cells that was closely comparable to that seen in the naturally EBV transformed NPC cell line C666 for 17 of the 22 miR-BARTs . Analysis of these cell lines then showed that five of the EBV miRNAs , that is miR-BART3 , 6 , 8 , 16 and 22 , were able to reduce the level of apoptosis seen after etoposide treatment ( Fig 2 and S1 Fig ) . We next globally identified the mRNA targets bound by RISC-loaded miR-BART miRNAs by PAR-CLIP analysis of the EBV-transformed epithelial cell line C666 , using a pan-Ago antibody . This resulted in the identification of several cellular mRNA targets bound by the five anti-apoptotic miR-BARTs ( Figs 4D and 5 ) , nine of which were predicted to encode proteins with pro-apoptotic activity ( S4 Table ) . We were able to further validate these cellular mRNAs as authentic targets for the five anti-apoptotic EBV miR-BART miRNAs by several criteria: Insertion of the 3’ UTR , including the PAR-CLIP identified miR-BART seed target , 3’ to the FLuc indicator gene conferred specific downregulation of FLuc when the cognate miR-BART miRNA was expressed in trans ( Fig 5B ) . Moreover , this downregulation was dependent on the integrity of the seed target ( S3 Fig ) . Expression of any one of these miR-BART miRNAs at physiological levels in AGS cells ( Fig 1 ) resulted in a specific and significant downregulation of the level of expression of the endogenous mRNA and protein encoded by the predicted target gene ( Figs 6 and 7 ) . While these four lines of evidence provide strong support for the hypothesis that the 10 genes listed in Figs 6 and 7 ( nine of which are novel while one , DICE1 , has been previously described [23] ) are indeed authentic targets for one of the five anti-apoptotic miR-BARTs , they do not address whether these mRNA targets are directly relevant to the observed anti-apoptotic phenotype ( Fig 2 ) . To address this question , we constructed two artificial miRNA ( amiRNA ) lentiviral expression vectors specific for each of the potential miR-BART mRNA targets tested in Figs 6 and 7 . These lentiviral vectors , which are closely similar to the miR-BART lentivectors used in Figs 1 , 2 , 6 and 7 , were then used to generate stably transduced AGS cell lines expressing these amiRNAs . As shown in S4 Fig , we obtained two amiRNAs that each effectively and stably downregulated the expression of eight of these potentially pro-apoptotic genes in AGS cells ( we did not obtain amiRNAs able to stably downregulate FEM1B or PPP3R1 , either because our amiRNA designs were ineffective or because these proteins are required in AGS cells ) . Analysis of the resultant 16 stable knockdown AGS cell lines obtained showed a significant reduction in apoptosis levels after etoposide treatment in both cell lines expressing an amiRNA specific for CASZ1 , DICE1 , SH2B3 , PAK2 or TP531NP1 and in one of the two cell lines expressing an amiRNA specific for OCT1 or CREBBP1 . Neither amiRNA specific for ARID2 showed an anti-apoptotic phenotype , though both effectively inhibited ARID2 protein expression ( S4D Fig ) . We therefore conclude that we have identified at least seven authentic pro-apoptotic cellular mRNA targets that are significantly downregulated upon expression of one of the anti-apoptotic miR-BART miRNAs at physiological levels in human epithelial cells . These findings can at least partly explain the previously reported anti-apoptotic activity of the miR-BART miRNA cluster in AGS cells [20] . Of the seven anti-apoptotic mRNA targets validated in this manuscript , i . e . , CASZ1 , DICE1 , OCT1 , CREBBP , SH2B3 , PAK2 and TP53INP1 , only one , DICE1 has been previously reported as an mRNA target for miR-BART3 [23] . This was surprising , as a number of other pro-apoptotic cellular mRNAs have also been reported to be targets for miR-BARTs [14 , 20–22 , 24 , 33] . However , as shown in Fig 3 and S2 Fig , we were not able to validate STAT1 , CASP3 or BIM as targets for any of the five pro-apoptotic EBV miRNAs miR-BART3 , 6 , 8 , 16 and 22 . It remains possible , as previously proposed [20] , that the simultaneous expression of multiple miR-BARTs , as seen in EBV-transformed epithelial tumors , would result in significantly reduced expression of STAT1 , CASP3 and/or BIM . However , we note that the 3’UTRs of STAT1 and CASP3 , which have been reported to be targets for miR-BART8 and miR-BART16 respectively [14 , 33] , do not contain full seed targets for either of these two EBV miRNAs and neither 3’UTR was in fact identified as a target for miR-BART8 or miR-BART16 binding in the PAR-CLIP analysis reported in Fig 4D and S3 Table . In addition to the previously reported mRNA targets for the anti-apoptotic miR-BARTs analyzed in Fig 3 , several other potentially pro-apoptotic cellular mRNAs have also been previously reported as targets for other miR-BARTs that did not exert a detectable anti-apoptotic phenotype when expressed individually in AGS cells ( Fig 2 ) . These include BID , a proposed target for miR-BART4 [21]; PUMA , a proposed target for miR-BART5 [22]; PTEN , a proposed target for miR-BART7 [24]; E-Cadherin ( E-CAD ) , a proposed target for miR-BART9 [43]; and finally , EBF1 , a proposed target for miR-BART11 [44] . All of these miRNAs , except miR-BART4 , were expressed at physiological levels in the AGS transductants ( Fig 1 ) , so the lack of a detectable anti-apoptotic phenotype was unexpected . Analysis of our PAR-CLIP data , obtained in C666 cells , as well as previous PAR-CLIP experiments , using Ago-specific antibodies , performed using LCLs or PEL cells latently infected with wildtype EBV [11 , 37] , and expressing readily detectable levels of the miR-BARTs , failed to identify miR-BART binding sites at the proposed locations in the 3’UTRs of any of these mRNAs ( S3 Table ) . Moreover , FLuc-based indicator constructs containing 3’UTRs derived from these five mRNA species either failed to show any evidence of downregulation in the presence of the cognate miR-BART expression plasmid ( BID/miR-BART4; PTEN/miR-BART7; E-CAD/miR-BART9 ) or showed a minimal level of inhibition ( PUMA/miR-BART5 and EBF1/BART11 ) ( S5 Fig ) . We note that the EBF1 3’UTR does not , in fact , contain a seed target for miR-BART11 and is therefore not predicted to be highly responsive to this miRNA . Others have also failed to confirm the identification of PUMA as an authentic target for miR-BART5 using RISC immunoprecipitation or indicator assays [14 , 20] , so the PUMA 3’UTR , despite the presence of a highly complementary potential 3’UTR target , may not in fact be an authentic target for downregulation by miR-BART5 . In conclusion , we have identified a series of at least seven mRNA targets for EBV miR-BART miRNAs that encode pro-apoptotic proteins . The BART miRNA-induced reduction in the expression of these proteins can at least partly explain the previously reported anti-apoptotic activity of the EBV miR-BART locus in EBV latency II epithelial cells [20] . Clearly , this activity could be highly advantageous to EBV in ensuring the survival of these latently infected cells despite the known ability of EBV to activate innate immune pathways that have the potential to induce programmed cell death pathways [45] and may also contribute to the development of resistance seen in a significant percentage of EBV+ NPC tumors in patients undergoing chemotherapy or radiation therapy [46 , 47] .
C666 cells ( a gift from Dr . Nancy Raab-Traub ) [7] , AGS cells ( a gift from Dr . Lindsey Hutt-Fletcher ) [16] and 293T cells ( Duke Cancer Institute Cell Culture Facility ) were cultured using RPMI 1640 , Ham’s F-12 and Dulbecco's Modified Eagle Medium ( Gibco ) , respectively , supplemented with 10% fetal bovine serum and 10 μg/ml gentamicin . All cell cultures were maintained at 37C with 5% CO2 . Lentiviral miR-BART miRNA expression vectors used for FLuc-based 3’UTR reporter assays were generated in the pLenti-CMV-Blasticidin ( pLCB ) backbone , and individual ~300 bp EBV miRNA expression regions , as previously described [30] , were inserted into the 3’UTR of the Blasticidin gene using unique XhoI and XbaI sites . Functional expression of individual miR-BART miRNAs was confirmed using miRNA indicator assays [30] and stem-loop-qRT-PCR ( Fig 1 ) . Lentiviral miR-BART miRNA expression vectors used for transduction of AGS cells were generated in the pTRIPZ backbone ( Open Bioystems ) ( doxycycline inducible turboRFP , puromycin selectable ) , with EBV miRNA expression regions inserted into the 3’ UTR of the turboRFP gene using XhoI and EcoRI sites . FLuc-based 3’UTR reporter plasmids were generated using the pLenti-SV40-GL3 ( pLSG ) backbone [37] by inserting 3’UTRs of candidate cellular target mRNAs ( see S1 Table for full description of the inserted sequences ) into the 3’UTR of FLuc between unique XhoI and XbaI sites . PCR primers used to clone the 3’UTRs are listed in S1 Table . To generate mutant 3’UTR reporter plasmids , internal primers bearing transversion mutations of the nucleotides pairing to seed positions 2 , 4 and 6 of the miRNA were utilized , together with the primers listed in S1 Table , to clone mutant forms of the 3’UTR regions from the wild-type 3’UTR reporter plasmids by overlap extension PCR . To clone a truncated SH2B3 3’UTR , an internal forward primer and the reverse primer listed in S1 Table were utilized . The PCR primers used to clone the mutant 3’UTRs are listed in S7 Table . Lentiviral amiRNA expression vectors were generated in the pLenti-CMV-Blasticidin-Hairpin ( pLCBH ) vector . pLCBH was derived from pLCB by inserting a miR-30-based amiRNA cassette [42] into the 3’UTR of the blasticidin gene between the unique XhoI and EcoRI sites . Oligonucleotides used to clone specific amiRNAs are listed in S6 Table . qRT-PCR for determination of relative mRNA expression and stem-loop qRT-PCR for relative miRNA expression were performed based on vendor protocols . Briefly , total RNA was first harvested using TRIzol ( Ambion ) . For qRT-PCR analysis , RNAs were reverse transcribed using a high capacity reverse transcription kit ( Applied Biosystems ) and assayed with Power SYBR Green PCR Master Mix ( Applied Biosystems ) . Relative gene expression was first normalized to GAPDH and was then compared to the negative control . Primers used to detect distinct isoforms of CASZ1 mRNA were as previously described [39] . All the qPCR primers used are listed in S5 Table . For stem-loop qRT-PCR , total RNA preparations were reverse transcribed using a Taqman miRNA reverse transcription kit ( Applied Biosystems ) , and assayed with Taqman Universal PCR Master Mix , no UNG ( Applied Biosystems ) . The relative miRNA expression level of individual miR-BART miRNAs expressed in transduced AGS cells was first normalized to endogenous U6 , and then to the miR-BART miRNA level detected in C666 cells . All the EBV miR-BART reverse transcription primers and stem-loop qPCR probes were purchased from Life Technologies . 105 AGS cells were plated into each well of a 6-well plate , and after 24 h , a final concentration of 5 μM/ml etoposide ( SigmaAldrich ) ( 25 μM/ml for PARP cleavage ) was added to the medium . After 24 h , both floating and adherent cells were harvested , pooled together and fixed with 80% ethanol for ~4 h . Cells were then stained with PI solution with RNase A ( BD Pharmingen ) and analyzed by flow cytometry . Data were further analyzed by FlowJo ( Treestar ) . Cells were lysed in NP40 lysis buffer supplemented with Complete Mini EDTA-free proteinase inhibitors ( Roche ) . Cell lysates were separated by SDS-PAGE and subsequently transferred to nitrocellulose membranes . Western blots were probed using primary antibodies including anti-FEM1B ( sc-67568 , Santa Cruz ) , anti-CASZ1 ( sc-135453 , Santa-Cruz ) , anti-DICE1 ( sc-376524 , Santa Cruz ) , anti-OCT1 ( sc-232 , Santa Cruz ) , anti-ARID2 ( sc-166117 , Santa Cruz ) , anti-CREBBP ( sc-369 , Santa Cruz ) , anti-SH2B3 ( sc-393709 , Santa Cruz ) , anti-PAK2 ( sc-1872 , Santa Cruz ) , anti-PPP3R1 ( sc-6119 , Santa Cruz ) , anti-TP53INP1 ( sc-689919 , Santa Cruz ) , anti-beta-Actin ( sc-47778 , Santa Cruz ) , and anti-PARP ( 9542p , Cell Signaling ) . The secondary antibodies used included anti-Goat IgG ( sc-2020 , Santa Cruz ) , anti-Mouse IgG ( A9044 , Sigma ) and anti-Rabbit IgG ( A6145 , Sigma ) . All images were obtained using G:BOX ( Syngene ) and GeneSys ( Syngene ) acquisition software , and were subsequently analyzed by Genetools software ( Syngene ) . 10 ng of a pLSG-based 3’UTR reporter , 10 ng pLenti-SV40-Rluc , along with either 500 ng of a miR-BART expression vector or a matched negative control , were co-transfected into 293T cells in 24-well plates using polyethylenimine ( PEI ) . Cells were lysed ~72 h post-transfection with passive lysis buffer ( Promega ) and FLuc and RLuc expression analyzed using a dual luciferase assay kit ( Promega ) . All 3’UTR reporter assays were performed on three separate occasions using technical triplicates . The small RNA deep sequencing library for C666 cells was generated as previously described [11] . Briefly , C666 total RNA was first harvested using TRIzol ( Ambion ) , and the small RNA fraction ( ~18 to ~24 nt ) was subsequently isolated using 15% TBE-Urea polyacrylamide gels ( Bio-Rad ) . The harvested RNAs were then ligated to 3’ and 5’ Illumina adapters , reverse transcribed using SSIII ( Invitrogen ) and subjected to Illumina deep sequencing . The PAR-CLIP library for C666 was generated as previously described [11 , 37] . Briefly , C666 cells were first expanded to 30 150-mm dishes at ~80% confluency , and were then cultured in the presence of 100 μM 4-thiouridine ( 4SU ) for ~20 h . The cells were then UV radiated at 365 nm for 1 minute , harvested and lysed on ice in NP40 lysis buffer . Cross-linked Ago:RNA complexes were then immunoprecipitated using a pan-Ago antibody ( ab57113; Abcam ) and protein G Dynabeads ( Invitrogen ) . Ago-bound RNAs were digested with RNaseT1 , radio-labeled , gel purified , proteinase K treated , phenol-chloroform extracted , ethanol precipitated and ligated to 3’ and 5’ Illumina adapters . After reverse transcription and limited PCR amplification , the recovered cDNAs were deep-sequenced . The C666-derived small RNA deep sequencing library and PAR-CLIP library were processed as previously described [11 , 37] . Briefly , sequencing reads were pre-processed using the FAST-X toolkit ( http://hannonlab . cshl . edu/fastx_toolkit/ ) , and were aligned to the human genome ( hg19 ) and EBV1 wild type genome using Bowtie with up to two ( three for PAR-CLIP ) mismatches allowed . The PAR-CLIP library was further processed using the PARalyzer program , as previously described [11 , 37 , 38] . Briefly , reads were first filtered allowing for up to three mismatches but with only one or zero non-T-to-C mutations . Subsequently , reads that aligned to a unique genomic location , that contained at least one T-to-C mutation and that overlapped by at least one nucleotide were grouped together as clusters . Clusters with a read depth of at least five reads were presented as miRNA:mRNA interaction sites in the PAR-CLIP dataset . Each cluster in the PAR-CLIP dataset was further examined for canonical miRNA seed match sites , using the miRNA expression data generated from the small RNA deep sequencing library derived in parallel , and miRNAs with seed matches equal to or greater than 7mer1A ( perfect base pairing to seed nt 2–7 with an A across from nt 1 of the miRNA , see [5] ) to the cluster were identified as candidate miRNAs putatively responsible for the cluster . The raw sequencing data from the C666 small RNA deep sequencing and PAR-CLIP analysis have been submitted to the NCBI Sequence Read Archive ( SRA ) , and both dataset can be accessed with the accession number GSE67990 . The sub-accession numbers of the individual C666 small RNA deep sequencing and PAR-CLIP libraries are GSM1660655 and GSM1660656 , respectively . | One important innate immune response to viral infection is apoptosis , also called programmed cell death , whereby the infected cells commit suicide rather than serve as factories for virus production . As a result , many viruses have developed strategies to inhibit apoptosis . Here , we demonstrate that five of the Epstein-Barr virus ( EBV ) miR-BART microRNAs that are expressed in EBV-transformed epithelial cell tumors display anti-apoptotic activity . We have identified ten cellular mRNAs that are bound and downregulated by one of these five anti-apoptotic microRNAs and show that this downregulation can explain the observed reduction in apoptosis in miR-BART-expressing cells . Together , these data demonstrate that the EBV miR-BARTs can help sustain latently EBV-infected cells in the face of pro-apoptotic innate immune signals and this may explain the resistance to DNA damaging agents , including chemotherapeutics and radiation , seen in a subset of EBV-induced epithelial tumors . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | EBV BART MicroRNAs Target Multiple Pro-apoptotic Cellular Genes to Promote Epithelial Cell Survival |
The latent EBV nuclear antigen 3C ( EBNA3C ) is required for transformation of primary human B lymphocytes . Most mature B-cell malignancies originate from malignant transformation of germinal center ( GC ) B-cells . The GC reaction appears to have a role in malignant transformation , in which a major player of the GC reaction is Bcl6 , a key regulator of this process . We now demonstrate that EBNA3C contributes to B-cell transformation by targeted degradation of Bcl6 . We show that EBNA3C can physically associate with Bcl6 . Notably , EBNA3C expression leads to reduced Bcl6 protein levels in a ubiquitin-proteasome dependent manner . Further , EBNA3C inhibits the transcriptional activity of the Bcl6 promoter through interaction with the cellular protein IRF4 . Bcl6 degradation induced by EBNA3C rescued the functions of the Bcl6-targeted downstream regulatory proteins Bcl2 and CCND1 , which resulted in increased proliferation and G1-S transition . These data provide new insights into the function of EBNA3C in B-cell transformation during GC reaction , and raises the possibility of developing new targeted therapies against EBV-associated cancers .
B-cell development through the germinal center ( GC ) is controlled strictly by sequential activation or repression of crucial transcription factors , executing the pre- and post-GC B-cell differentiation [1] . The deregulation of induced GC reactions during B-cell development is associated with malignant transformation giving rise to different types of lymphoma and leukemia [2] . Most mature B-cell malignancies , including diffuse large B-cell lymphoma ( DLBCL ) , follicular lymphoma ( FL ) and Burkitt’s lymphoma ( BL ) are derived from malignant transformation of GC B-cells [2 , 3] . Furthermore , DLBCL is the most common subtype of non-Hodgkin’s lymphoma ( NHL ) , accounting for approximately 40% of all cases [4] . DLBCL is considered a heterogeneous group of tumors , with some specific clinicopathological variants of DLBCLs being associated with the presence of EBV [5 , 6] . A major regulator of the GC reaction is represented by B-cell lymphoma 6 ( Bcl6 ) , a sequence specific transcriptional repressor [7–9] . Knock-out of Bcl6 in vivo results in lack of GC formation and the maturation of high-affinity antibodies [10 , 11] . Interestingly , deregulation of Bcl6 expression can be found in BL , FL and DLBCL [12 , 13] . In addition , Bcl6 is the most frequent oncogene involved in roughly 40% of the cases of DLBCLs , and its locus is frequently rearranged due to chromosomal translocations in DLBCL [14 , 15] . As a key transcriptional repressor in normal B-cell differentiation , Bcl6 was shown to repress NF-κB and the positive regulatory domain I element ( PRDM1 ) also known as Blimp-1 in DLBCLs [16–18] . Also , Bcl6 is now been investigated as a potential therapeutic target for the treatment of tumors with rationally designed specific Bcl6 inhibitors [19–21] . EBV is a lymphotropic virus that is linked to many kinds of B-cell malignancies , including BL , FL and DLBCL [22 , 23] . EBV infection transforms primary human B-cells into continuously growing lymphoblastoid cells ( LCLs ) and different latent types were established in EBV-infected cells [23 , 24] . During latency III or the growth program , EBV expresses the full complement of oncogenic latent proteins , including EBV nuclear antigens EBNA1 , EBNA2 , EBNA3A , EBNA3B , EBNA3C and EBNA-LP , as well as latent membrane proteins LMP1 , LMP2A and LMP2B in addition to numerous RNAs and miRNAs [25] . Genetic studies using recombinant virus strategies demonstrated that EBNA1 , EBNA2 , EBNA3A , EBNA3C , EBNA-LP and LMP1 are essential or very important for EBV-mediated transformation of primary B-cells in vitro [26–28] . Specifically , EBNA3C has the ability to function as a transcriptional activator and repressor , and can regulate the transcription of both cellular and viral genes [29 , 30] . A number of earlier studies have shown that EBNA3C interacts with a wide range of transcription factors and modulators , such as c-Myc [31] , IRF4 [32] , CtBP [33] , p53 [34] , E2F1 [35] and E2F6 [36] , which leads to dysregulation of their associated cellular functions . Previous studies have indicated that expression of EBV latent proteins were associated with Bcl6 expression [37–39] . For example , in some B-cell lymphomas , Bcl6 expression was inversely correlated with LMP1 expression , and some data suggested that LMP1 can cause downregulation of Bcl6 [6 , 37] . However , the link between LMP1 and Bcl6 was not fully explained as Bcl6 expression was also inhibited in LMP1-negative cells [38] . Similar studies have shown that LMP1 through heterologous expression was unable to suppress expression of Bcl6 in DLBCLs [39] . In addition , EBNA2 can interfere with the germinal center phenotype by downregulating Bcl6 in Non-Hodgkin's Lymphoma cells [39] . Furthermore , EBV encoded microRNAs can repress Bcl6 expression in DLBCL [38] . However , the mechanism by which Bcl6 is down-regulated in EBV-infected cells is still not fully understood . Our goal is to determine the role of EBNA3C in regulating expression of Bcl6 oncoprotein in B-cells , and further uncover novel molecular mechanisms by which EBNA3C-mediated regulation of cellular functions can lead to B-cell transformation .
To determine the expression levels of Bcl6 during EBV infection of primary B-cells , 10 million human peripheral blood mononuclear cells ( PBMCs ) from two donors , respectively , were infected with wild-type BAC-GFP-EBV or EBNA3C-deleted ( ΔE3C ) BAC-GFP-EBV . The infected cells were harvested at different time points ( 0 , 2 , 4 , 7 and 15 days post-infection ) , then mRNA was extracted and Real-time PCR was performed for Bcl6 detection . The results showed that for both donors Bcl6 expression was down-regulated and expressed at a relatively low level after wild-type EBV infection . However , its expression was consistently enhanced with the EBNA3C-deleted EBV infection as much as 20–35 fold over wild-type infection ( Fig 1A and 1B ) . This suggested that EBNA3C can play a key role in Bcl6 expression during EBV infection . To determine the effect of EBNA3C on Bcl6 expression in Burkitt’s lymphoma cells , western blot analysis was also performed in EBV-negative BL41 and Akata cells , as well as the corresponding EBV-positive BL41/B95 . 8 , Akata-EBV cells . We found a significant downregulation of Bcl6 expression in the presence of EBV-infected BL41 and Akata cell lines of approximately 2–4 fold ( Fig 1C ) . To further investigate whether the differential expression was due to the presence of EBNA3C , EBV-negative Ramos and BJAB cells; EBNA3C expressing stable BJAB cells ( BJAB7 and BJAB10 ) ; and EBV-transformed lymphoblastoid cell lines ( LCL1 and LCL2 ) , were analyzed by western blot . Similarly , Bcl6 protein expression was down-regulated close to 50% in the presence of EBNA3C in Burkitt’s lymphoma cells and were dramatically suppressed in EBV-transformed LCLs ( Fig 1D ) . These results strongly suggested that EBNA3C contributes to inhibition of Bcl6 expression . To further examine if Bcl6 expression was regulated by EBNA3C specifically , HEK293T and BJAB cells were transfected with a dose-dependent increase of EBNA3C in addition to heterologous expression of Bcl6 . The western blot results demonstrated that EBNA3C expression led to a strong reduction in Bcl6 protein expression in human cells , including B-cell lines of about 4–10 fold ( Fig 1E and 1F ) . To further explore the role of EBNA3C in modulating Bcl6 expression levels in EBV transformed LCLs , EBNA3C knocked-down LCL1 cell line was generated with specific EBNA3C short hairpin RNA ( sh-E3C ) [40] . Compared to the control LCL1 ( sh-Ctrl ) , Bcl6 protein expression was significantly increased by at least 2-fold in sh-E3C LCL1 cells ( Fig 1G ) . The results provide additional supporting evidence demonstrating that down-regulation of Bcl6 expression can be specifically linked to EBNA3C . Next , we examined whether EBNA3C could interact directly with Bcl6 . Two experiments using Co-Immunoprecipitation ( Co-IP ) assays were performed in different cell types . First , HEK293T cells were transfected with Myc-tagged EBNA3C and HA-tagged Bcl6 . The Co-IP results showed that EBNA3C associated in a complex with Bcl6 ( Fig 2A ) . Similarly , an experiment using Saos-2 cells also showed the formation of a complex of EBNA3C and Bcl6 in these cells ( Fig 2B ) . Second , to further determine the interaction in B-cell lines , BJAB , EBNA3C stably expressed BJAB cells ( BJAB7 and BJAB10 ) , and EBV-transformed cells ( LCL1 and LCL2 ) were used in our Co-IP experiment . Western blot analysis also validated the above results demonstrating that endogenous EBNA3C can physically associate with Bcl6 in the background of B-cells and more importantly in EBV-transformed lymphoblastoid cell lines ( Fig 2C and 2D ) . To determine the functional binding domain of EBNA3C that specifically interacts with Bcl6 , Myc-tagged full length and truncated regions of EBNA3C ( 1-365aa , 366-620aa , 621-992aa ) were co-transfected into HEK293T cells with HA-tagged Bcl6 . The targeted protein was immunoprecipitated with anti-Myc or anti-HA antibody , respectively . The results demonstrated that Bcl6 was associated with EBNA3C ( 366-620aa ) along with the full-length EBNA3C protein ( 1-992aa ) ( Fig 2E and 2F ) . Little or no detectable co-immunoprecipitation was observed with the control vector indicating the specificity of the complex between EBNA3C and Bcl6 . These results showed that EBNA3C amino acid residues 366-620aa which includes the acidic domain were responsible for the interaction of EBNA3C and Bcl6 protein ( Fig 2G ) . Previous studies have shown that EBNA3C binds to Bcl6 specifically in human cells , so it is expected that these two proteins would be localized within the same cellular compartments . To determine the co-localization of EBNA3C and Bcl6 , HEK293T cells were transfected with constructs expressing Myc-tagged EBNA3C and HA-tagged Bcl6 , and the cellular localization of the expressed proteins was studied using immunofluorescence microscopy . In cells transfected independently with Myc-EBNA3C or HA-Bcl6 alone , both were found to be primarily expressed in the nucleus ( Fig 3A ) . In cells co-transfected with Myc-EBNA3C and HA-Bcl6 , the merged yellow fluorescence demonstrated that EBNA3C co-localized with Bcl6 in human cells ( Fig 3A ) . Similar results were also observed in Saos-2 cells ( Fig 3B ) . To further determine the co-localization of EBNA3C and Bcl6 proteins in more relevant B-cells , immunofluorescence assays were performed using antibodies specific to EBNA3C and Bcl6 proteins in order to examine the endogenous expression in different B-cell lines . The results further confirmed that EBNA3C co-localized with Bcl6 in nuclear compartments of EBV-transformed LCLs ( Fig 3C ) . This was consistent with the results of the above studies which demonstrated that EBNA3C associated with Bcl6 in nuclear complexes as seen in the Co-IP experiments in human cells . To explore the potential mechanism of EBNA3C-mediated down-regulation of Bcl6 , a stability assay was performed to determine whether EBNA3C regulated Bcl6 expression at the protein level . HEK293T cells were transfected with HA-tagged Bcl6 and Myc-tagged EBNA3C or Myc-tagged empty vector . Twenty-four hours post-transfection , cells were incubated with protein synthesis inhibitor cycloheximide ( CHX ) and monitored for Bcl6 protein levels at 0 , 4 , 8 , 12 hours by western blot analysis . As expected , the results showed that the stability of Bcl6 protein was significantly decreased by greater than 50% in the presence of EBNA3C by 12 hours , while the Bcl6 protein level was more stable in the absence of EBNA3C ( compare right and left panels , Fig 4A ) . To further support these results , BJAB ( EBNA3C negative B-cells ) , and BJAB7 ( BJAB stably expressing EBNA3C cells ) were treated with CHX for 0 , 2 , 4 , and 6 hours . The following western blot analysis also demonstrated that there was a dramatic reduction in the stability of Bcl6 protein which was directly associated with EBNA3C expression as seen by the significant change in the Bcl6 levels by 2 hours post cycloheximide treatment and greater than 50% by 6 hours ( compare right and left panels , Fig 4B ) . Bcl6 expression is strictly regulated during GC reaction , and its degradation through the ubiquitin-proteasome pathway is crucial for B-cell development or lymphomagenesis in temporal function . Earlier studies showed that Bcl6 could be degraded by the ubiquitin-mediated proteasome [41 , 42] . Therefore , it is expected that Bcl6 degradation is likely mediated by EBNA3C utilizing a similar pathway as EBNA3C has been shown to recruit E3 ligases for targeted degradation of cellular substrates [43 , 44] . To determine whether this is the case , HA-Bcl6 was transfected along with Myc-EBNA3C or control vector . Twenty four hours post-transfection , the cells were treated with the proteasome inhibitor MG132 for 12 hours or vehicle control . The following western blot analysis showed that EBNA3C promoted the degradation of Bcl6 protein , which was similar to the above results . However , Bcl6 protein expression was increased after MG132 incubation , even in the presence of EBNA3C ( Fig 4C ) . These results demonstrated that the stability of the Bcl6 protein is regulated by EBNA3C via the ubiquitin-proteasome pathway . To further support our hypothesis , ubiquitination assays were performed with different expression plasmids for Myc-E3C , HA-Ub and HA-Bcl6 , and incubated for 24 hours followed by MG132 treatment for another 12 hours . This was followed by immunoprecipitation and western blot analysis . The results demonstrated enhanced ubiquitination of Bcl6 when EBNA3C was expressed , when compared with control vector or HA-Ub alone ( Fig 4D ) . This strongly indicated that Bcl6 is likely degraded by expression of EBNA3C through the ubiquitin-proteasome dependent pathway . Bcl6 gene expression is tightly regulated during mature B-cell development [2 , 45 , 46] . Our above studies showed that Bcl6 mRNA expression was down-regulated after EBV infection , and that this was associated with EBNA3C expression . To further define how EBNA3C can regulate Bcl6 expression at the mRNA level , different B-cell lines ( BJAB , BJAB7 , BJAB10 , LCL1 and LCL2 ) were used to monitor endogenous Bcl6 mRNA expression . Bcl6 mRNA expression was significantly greater ( >20 fold ) in BJAB cells compared to EBNA3C stably expressed BJAB7 and BJAB10 cell , and also EBV-transformed LCL1 and LCL2 cells ( Fig 5A ) . To verify that the inhibition was related to the presence of EBNA3C , EBNA3C stably knocked-down LCL1 ( sh-E3C ) and the control LCL1 ( sh-Ctrl ) were used to detect Bcl6 mRNA expression . The results showed that Bcl6 mRNA expression was upregulated significantly after the knockdown of EBNA3C ( Fig 5B ) . In addition , BJAB10 cells were then transfected with specific EBNA3C short hairpin RNA ( sh-E3C ) to knock down EBNA3C expression . Expectedly , Bcl6 mRNA expression was increased in the EBNA3C knockdown cell lines ( Fig 5C ) . These findings undoubtedly provide new evidence that EBNA3C can inhibit Bcl6 mRNA expression . Bcl6 promoter transcriptional activity could not only be controlled by Bcl6 through binding to the upstream regulatory region of its gene [15] , but is also inhibited directly by the transcription factor IRF4 via binding to multiple sites within its promoter [47] . To test whether EBNA3C-mediated Bcl6 mRNA down-regulation was related to its transcriptional activity at the Bcl6 promoter , a dual-luciferase reporter system was implemented . The reporter construct containing a wild-type Bcl6 promoter ( pLA/B9 ) and a dose-dependent increase of Myc-EBNA3C were transfected into cells . Meanwhile , the thymidine kinase promoter-Renilla luciferase reporter plasmid ( pRL-TK ) was additionally transfected and used as a control for transfection efficiency . The luciferase assay results clearly revealed that the Bcl6 promoter activity was inhibited by EBNA3C in a dose-dependent manner ( Fig 5D ) . Previous experiments showed that EBNA3C did not bind with DNA directly and functions through binding of other cellular transcription proteins to regulate gene expression [48] . Therefore , other transcription proteins mediate the inhibition of viral and cellular genes . EBNA3C interacted with p53 , attenuated its function and mediated its degradation [34 , 44 , 49] . Furthermore , p53 can activate Bcl6 transcription [50] . It suggests that the transcription activity of the Bcl6 promoter may be inhibited by EBNA3C-induced p53 degradation . However , our results using MEF ( p53-/- ) cells showed that the regulation of Bcl6 promoter by EBNA3C was independent of the function of p53 protein ( S1 Fig ) . Among several other transcriptional proteins that inhibited Bcl6 promoter , we found that IRF4 , a DNA-binding protein , was an important transcription factor for regulating Bcl6 promoter activity [47] . Interestingly , EBNA3C also interacted with IRF4 and contributed to stabilization of IRF4 [32] . One study showed that a high level of IRF4 was expressed in LMP1-KO EBV-induced lymphoma [51] . Thus , we further assessed the possible function of EBNA3C on IRF4-mediated Bcl6 promoter activity . To specifically test the Bcl6 promoter activity , the wild-type and DNA binding domain ( DBD ) -deleted IRF4 plasmids were used ( Fig 5E ) . The results showed that EBNA3C enhanced the IRF4-mediated inhibition of the Bcl6 promoter activity , and the effect was dependent on the DNA binding domain of IRF4 as the promoter repression was rescued when EBNA3C and IRF4-ΔDBD were co-expressed ( Fig 5E ) . It also suggested that IRF4 is one of the major transcription factors that mediate EBNA3C-regulated inhibition of Bcl6 promoter activity . To examine the effect of EBNA3C on Bcl6-mediated cell proliferation , Saos-2 cells were transfected with expression constructs of EBNA3C and Bcl6 , and selected with G418 for two weeks to monitor colony formation . We observed a significant increase in colony numbers when EBNA3C and Bcl6 were co-transfected in comparison to those transfected with only EBNA3C or Bcl6 ( Fig 6A ) . We further extended these studies by performing cell proliferation assays as determined by cell counting for 10 days in Saos-2 cells ( Fig 6B ) . A similar experiment was also repeated in HEK293 cells ( Fig 6C ) . These results demonstrated that expression of EBNA3C and Bcl6 results in a strong induction in cell proliferation . The anti-apoptotic proto-oncogene Bcl2 protein is a critical regulator protein and its expression is inhibited by Bcl6 in GC B-cells [52 , 53] . Therefore , it was reasonable to believe that EBNA3C-mediated Bcl6 down-regulation will lead to up-regulation of Bcl2 expression . Therefore , its anti-apoptotic function will be activated and leads to promotion of cell proliferation . To determine the expression of Bcl2 in B-cells , LCL1 was treated with a Bcl6-specific inhibitor ( 79–6 ) to suppress Bcl6 activity [20] . The Bcl6 inhibitor disrupts Bcl6 transcription activity by binding to its BTB/POZ domain [20] . The following western blot results showed that Bcl2 expression was up-regulated after Bcl6 inhibitor incubation in B-cells by approximately 2-fold ( Fig 6D ) . To further support the results , Bcl2 mRNA expression was determined in the stable EBNA3C or Bcl6 knock-down LCL1 cells to verify that EBNA3C promoted Bcl2 up-regulation through Bcl6 down-regulation in LCLs ( S2 Fig ) . This suggests that EBNA3C-mediated Bcl6 inhibition can contribute to cell proliferation through the Bcl2-associated signaling pathway . The soft agar assay for colony formation measures anchorage-independent in vitro transformation . The oncogene Bcl6 can confer anchorage-independent growth to immortalized cells [54] . To investigate the effects of EBNA3C on Bcl6-related transforming activity , the stable Bcl6 knock-down BJAB and LCL1 cells were generated with lentivirus transduction and puromycin selection ( Fig 7A–7C ) . Soft agar assays were performed using the Bcl6 knock-down BJAB and LCL1 cells . The down-regulation of Bcl6 inhibited the ability of colony formation in BJAB cells ( Fig 7D ) , but oppositely , the ability was enhanced in EBV-transformed LCLs ( Fig 7E ) . The results indicate that EBV promotes transformation and anchorage-independent growth through the inhibition of Bcl6 expression . Our previous study showed that EBNA3C could only stabilize Cyclin D1 ( CCND1 ) protein , but not promote its transcription activity [40] . This suggests that other cellular factors may regulate CCND1 expression . Interestingly , CCND1 is induced by Bcl6 in human B-cells [55] . To further investigate the function of Bcl6 in cell cycle , we analyzed CCND1 mRNA expression in stable B-cells . The results show that CCND1 expression is also suppressed when Bcl6 is knocked-down in EBV-negative BJAB cells . However , its expression is upregulated in stable Bcl6 knock-down EBV-transformed LCL1 cells ( Fig 8A and 8B ) . These results suggest that Bcl6 plays a critical role in controlling CCND1 mRNA expression in a B-cell background . The following cell cycle experiments also demonstrated that the upregulation of CCND1 through Bcl6 inhibition facilitates G1-S transition in EBV-transformed LCL1 cells , but not in the EBV-negative BJAB cells ( Fig 8C and 8D ) . Similar results were observed with other sh-Bcl6 clones .
Bcl6 is a nuclear phosphoprotein of the BTB/POZ/Zinc Finger ( ZF ) protein family , and functions as a transcription repressor to repress target genes by binding to specific DNA sequences and recruiting corepressors [8 , 56] , including SMRT , MTA3 , N-CoR and HDAC [57–60] . Bcl6 is indispensable for GC formation and somatic hypermutation ( SHM ) during B-cell development , thus chromosomal translocations and mutations of Bcl6 regulatory region lead to the deregulation of Bcl6 expression in about 40% DLBCL and 5–10% FL [46] . Although Bcl6 expression is associated with EBV latent antigen EBNA2 and LMP1 , the reported conflicting results did not provide a reasonable explanation or a detailed mechanism on EBV-mediated Bcl6 degradation in B-cell lymphoma [37–39] . A recent study indicated that EBNA3C had no effects on Bcl6 expression , but a previous paper also showed that Bcl6 expression can be increased more than 10-fold in EBNA3C-deleted EBV infection [61 , 62] . Here , our data clearly show that Bcl6 expression can be down-regulated by EBNA3C specifically at transcriptional and post-transcriptional levels ( Fig 9 ) . This is different from the well-known Bcl6 translocations or mutations identified on the human genome associated with oncogenesis . First of all , our results demonstrated that EBNA3C was specifically associated with Bcl6 , and mediated Bcl6 protein degradation through the ubiquitin-proteasome dependent pathway . A previous study showed that Bcl6 protein can be targeted for degradation by cellular factor FBXO11 in DLBCL [63] . However , the role of EBNA3C on FBXO11-related Bcl6 stability is still unclear . It is possible that FBXO11 is the E3 ligase recruited by EBNA3C for Bcl6 degradation . In addition , the acetylation of Bcl6 within the PEST domain inactivates its function of recruiting co-repressors [54] , and the activation of MAPK signaling pathway induces phosphorylation of Bcl6 followed by degradation through the ubiquitin-proteasome pathway [64] . Further studies are warranted to determine whether EBNA3C-mediated Bcl6 degradation is related to Bcl6 acetylation and phosphorylation . BCL6 activity is also dysregulated by translocation or mutation in a remarkably high proportion of DLBCL and FL [65] . The chromosomal translocations of Bcl6 regulatory region referred to as promoter substitution , and frequent mutations of the 5’ noncoding region of Bcl6 result in its deregulated expression , suggesting a key role for Bcl6 in pathogenesis of B-cell lymphoma [12 , 13 , 66 , 67] . However , our results indicate that Bcl6 mRNA expression is down-regulated through EBNA3C-mediated inhibition of the transcription activity of Bcl6 promoter by recruiting another cellular factor IRF4 . This is consistent with a previous study showing that the CD40 receptor signaling pathway leads to NF-κB-mediated IRF4 activation , and furthermore Bcl6 downregulation [47] . Meanwhile , we also showed that EBNA3C could interact with IRF4 and was critical for IRF4 stabilization [32] . This suggests that EBNA3C may mimic the activities of the CD40 ligand to induce NF-κB-IRF4 signaling pathway or enhance the stability of IRF4 protein directly to repress the transcription activity of the Bcl6 promoter . Moreover , EBNA3C was associated with IRF8 and mediated its destabilization and degradation [32] . Interestingly , IRF8 is the only transcriptional activator of Bcl6 to upregulate its mRNA expression in GC reaction [68] . Therefore , it is conceivable that EBNA3C may downregulate Bcl6 expression by activating CD40 signaling pathway as well as regulating IRF4/IRF8 stability . The importance of Bcl6 function in GC B-cells is reflected in the multiple functional pathways it can regulate in the cell . To date , more than one thousand genes are found to be targeted by Bcl6 through binding on their promoters and further modulating the downstream signaling pathways during GC development , involved in cell apoptosis , cell cycle and cell differentiation [69 , 70] . Among the targeted proteins , Bcl2 is a critical anti-apoptosis protein and the direct target of Bcl6 that can interact with Miz1 and bind to Bcl2 promoter to inhibit Miz1-induced Bcl2 transcription activity in GC B-cells [52 , 53] . The dysregulation of Bcl6-mediated Bcl2 expression is often found in DLBCL and FL [2] . Our results show that Bcl2 , a Bcl6 target protein , is regulated by EBNA3C and is increased in LCLs treated with a Bcl6 inhibitor . Thus EBNA3C can induce cell proliferation by degrading and inhibiting the expression of Bcl6 and so releasing the suppression of Bcl2 , therefore activating the anti-apoptosis pathway for tumorigenesis . Moreover , CCND1 , a direct target of Bcl6 in human B-cells , is de-repressed to promote G1-S transition in EBV-transformed LCLs . Whether other cyclin proteins are also under the control of Bcl6 is still unknown . Interestingly , several studies have shown that CCND2 is another target of Bcl6 , but its expression is negatively correlated [71–75] . Activation-induced cytidine deaminase ( AID ) which is responsible for somatic hypermutation and class-switch recombination is also required in GC-derived lymphomas , and its expression is upregulated by EBNA3C in EBV-infected cells [61 , 76] . Bcl6 could promote AID expression by inhibiting miR-155 and mir-361 , so how EBNA3C regulates AID expression without the help of Bcl6 needs to be further explored [2] . A recent study concluded that Bcl6 targeted genes in T follicular helper ( Tfh ) cells through analysis of its genome-wide occupancy and transcriptional regulatory networks [77] . The current development of Bcl6 small-molecular inhibitor indicates a huge potential for Bcl6 as a therapeutic target to treat human lymphomas [19] . However , the Bcl6-mediated regulatory networks are still unknown in EBV-transformed LCLs . Next , xenografts of LCLs in Bcl6 knock-out mice will further reveal the biological function of Bcl6 in EBV-related lymphomagenesis . However , a more efficient in vivo model will be necessary to uncover the crucial functions of EBNA3C or other latent antigens in GC reaction . In summary , the inhibition of Bcl6 expression by the essential EBV antigen EBNA3C may provide a novel insight into the current understanding of EBV contribution on lymphomagenesis by blocking GC reaction . Importantly , a number of EBV latent proteins are expressed in EBV infected cells , but how these latent proteins cooperate with each other to regulate B-cell development , or lead to B-cell lymphoma still needs further investigation . Nevertheless , our observations have implications for emerging strategies targeted at the EBV-associated cancers .
The University of Pennsylvania Immunology Core ( HIC ) provided us human peripheral blood mononuclear cells ( PBMC ) from different unidentified and healthy donors with written , informed consent . All the procedures were approved by the Institutional Review Board ( IRB ) and conducted according to the declarations of Helsinki protocols [36 , 78] . Myc-tagged full-length EBNA3C or its truncations such as 1-365aa , 366-620aa , 621-992aa , and Flag-tagged IRF4 plasmids have been described previously [32] . Myc-tagged constructs expressing full length or DNA binding domain ( DBD ) mutant IRF4 , and HA-tagged full length Bcl6 , wild-type Bcl6 promoter plasmids [9 , 54] were kindly provided by Dr . Riccardo Dalla-Favera ( Columbia University , New York , USA ) . HEK293 or HEK293T ( human embryonic kidney cell line ) , Saos-2 ( human osteosarcoma cell line ) , EBV-negative or -positive cells have been described earlier in detail [32 , 36] . MEF ( mouse embryonic fibroblast cell line ) was a gift from Xiaolu Yang ( University of Pennsylvania ) [79] . HEK293 , HEK293T , Saos-2 , MEF ( p53-/- ) and MEF ( p53+/+ ) cells were grown in Dulbecco's modified Eagle's medium ( DMEM ) , while B-cell lines were maintained in RPMI 1640 media . All the above-mentioned cells were incubated at 37°C in a humidified 5% CO2 environment . The Bcl6 inhibitor ( 79–6 ) was purchased from EMD Millipore ( Billerica , MA , USA ) . Bcl6 antibody ( N-3 ) and Bcl2 antibody ( C-2 ) were purchased from Santa Cruz biotechnology ( Santa Cruz , CA , USA ) . Bcl6 antibody ( ab19011 ) were purchased from Abcam ( Cambridge , UK ) . Antibodies for IRF4 , Ub , GAPDH have been described earlier [32] . Flag antibody ( M2 ) was purchased from Sigma-Aldrich ( St . Louis , MO , USA ) . Other antibodies to mouse anti-Myc ( 9E10 ) , anti-HA ( 12CA5 ) , anti-EBNA3C ( A10 ) were prepared from hybridoma cultures and mentioned previously [80] . 10 million transfected cells or 50 million B-cells were harvested , washed with ice-cold 1×PBS twice , lysed in 400μl ice-cold RIPA buffer [1% Nonidet P-40 ( NP-40 ) , 10 mM Tris ( pH8 . 0 ) , 2 mM EDTA , 150 mM NaCl , supplement with protease inhibitors ( 1 mM phenylmethylsulphonyl fluoride ( PMSF ) , 1 μg/ml each aprotinin , pepstain and leupeptin] . Lysates were precleared with normal control serum plus 30 μl of a 2:1 mixture of Protein-A/G Sepharose beads ( GE Healthcare Biosciences , Pittsburgh , PA ) for 1 h at 4°C . Approximately 5% of the lysate was saved as input . About 1 μg of specific antibody was used to capture the protein of interest by overnight rotation at 4°C . Input and IP samples were boiled in laemmli buffer , resolved on SDS-PAGE gel and transferred to a 0 . 45 μm nitrocellulose membrane . The membrane was blocked in 1×TBS-Tween with 5% w/v non-fat dry milk probed with appropriate primary antibody , subsequently incubated with corresponding secondary antibody , and visualized on a Licor Odyssey imager ( LiCor Inc . , Lincoln , NE ) . Image analysis and quantification measurements were performed using Image Quant application software ( LiCor Inc . , Lincoln , NE ) . The relative density ( RD ) of indicated proteins were shown . HEK293T or Saos-2 cells plated on coverslips were transfected with expression plasmids or not as indicated . Forty eight hours post-transfection , cells were fixed by 4% paraformaldehyde ( PFA ) including 0 . 1% Triton X-100 for 15–20 mins at room temperature [81] . B-cells were air-dried and fixed similar to above . The fixed cells were washed with 1×PBS for three times , and 5% Bovine serum albumin ( BSA ) was used for blocking . EBNA3C and Bcl6 were detected by mouse anti-EBNA3C ( A10 ) and rabbit anti-Bcl6 antibody , respectively . The slides were examined using an Olympus Fluoview 300 confocal microscope , and Images were analyzed by Fluoview software ( Olympus Inc . , Melville , NY ) . HEK293T cells were co-transfected with pLA/B9 plasmid ( Bcl6 promoter , a gift from Dr . Riccardo Dalla-Favera ) [47] , pRL-TK ( Promega , Madison , WI , USA ) , Myc-tagged EBNA3C , and control or Myc-tagged IRF4/IRF4-ΔDBD plasmids . Forty eight hours post-transfection , cells were harvested and the dual-luciferase reporter assay was performed according to the manufacture’s protocols ( Promega , Madison , WI , USA ) . At the same time , the supernatant was collected and prepared for detection by western blot . Cells were collected and washed with ice-cold 1×PBS prior to RNA isolation . Then total RNA extraction was performed using Trizol reagent ( Invitrogen , Inc . , Carlsbad , CA ) and treated with Dnase I ( Invitrogen , Inc . , Carlsbad , CA ) , then cDNA was prepared with Superscript II reverse transcriptase kit ( Invitrogen , Inc . , Carlsbad , CA ) according to the manufacturer’s protocol . Primers for GAPDH were 5′-TGCACCACCAACTGCTTAG-3′ and 5′-GATGCAGGGATGATGTTC-3′ [40] . Quantitative Real-time PCR analysis was performed by using SYBR green Real-time master mix ( MJ Reserch Inc . , Waltham , MA ) . The assays were performed in triplicate . Transfected HEK293T cells were treated with protein synthesis inhibitor cycloheximide ( CalBiochem , Gibbstown , NJ ) after 24 hours transfection as 40 μg/ml concentration . Cells were harvested after 16 hours incubation and lyse with RIPA buffer , then protein samples were quantitated and used for western blot analysis . Protein band intensities were quantified using Image Quant 3 . 0 software . 10 million HEK293 or Saos-2 cells were transfected with control vector , Myc-EBNA3C , Myc-Bcl6 and GFP vector by electroporation and allowed to grow in DMEM supplemented with 1 mg/ml G418 ( Sigma-Aldrich , St . Louis , MO , USA ) . After two weeks selection , GFP fluorescence of every plate was scanned by PhosphorImager ( Molecular Dynamics , Piscataway , NJ ) and the area of the colonies measured by using Image J software ( Adobe Inc . , San Jose , CA ) . Three independent experiments were performed . The two sense strands of Bcl6 shRNA are 5’-tcgagtgctgttgacagtgagcgaGCCTGTTCTATAGCATCTTTAtagtgaagccacagatgtaTAAAGATGCTATAGAACAGGCgtgcctactgcctcggaa–3’ ( sh-Bcl6-1 ) , and 5’- tcgagtgctgttgacagtgagcgaCCACAGTGACAAACCCTACAAtagtgaagccacagatgtaTTGTAGGGTTTGTCACTGTGGgtgcctactgcctcggaa–3’ ( sh-Bcl6-2 ) , respectively . The upper-cases designate Bcl6 target sequences , while lower cases specify hairpin and enzyme sequences . These sense stranded oligos were annealed with their respective anti-sense stranded oligos , and then cloned into pGIPZ vector with Xho I and Mlu I restriction sites . Besides , a negative control was set using a sh-Ctrl plasmid including control shRNA sequence 5’-TCTCGCTTGGGCGAGAGTAAG–3’ ( Dharmacon Research , Chicago , IL ) . Lentivirus production and transduction of B-cell lines has been described previously with a slight modification [32] . A pool of two shRNAs that targeted different regions of the Bcl6 mRNA were co-transfected to generated shRNA-expressing lentiviruses . The BJAB or LCL1 stable cell lines were generated according to the above-mentioned protocols . Approximately , 5 million BJAB or LCL1 stable cells were collected , fixed with 80% ethanol for 2 hours or overnight at -20°C , then washed with 1×PBS and incubated with PI staining buffer ( 0 . 5 mg/ml propidium iodide in 1×PBS , 50 μg/ml RNase A ) for 30 minutes to 2 hours at room temperature . The indicated cells were washed with 1×PBS once , resuspended in 500 μl 1×PBS , and analyzed on FACS Calibur ( Becton Dickinson , San Jose , CA , USA ) using FlowJo software ( TreeStar , San Carlos , CA , USA ) . The soft agar assays were performed using BJAB or LCL1 cells . Briefly , 1 ml of 0 . 5% agar in supplemented RPMI media was poured into 6-well plate and set aside to solidify . 0 . 5 ml 0 . 3% agar/medium containing 2×105 cells was added to the previously plates as the middle layer . Then cells were covered with a top layer of another 1ml 0 . 5% agar/medium . After two weeks , colonies were stained with 0 . 005% crystal violet for 1 hour , and scanned using a Licor Odyssey system ( LiCor Inc . , Lincoln , NE ) . The number of colonies was counted using ImageJ software . Data represented here are the mean values with standard deviation ( SD ) . The significance of differences in the mean values was calculated by performing 2-tailed student's t-test . P-value of <0 . 05 was considered as statistically significant in all our results ( *P < 0 . 05; **P < 0 . 01; ***P < 0 . 001; NS , not significant ) . Epstein-Barr virus ( EBV ) genome , strain B95-8-GenBank: V01555 . 2; EBNA3C ( Human herpesvirus 4 ) -NCBI Reference Sequence: YP_401671 . 1; Bcl6 ( Homo sapiens ) -NCBI Reference Sequence: NM_001130845 . 1; IRF4 ( Homo sapiens ) -NCBI Reference Sequence: NM_002460 . 3; Bcl2 ( Homo sapiens ) -NCBI Reference Sequence: NM_000633 . 2; CCND1 ( Homo sapiens ) -NCBI Reference Sequence: NM_053056 . 2; p53 ( Homo sapiens ) -NCBI Reference Sequence: NM_000546 . 5 . | Epstein-Barr virus ( EBV ) is the first characterized human tumor virus , which is associated with a broad range of human cancers . One of the latent proteins , EBV encoded nuclear antigen 3C ( EBNA3C ) plays a critical role in EBV-mediated B-cell transformation . Bcl6 is a master regulator required in mature B-cells during germinal center ( GC ) reaction . As a transcriptional repressor , Bcl6 can be targeted during malignant transformation and therefore contributes to its function as an oncoprotein during lymphomagenesis . In this study , we demonstrated that EBNA3C interacts with Bcl6 and facilitates its degradation through the ubiquitin-proteasome dependent pathway , and suppresses Bcl6 mRNA expression by inhibiting the transcriptional activity of its promoter . Furthermore , EBNA3C-mediated Bcl6 regulation significantly promotes cell proliferation and cell cycle by targeting Bcl2 and CCND1 . Therefore , our findings offer new insights into the functions of EBNA3C during B-cell transformation in GC reaction and B-cell lymphoma development . This increases the possibility of developing new therapies for treating EBV-associated cancers . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immunology",
"messenger",
"rna",
"cancers",
"and",
"neoplasms",
"cell",
"processes",
"rna",
"extraction",
"dna-binding",
"proteins",
"oncology",
"hematologic",
"cancers",
"and",
"related",
"disorders",
"immunoprecipitation",
"lymphomas",
"co-immunoprecipitation",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"cell",
"proliferation",
"animal",
"cells",
"proteins",
"gene",
"expression",
"hematology",
"precipitation",
"techniques",
"antibody-producing",
"cells",
"biochemistry",
"rna",
"cell",
"biology",
"b",
"cells",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types"
] | 2017 | An essential EBV latent antigen 3C binds Bcl6 for targeted degradation and cell proliferation |
The advent of high-throughput sequencing techniques has made it possible to follow the genomic evolution of pathogenic bacteria by comparing longitudinally collected bacteria sampled from human hosts . Such studies in the context of chronic airway infections by Pseudomonas aeruginosa in cystic fibrosis ( CF ) patients have indicated high bacterial population diversity . Such diversity may be driven by hypermutability resulting from DNA mismatch repair system ( MRS ) deficiency , a common trait evolved by P . aeruginosa strains in CF infections . No studies to date have utilized whole-genome sequencing to investigate within-host population diversity or long-term evolution of mutators in CF airways . We sequenced the genomes of 13 and 14 isolates of P . aeruginosa mutator populations from an Argentinian and a Danish CF patient , respectively . Our collection of isolates spanned 6 and 20 years of patient infection history , respectively . We sequenced 11 isolates from a single sample from each patient to allow in-depth analysis of population diversity . Each patient was infected by clonal populations of bacteria that were dominated by mutators . The in vivo mutation rate of the populations was ∼100 SNPs/year–∼40-fold higher than rates in normo-mutable populations . Comparison of the genomes of 11 isolates from the same sample showed extensive within-patient genomic diversification; the populations were composed of different sub-lineages that had coexisted for many years since the initial colonization of the patient . Analysis of the mutations identified genes that underwent convergent evolution across lineages and sub-lineages , suggesting that the genes were targeted by mutation to optimize pathogenic fitness . Parallel evolution was observed in reduction of overall catabolic capacity of the populations . These findings are useful for understanding the evolution of pathogen populations and identifying new targets for control of chronic infections .
The opportunistic pathogen Pseudomonas aeruginosa is found in many environments and can cause acute or chronic infections in a range of hosts from protozoans to plants to humans [1] , [2] . In particular , patients with cystic fibrosis ( CF ) are highly susceptible to chronic colonization by P . aeruginosa , which is frequently fatal because of a persistent inflammatory response leading to gradual decline of lung function [3] , [4] . In most cases , following a period of recurrent colonizations , a single strain of P . aeruginosa becomes predominant and persists for the rest of the patient's life [5] , [6] . Genetic adaptation has been shown to play a major role in successful establishment of long-term chronic P . aeruginosa infections of CF patients , and natural selection acts on these bacteria in CF airways to accommodate the fixation of mutations that cause beneficial phenotypic changes [7] , [8] , [9] . The selected phenotypes display traits that differ from those of environmental isolates but are common in populations found in CF patients , suggesting repeatable patterns of long-term adaptation to the CF lung [10] , [11] , [12] . A trait frequently observed in chronic infections is an increased mutation rate leading to a mutator phenotype [13] , [14] . P . aeruginosa from chronically infected CF airways was the first natural model to reveal a high proportion of mutators in contrast to reported proportions in acute infections [13] . Hypermutability in CF P . aeruginosa is due primarily to inactivation of the mismatch repair system ( MRS ) through lost function of the antimutator mutS and mutL genes [15] , and 36–54% of CF patients have been shown to be infected by mutator isolates [13] , [16] , [17] , [18] , [19] . Theoretical and experimental approaches have attempted to explain the selection of MRS-mutators as the result of co-selection ( hitchhiking ) with linked beneficial mutations [20] , [21] , [22] , [23] , [24] , [25] , and their overrepresentation as a consequence of high recombination rates [26] . Mutators have been linked to the development of antibiotic resistance both in vitro and in vivo [13] , [27] , [28] , [29] , [30] , [31] , and have been reported to enhance genetic adaptation to CF airways through increased accumulation of new mutations [32] . However , comparisons between mutators and normo-mutators did not reveal any association between hypermutability and a particular distribution of mutations among genes , even for antibiotic resistance-related genes [32] . No study to date has linked hypermutability in CF adaptation to any specific adaptive mutation [17] , [32] , [33] , nor to any key adaptive trait in the transition to a chronic state of infection [33] . Our previous studies demonstrated a role of MRS deficiency in the acquisition of CF-related phenotypes under in vitro conditions such as mucoid conversion [34] , lasR inactivation [35] , [36] , and enhanced adaptability in biofilms [37] – all hallmarks of P . aeruginosa chronic airway infection . We also reported the ability of MRS deficiency to bias mutagenic pathways toward DNA simple sequence repeats ( SSRs ) , which gave specific mutational spectra under both in vitro [34] , [38] and in vivo conditions [18] , [39] . In view of the widespread effect of hypermutability on the process of adaptation to the CF lung [32] , it is important to elucidate the evolution of MRS-mutator strains in the course of CF chronic lung infections . Previous genome analyses of longitudinally collected P . aeruginosa from CF patients demonstrated intra-patient genomic diversity of clonal isolates , suggesting that within-host P . aeruginosa population dynamics are driven by clonal competition ( clonal interference ) and/or niche specialization ( adaptive radiation ) [18] , [40] , [41] , [42] . To further investigate these processes , Chung et al . compared the genomes of pairs of randomly selected contemporary isolates sampled from three chronically infected adult CF patients , and found that the pairs were differentiated by 1 , 54 , and 344 SNPs , respectively . In the latter case , both isolates were mutators [43] . Although mutators are frequently found in CF infections , no whole-genome studies have focused on the within-host evolution of mutators . Similarly , diversity of within-patient pathogen populations is relevant to planning of clinical intervention strategies , elucidation of transmission networks , and understanding of evolutionary processes , but no study to date has involved genome sequencing of a sufficiently large collection of P . aeruginosa isolates taken from the same patient at the same time point to facilitate an in-depth analysis of population diversity . We combined two distinct strategies for a genome-wide analysis of P . aeruginosa MRS-mutators: ( i ) a longitudinal analysis of two separate clonal lineages of mutators obtained from two CF patients; ( ii ) a within-host population analysis of a large collection of isolates obtained from a single sputum sample from each of the patients , to provide a snapshot of mutator population structure in the CF lung at a single time point . Whole-genome sequencing of 27 P . aeruginosa isolates allowed us to quantify the nature and extent of the genomic changes of MRS-mutator clones and provide a panorama of the high genomic diversity that shapes the structure of P . aeruginosa mutator populations during long-term adaptation to the CF airway environment .
To quantitatively describe the evolutionary processes of MRS-deficient strains during chronic airway infections , we performed longitudinal and cross-sectional analyses of clonal P . aeruginosa isolates collected from two CF patients , referred to here as CFA and CFD ( Figure 1 , and Materials and Methods ) . The cross-sectional study included 90 isolates obtained from a single sputum sample from each patient . These large collections were used to investigate the clonal genomic diversity within mutator populations in a single host at a single time point . Two different non-epidemic P . aeruginosa strains were collected from geographically distant locations , Argentina ( CFA ) and Denmark ( CFD ) , in which different therapeutic protocols are applied . We were thus able to analyze two independent mutator populations whose evolutionary histories were presumably subjected to common as well as patient-specific selective pressures . The collection from CFA included: ( i ) Two sequential isolates obtained in 2004 ( CFA_2004/01 ) and 2007 ( CFA_2007/01 ) . These isolates were characterized as MRS-mutators because they harbored missense mutations in the mutS and mutL genes and showed increased mutation rate ( Table 1 and Text S1 ) . ( ii ) A collection of 90 P . aeruginosa isolates obtained from a single sputum sample in 2010 ( CFA_2010 ) . The collection from CFD included: ( i ) One normo-mutable isolate obtained in 1991 ( CFD_1991/01 ) . ( ii ) Two sequential MRS-deficient mutators from 1995 ( CFD_1995/01 ) and 2002 ( CFD_2002/01 ) ( Table 1 and Text S1 ) that harbored the same mutS missense mutation [33] . ( iii ) A collection of 90 P . aeruginosa isolates obtained from a single sputum sample in 2011 ( CFD_2011 ) . The CFA and CFD collections covered periods of 6 and 20 yrs in the patients' lives , respectively . Based on previous studies indicating a doubling time of 115 min for P . aeruginosa in sputum [44] , we estimated that ∼36 , 500 and ∼91 , 400 duplication events occurred between the first and last isolates collected from CFA and CFD , respectively . Genotypic characterization of the two collections by various molecular methods ( Materials and Methods ) showed that each patient was chronically infected by a single , unrelated P . aeruginosa clone that persisted throughout the study period . The hexadecimal codes [45] for the SNP patterns of the CFA and CFD isolates analyzed are 2C32 and 249A , respectively . The proportion of MRS-deficient mutators present in each patient was determined based on the rifampicin mutation frequency of the 90 P . aeruginosa isolates of the CFA_2010 and CFD_2011 panels . All the CFA_2010 isolates showed mutation frequencies ( 1 . 7×10−5–7 . 6×10−6 ) consistent with a strong mutator phenotype . Similarly , in the CFD_2011 panel , strong mutator isolates ( 1 . 4×10−5–8 . 3×10−6 ) comprised ∼94% of the population . The remaining 6% showed mutation frequencies close to those observed in the prototypic wild-type normo-mutable strain PAO1 ( 3×10−8–1×10−7 ) . These findings were supported by the observed prevalence of mutators in 90 isolates obtained ≥6 months later from new sputum samples ( CFA_2011 and CFD_2012 ) ; 100% of CFA and 90% of CFD isolates displayed a strong mutator phenotype . To our knowledge , these are the highest proportions of mutators reported to date in large intra-patient populations of P . aeruginosa isolated from CF patients . A previous study , which analyzed sets of 40 isolates per sputum sample from 10 CF patients , reported proportions of 27% or less [8] . These findings indicate that P . aeruginosa can persist in chronic airway infections without the mutator phenotype , but in certain CF populations , such as those described here , MRS-deficient isolates may be prevalent and dominate the entire infecting population . To analyze the genomic evolution of CFA and CFD P . aeruginosa mutator lineages , we performed whole-genome sequencing of 13 CFA and 14 CFD isolates . From the CFA collection , we selected and sequenced the initial isolate CFA_2004/01 , the intermediate CFA_2007/01 , and 11 mutator isolates chosen randomly from the CFA_2010 population ( CFA_2010/01 , CFA_2010/11 , CFA_2010/26 , CFA_2010/31 , CFA_2010/32 , CFA_2010/40 , CFA_2010/43 , CFA_2010/72 , CFA_2010/78 , CFA_2010/82 , CFA_2010/87 ) . From the CFD collection , we selected and sequenced the initial normo-mutable isolate CFD_1991/01 , the intermediate mutators CFD_1995/01 and CFD_2002/01 , and 11 isolates from the CFD_2011 population consisting of five normo-mutators ( CFD_2011/04 , CFD_2011/11 , CFD_2011/45 , CFD_2011/57 , CFD_2011/95 ) and six randomly selected mutators ( CFD_2011/27 , CFD_2011/28 , CFD_2011/33 , CFD_2011/34 , CFD_2011/83 , CFD_2011/94 ) . The reads of CFA_2004/01 and CFD_1991/01 were assembled de novo , yielding genomes of 6 , 294 , 248 bp and 6 , 313 , 855 bp , respectively ( Table S1 ) , which were used as references in subsequent analyses . The sequences of the remaining CFA and CFD isolates were aligned against the corresponding references to assess the genetic changes accumulated in the two mutator lineages during the infection process ( Table S2 ) . Both lineages accumulated a high number of mutations during their evolution in CF airways in comparison with previously reported normo-mutable CF isolates such as the DK2 and PA14 clones [18] , [40] . The CFA collection had a total of 2 , 578 single-nucleotide polymorphisms ( SNPs ) and 544 1- to 10-bp insertion/deletion mutations ( microindels ) . The CFD collection had a total of 5 , 710 SNPs and 1 , 078 microindels ( Table S3 ) . We applied Bayesian statistical analysis to infer time-measured phylogenies [46] , resulting in estimated mutation rates of 106 SNPs/yr ( 4 . 2×10−9 SNPs/bp per generation ) for CFA and 89 SNPs/yr ( 3 . 2×10−9 SNPs/bp per generation ) for CFD . These findings indicate a mutation rate ∼40-fold higher than that ( 2 . 6 SNPs/yr ) reported previously for normo-mutable isolates obtained from CF chronic infections [18] . SNPs were used as phylogenetic markers to perform a maximum-parsimony reconstruction of the evolutionary relationship of CFA and CFD isolate groups , and to evaluate temporal changes in their population genetic structure ( Figure 2 ) . Alleles of P . aeruginosa reference strain PAO1 were used to root the trees . In both cases , essentially all SNPs ( >99 . 5% ) supported single phylogenetic trees . Interestingly , high genetic diversity was observed in both CFA and CFD P . aeruginosa intra-patient populations . CFA_2010 and CFD_2011 contemporary clones were grouped together into three and four distinguishable clusters , respectively: Clusters II , III , and IV for CFA ( Figure 2A ) and Clusters I , IV , V , and VI for CFD ( Figure 2B ) trees . Clusters were composed of genetically similar isolates , whereas more extensive genetic dissimilarities were observed between clusters . The branch lengths among clusters differed substantially , indicating uneven mutational loads in coexisting P . aeruginosa intra-patient populations . All CFD normo-mutable isolates were grouped together in Cluster IV ( Figure 2B ) . In spite of the normo-mutable phenotype , this cluster shared common branches with the intermediate mutator isolates CFD_1995/01 and CFD_2002/01 and with their contemporary mutator clones ( branches D , F , and H ) ( Figure 2B ) . These findings suggest that the normo-mutators arose from a mutator population of branch J at some point . Cluster IV showed the highest accumulation of mutations during the infection process , despite the low mutation frequencies of its members ( Figure 2B and Table S3 ) . The estimated time points of the most recent common ancestor ( MRCA ) of the CFA and CFD populations were 2002 and 1988 , respectively . Interestingly , each of these estimates coincided with the year at which the patient was diagnosed as chronically infected ( Figure 1 ) . This finding indicates that the divergent sub-lineages coexisted for many years – the same as the colonization period of the patient ( Figure 2 ) . The question arose whether the genetic diversity observed in the two CF intra-patient populations was associated with differing repertoires of mutated genes among contemporary isolates , or whether most mutations occurred in common genes among the clones . To distinguish between these possibilities , we first determined the set of genes of each CFA and CFD isolate that were altered by nonsynonymous SNPs and microindels . Minimum Spanning Trees ( MSTs ) were then constructed to illustrate the relationships among contemporary isolates and their corresponding ancestors based on the number of distinctive mutated genes . The CFA_2010 intra-patient P . aeruginosa mutator population ( Figure 3A ) was distributed in three main clusters , whereas the CFD_2011 population ( Figure 3B ) showed four clusters with all normo-mutable isolates grouped together . The structures of the two MSTs indicate high genetic diversification and a scenario in which even contemporary CFA and CFD mutator populations diversified through mostly different evolutionary pathways . We classified the SNPs according to their distribution in coding and noncoding regions and their effect in translation , to assess the selective forces acting on the CFA and CFD P . aeruginosa lineages . Most of the SNPs in CFA and CFD were found to occur within coding regions , and 58 . 0% and 60 . 2% ( respectively ) corresponded to missense mutations ( Figure S1 ) . The rates of nonsynonymous to synonymous mutations ( dN/dS ) were 0 . 68 for CFA and 0 . 79 for CFD lineages . Thus , most of the SNPs that became fixed in the P . aeruginosa mutator lineages were neutral mutations , with most of each genome showing a signature of purifying selection and/or genetic drift ( dN/dS<1; P = 5 . 2×10−56 and P = 1 . 5×10−47 , respectively ) . However , it is conceivable that positively selected genes remain “hidden” among the much larger number of genes that have accumulated mutations by genetic drift . Evolving populations of P . aeruginosa presumably accumulate adaptive mutations in response to the human host environment in which they propagate . We would therefore expect to observe parallelism in the adaptive genetic routes of the different lineages . To confirm such convergent evolution , we attempted to identify genes that underwent parallel mutation in the 10 sub-lineages ( four CFA sub-lineages and six CFD sub-lineages ) that coexisted over many years ( Figure 2 ) . We analyzed our dataset by selecting those genes that were independently mutated in at least half of the parallel evolving sub-lineages ( see Materials and Methods ) . Forty genes were found to be frequently mutated across the sub-lineages ( Table 2 ) , suggesting that the parallel mutation of these genes was due to positive selection for mutations . Consistently , the signature for selection for SNPs accumulated in the 40 genes ( dN/dS = 0 . 97 ) was significantly higher than the ratio obtained for SNPs affecting all other genes ( dN/dS = 0 . 75; P = 0 . 029 by Fisher's exact test ) , suggesting that these mutations were positively selected during evolution . Analysis of the 40 genes was further focused on those that were non-synonymously mutated in at least half of the 10 sub-lineages ( Figure 4 ) . Several of these genes were associated with functions related to CF host adaptation . In particular , ftsI , ampC , fusA1 , mexY , PA1874 , and PA0788 are involved in resistance to antibiotics commonly used in CF therapies , i . e . , betalactams , aminoglycoside , quinolones , chloramphenicol , trimethoprim , and imipenem [29] , [47] , [48] , [49] , [50] , [51] , [52] , [53] . Even though the GacA/GacS system is required for activation of genes involved in chronic persistence , gacS mutants are prone to generate stable and stress-tolerant small colony variants ( SCVs ) when growing in biofilms , exposed to stress factors [54] , [55] , or in vivo [56] , suggesting that the absence of GacS may confer some additional advantage for persistence in the CF lung . fusA1 encodes the elongation factor EF-G1A , which confers resistance to the antibiotic argyrin in P . aeruginosa [57] , [58] . Chung et al . recently reported independent fusA1 mutations in two CF patients and suggested that these mutations are involved in regulation of virulence through a ppGpp-dependent stringent response [43] . On the other hand , the gene pslA is involved in biofilm formation [59] , and cupC3 is associated with motility/attachment [60] . Lack of motility is a trait frequently observed in isolates from chronically colonized patients , and may give P . aeruginosa a survival advantage in chronic CF infection by enabling it to resist phagocytosis and conserve energy [61] . Alterations in several genes related to bacterial catabolism ( e . g . , aceE , gcvP1 , soxA , xdhB , PA0794 ) were also observed , suggesting that the inactivation of certain metabolic functions may be a common trait related to CF host adaptation ( see below ) . The concurrent alteration of specific genes or functions related to adaptation to the CF airway environment provides strong evidence for parallel evolution not only across CFA and CFD lineages , but also across intra-patient coexisting sub-lineages . Certain genes were convergently but exclusively mutated among CFD sub-lineages , e . g . , ampC ( beta-lactamase precursor ) , PA0788 ( penicillin binding protein ) , and PA3271 ( two-component sensor ) . These findings suggest the occurrence of in-host parallel evolutionary processes resulting from specific selective pressures from differential antibiotic treatments . Mutations in the global regulators mucA , algT , rpoN , and lasR are related primarily to adaptation to the CF airway environment [10] , [17] , [33] , [62] , [63] , [64] . However , our analyses did not reveal such mutations because they arose in the ancestral isolates before diversification into sub-lineages ( Table S4 ) . The entire population from CFA had a mutation in mucA , whereas the population from CFD had mutations in lasR and rpoN ( Table S4 ) . These findings indicate that mutations in these regulator genes were specifically fixed in the respective bacterial populations during early in-host evolution . Our previous in vitro studies showed that , in a MRS-deficient background , G∶C SSRs constitute hotspots capable of biasing mutagenesis toward a specific genetic pathway [34] , [38] . Our recent studies of P . aeruginosa PACS2 and epidemic DK2 strains demonstrated the same phenomenon at a genome-wide level in CF in vivo chronic airway infection [18] , [39] . The present study design allows examination of such a genome-wide effect of biased mutagenesis in large populations obtained at single time points , and observation of SSR instability in coexisting isolates . Analysis of types of mutations occurring in both the CFA_2010 and CFD_2011 isolates revealed a mutational spectrum typical of MRS-deficient strains . Transitions ( ∼80% ) and small indels ( 1–4 bp ) ( ∼16% ) were the most frequently observed mutations in both collections ( Figure 5C ) . The most prevalent transition was G∶C→A∶T , accounting for 65% and 62% of total transitions in CFA and CFD lineages , respectively . Of the 1–4 bp indels , >80% were located within a homopolymeric SSR , and ∼75% were located specifically in G∶C SSRs ( Figure 5A ) . In contrast , indels in A∶T SSRs accounted in average for 5 . 2% and 6 . 7% of the 1–4 bp indels in CFA and CFD isolates , respectively ( Figure 5A ) . Based on this strong skewing of MRS spectra toward small indels in G∶C SSRs , we selected homopolymeric G∶C SSRs of ≥6 bp , which were mutated in at least half of the coexisting isolates in both CFA and CFD lineages , and analyzed their mutational dynamics at the intra-population level . Eleven of these highly mutated G∶C SSRs harbored 2–5 distinct indel mutations accounting for independent mutational events ( Figure 5B ) . A single SSR was observed to be either unaltered or modified by different indel mutations even in coexisting isolates from the same cluster . Analysis of these 11 G∶C SSRs in the normo-mutable CFD_1991/01 isolate showed no mutations . Using genome data available online , we evaluated the occurrence of indel mutations in these G∶C SSRs in 12 normo-mutable P . aeruginosa strains ( PAO1 , PA14 , M18 , NCGM2 . S1 , B136-33 , RP73 , 39016 , PACSC2 , 2192 , C3719 , DK2 , LESB58; the latter five are normo-mutable isolates obtained from CF infections ) , whose genomes have been sequenced and are available in the Pseudomonas Genome Database ( www . pseudomonas . com ) [69] . Our survey revealed that these 12 strains harbored no indel mutations in the analyzed G∶C SSRs , even though large G∶C SSRs are considered to be “hotspots” for mutagenesis . One of the identified homopolymers is located in a gene ( PA4071/PADK2_03970 ) which has been previously suggested to be preferentially mutated in mutators and to represent a mutator-specific target of adaptive mutations [18] . These findings suggest a scenario in which MRS-deficient populations generates a vast of genetic diversity due to G∶C SSR instability . In this scenario , genes containing large G∶C SSRs constitute continual sources of genetic diversification primarily in mutator bacterial populations . We evaluated the dynamics of phenotypic changes in the 27 P . aeruginosa CFA and CFD isolates by determining global catabolic activities ( the “catabolome” ) . Biolog phenotype microarrays were used to monitor the catabolic profiles of each isolate with various C and N sources ( Table S7 ) . The total catabolic functions in the isolates were greatly reduced ( average reduction 73 . 5% and 63 . 8% , respectively ) in comparison with those of the CFA_2004/01 and CFD_1991/01 ancestors ( Figure 6 ) . This extensive loss of functions led to homogeneous populations in both the CFA and CFD lineages , with slight catabolome variation among clones . Catabolic function reduction thus appears to be a phenotypic pattern shared by CFA and CFD mutator lineages . Accordingly , genes related to catabolism were convergently mutated in both the CFA and CFD lineages ( Figure 4 ) . This phenomenon may be partially responsible for the decreased catabolic phenotype .
This study provides a complete panorama of the genomic diversity that shapes the structure of P . aeruginosa mutator populations during long-term adaptation to the CF airway environment . We combined a longitudinal study with an extensive cross-sectional approach , including multiple isolates obtained from single sputum samples , which allowed in-depth analysis of population diversity ( Figure 1 ) . We utilized P . aeruginosa panel collections from two chronically infected CF patients , CFA ( Argentinian ) and CFD ( Danish ) , with time spans of 6 yrs and 20 yrs , respectively , from initial to later stages of chronic infection . Our comprehensive study design included whole-genome sequencing and high-throughput phenotypic approaches , calculation of mutation frequencies , phylogenetic estimation of time points of sub-lineage diversification , and analysis of mutS and mutL genes to obtain a wide-ranging depiction of hypermutability in CF . We expected ( and confirmed ) that each of the two patients was infected by a single non-epidemic P . aeruginosa clone that did not present , during the initial stages of infection , the pathoadaptive mutations displayed by epidemic clones [42] , [70] , [71] , [72] . We were therefore confident that our analysis addressed specific and independent in-host evolutionary processes . Mutator strains were highly prevalent in both patients , essentially dominating the populations . The proportion of mutators was ≥90% in the single time point 90-isolate collections , indicating that mutators , once selected , dominated the CFA and CFD infecting populations . The observed prevalence of within-patient mutators was much higher than the values reported in previous studies [8] , [13] . These findings indicate that although P . aeruginosa may persist throughout the course of chronic infection without ever acquiring the mutator phenotype , mutator strains may become prevalent and even dominate the whole population under certain yet-unknown conditions . This concept is supported by the observation that two patients of different ages from geographically distant locations , infected with different non-epidemic P . aeruginosa clones and subjected to different therapeutic protocols , underwent overlapping evolutionary trajectories that led to complete domination of mutators . Recent reports have demonstrated high diversity at the phenotypic level among P . aeruginosa populations from CF lung infections [8] , [9] , [73] . However , there have been no genome-wide studies of such diversity in bacterial populations from the same clinical samples . The global picture of genetic structure of intra-patient mutator populations in the present study reveals significant genomic diversity driven by high accumulation of mutations ( Figures 2 and 3 ) , reflected by the typical MRS spectra ( Figure 5 ) . The distribution and combination of thousands of mutations result in a unique genotype for every isolate , allowing long-term persistence in the CF airway environment . The observed genomic variation into the CFA and CFD lineages indicates that the population structure in each case was not determined by homogeneous single dominating clones , but occurred through multiple evolutionary genetic pathways that adapted equally to the CF airway environment and allowed the coexistence of diverse subpopulations for many years . We determined that this high genomic diversity , to an equal degree in the two patients , spread out from the establishment of chronic infection . Interestingly , MRS genes were also characterized by the coexistence of multiple polymorphisms . However , underlying the polymorphisms within the MRS genes , there is an ancestral mutation that was fixed in each CFA and CFD population and is apparently responsible for hypermutability . These findings suggest the existence of common selective forces acting on MRS inactivation in the two patients . The long-term evolution of the P . aeruginosa CFA and CFD lineages was signed mainly by purifying selection and/or genetic drift . There are conflicting reports regarding whether genomic evolution of CF isolates shows signatures of positive selection [10] or ( in contrast ) genetic drift and/or purifying selection [32] . Our results strongly support the latter concept; i . e . , that genomic signatures of purifying selection and/or genetic drift are not inherent consequences of mutators , but are characteristic of the genetic adaptation processes underlying P . aeruginosa persistence in chronic lung infections . According to our observations , the large number of mutations were for the most part distributed randomly among the P . aeruginosa mutator genome ( Figure 3 ) . However , we also identified a group of genes that were convergently mutated in multiple genomes by independent mutational events ( Table 2 and Figure 4 ) . Most of these genes code for functions related to pathogenicity ( e . g . , antibiotic resistance , virulence , motility , attachment ) , suggesting that they were positively selected as beneficial mutations . We note that five of the genes ( ampC , ftsI , fusA1 , PA3271 , PA2018 ) were also found to be mutated in isolates obtained from the epidemic DK2 clone [18] , providing evidence of parallel evolution for certain specific traits among different P . aeruginosa lineages . As we have reported recently for the PACS2 [39] and epidemic DK2 P . aeruginosa strains [18] , the impact of hypermutability on the evolution of the CFA and CFD lineages is reflected by the high tendency of G∶C SSR-containing sequences to be mutated ( Figure 5A ) . This finding confirms that genes which maintain G∶C SSRs in their coding region and/or in neighboring regulatory sequences are highly unstable in an MRS-deficient background and may be mutator-specific targets of adaptive mutations . This concept is extended here by the demonstration that large G∶C SSRs , as DNA sequences per se , are highly polymorphic in single time point populations , indicating that they are continual sources for diversification ( Figure 5B ) . This SSR-driven diversity is not observed in genomes of other P . aeruginosa strains , even of normo-mutable CF clones . Our present and previous results [18] , [34] , [38] , [39] , taken together , demonstrate a clear association between MRS-deficiency and G∶C SSR instability , which exerts a global effect along the entire genome . The impact of hypermutability during evolution of P . aeruginosa in the CF airway environment is not simply a major , rapid acquisition of mutations in quantitative terms . In contrast with SNPs , indels that are not multiples of three produce frameshifts in the coding sequence of genes and thereby affect gene function . Indels in G∶C SSRs may play an important role in the evolutionary process and in relation to mutator competitiveness . Along this line , nine of the 11 analyzed G∶C SSRs ( Figure 5B ) were located in coding regions of the genome . Five of these genes are predicted to encode for hypothetical proteins with no assigned function . On the other hand , some G∶C SSRs were positioned in genes functionally related to transcriptional regulators ( PA1490 ) , adaptation-protection ( PA1127 ) , membrane proteins , and transport of small molecules ( PA1626 , PA2203 ) . A small percentage ( 6% ) of the CFD population has a reduced mutation frequency similar to that of normo-mutable strains . This subpopulation , which is grouped in a single cluster ( Cluster IV in Figure 2 ) , had the highest accumulation of mutations observed in the whole CFD collection ( Table S3 ) . This observation posed the question whether the mutational load of these clones is too heavy to continue supporting a mutator phenotype . However , the normo-mutable subpopulation carried the same mutS loss-of-function mutation ( Table 1 ) as the mutator isolates . Phylogenetic analysis indicated that isolates from Cluster IV had arisen from a mutator subpopulation at some undetermined point in branch J ( Figure 2 ) . Our sequencing data suggested that the most feasible explanation is the emergence of secondary mutations , in genes not belonging to the MRS , that compensate for mutS hypermutability , since neither reversion of the original −CG1551 nor duplication of the mutS gene was observed in normo-mutable genomes . Although the new +CC334 mutation restored in part the mutS reading frame , this fact did not account for the reduction in mutation frequency observed in these clones . In a previous study , E . coli MRS-deficient populations in vitro evolved a compensation of the mutator phenotype based on secondary mutations in genes related to oxidative stress response which , although selected for different increasing-fitness traits , resulted in a reduction of the mutation rate [74] . Future studies will elucidate the mechanisms underlying the observed reduction in mutation frequency of these mutS-deficient P . aeruginosa strains . How are these mutator populations maintained over extended periods of time , and even able to accumulate greater numbers of mutations ? The chronically infected CF lung is a uniquely challenging habitat in which P . aeruginosa must cope with aggressive immune system responses , intense antibiotic therapies , and/or competition with other resident microorganisms . This environment presents stressful and variable conditions , and multiple mutations may occur that increase fitness . On the other hand , favorable nutritional conditions of CF airways [75] , [76] may sustain the inactivation of certain functions that are no longer necessary for survival , particularly in this environment . In this regard , CF isolates have been reported to accumulate a high number of auxotrophies – higher than in other studied environments [77] . The results obtained here for CFA and CFD catabolomes indicate overall reduction of total catabolic activities ( Figure 6 ) . Reduction of catabolic functions appears to be part of a more general adaptive process of P . aeruginosa residing for long periods in the CF lung , because such reduction has also been observed in normo-mutable strains [42] . The large genome of P . aeruginosa may have the potential to undergo reductive evolution , with elimination of functions that are redundant and/or dispensable in the CF host environment , thereby buffering the heavy mutational load observed [78] . Based on these considerations , we hypothesize that , during long-term evolution of P . aeruginosa in CF airways , the increased availability of certain beneficial mutations , in combination with a whole-genomic signature of neutral evolution , provided favorable conditions to increase the percentage of mutators , until reaching a frequency at which mutators dominated both the CFA and CFD populations . P . aeruginosa adaptation to the CF lung is presumably manifested through selection of multiple genetic combinations [79] , and hypermutability is consequently maintained over time as a constant source of genetic variation .
Clinical P . aeruginosa isolates were obtained from sputum samples from two CF patients at the Hospital de Niños Santísima Trinidad ( Córdoba , Argentina ) ( patient CFA ) and the Copenhagen CF Centre at Rigshospitalet ( Copenhagen , Denmark ) ( patient CFD ) . Patient age at the time of the first isolate collection was 8 and 23 yrs , respectively . The onset of chronic infection with P . aeruginosa was 2001 and 1986 , respectively . Chronic pulmonary infection was defined as the persistence of P . aeruginosa in sputum for 6 consecutive months , or less if the persistence was combined with presence of two or more precipitating antibodies against P . aeruginosa [33] . The criteria for choosing these patients were: ( i ) chronic airway infection by P . aeruginosa; ( ii ) absence of transmissible P . aeruginosa clones; ( iii ) presence of single clonal P . aeruginosa infecting lineages throughout the course of infection; ( iv ) appearance of MRS-deficient P . aeruginosa mutators during the course of infection . Sputum samples were collected by expectoration during routine hospital visits , stored on ice , and processed immediately . Sputa were liquefied by addition of an equal volume of Sputolysin ( Calbiochem ) , diluted , and plated onto Pseudomonas isolation agar ( BD Biosciences ) , which promotes growth of P . aeruginosa and other Pseudomonas species . For cross-sectional analysis , 90 isolates were taken randomly from a single sputum sample . Isolates were maintained and stored at −70°C in glycerol stock solution . For assays , isolates were subcultured from the frozen stocks onto Luria Bertani ( LB ) agar plates , and inocula were prepared from overnight cultures in LB broth at 37°C with appropriate aeration . Genotyping analysis of the P . aeruginosa isolates was performed by Randomly Amplified Polymorphic DNA ( RAPD ) , using primer 272 ( Table S5 ) [5] , [80] . Banding patterns from electrophoretic gels were analyzed by GelPro Analyzer V . 6 . 3 software . Epidemiological relatedness was evaluated by pulsed-field gel electrophoresis ( PFGE ) using SpeI enzyme as described previously [81] . DNA macrorestriction patterns were interpreted according to the criteria of Tenover et al . [82]; i . e . , isolates with PFGE patterns differing by ( i ) 1–3 bands are closely related clones; ( ii ) 4–6 bands are possibly related clones; ( iii ) ≥7 bands belong to different strains . Clonal relatedness among the isolates selected for genome sequencing analysis was evaluated by single-nucleotide polymorphism ( SNP ) typing using AT biochips ( Clondiag Chip Technologies , Germany ) [83] . Strain assignment was performed by visual array analysis using a hexadecimal code as described previously [45] . Mutation frequencies were estimated as described previously [13] , [15] . In brief , five independent colonies of each isolate were grown overnight in 10 ml LB medium at 37°C with appropriate aeration . Cultures were collected by centrifugation at 3000 rpm for 10 min and resuspended in 1 ml LB medium . Serial ten-fold dilutions were plated on LB agar with and without 300 µg/ml rifampicin . After 36 h incubation ( 48 h for slow-growing strains ) at 37°C , colonies were counted and the mean percentage of mutants was estimated . Strains were considered hypermutable if the mutation frequency was at least 20-fold higher than that of control strain PAO1 [13] . Frequencies were determined from two independent experiments . Genomic libraries were prepared as described previously [42] and sequenced on an Illumina Hiseq2000 platform , generating 100 base paired-end reads using a multiplexed protocol . A total of 224 million paired-end reads were generated to yield average genomic coverage of 33–207 fold ( all average coverage depths were ≥82× except for CFD287_2011/94 ( 33× ) and CFD_2011/27 ( 52× ) ) . Genome sequence reads were deposited in the European Nucleotide Archive ( ENA/SRA ERP002379 ) . Illumina reads from isolates CFA_2004/01 and CFD_1991/01 were de novo assembled using Velvet software V . 1 . 2 . 07 [84] using the following settings: -ins_length 238 and 244 for CFA_2004/01 and CFD_1991/01 , respectively; -ins_length_sd 54 . 4 and 50 . 3 for CFA_2004/01 and CFD_1991/01 , respectively; -exp_cov 291 . 1 and 296 . 7 for CFA_2004/01 and CFD_1991/01 , respectively; -cov_cutoff 10; -min_contig_lgth 500 . Selected kmer sizes were 51 and 39 for CFA_2004/01 and CFD_1991/01 , respectively . Each de novo assembly was used as a reference genome sequence to map reads from the remaining CFA and CFD genome sequences using Novoalign V . 2 . 08 . 02 ( Novocraft Technologies ) [85] . Pileups of the mapped reads were processed by SAMtools V . 0 . 1 . 7 [86] . SNPs were identified by the varFilter algorithm in SAMtools ( samtools . pl varFilter -d 3 -D 10000 ) , and only unambiguous SNPs with quality scores ( Phred-scaled probability of sample reads used as homozygous reference ) of ≥50 ( i . e . , P≤10−5 ) were retained . Read alignments surrounding all putative indels were realigned using GATK V . 1 . 0 . 5083 [87] , and microindels were extracted from the read pileup using the following criteria: ( i ) quality score ≥500; ( ii ) root-mean-square ( RMS ) mapping quality ≥25; ( iii ) support by ≥20% of the covering reads . The false-negative rates obtained were 2% and 3% by in silico introduction of random base-substitutions and microindels ( lengths 1–10 bp ) , respectively . All sites with putative polymorphisms in the pileup of reads from the reference were excluded to avoid false-positives resulting from errors in the reference assembly or general mapping errors . MUMmer3 [88] was used for whole-genome alignments . Mutations were described according to their relative gene positions in orthologs of P . aeruginosa reference strains PAO1 , PA14 , and LESB58 with completed genome sequences [69] . If no ortholog was found in the reference strains , the mutation was described by the GenBank ID of the homolog sequence ( Genbank_ID:ORF_name: Position ) or as “Not annotated” . Phylogenetic trees based on the SNP mutations identified in each alignment were constructed using maximum-parsimony analysis as described previously [42] . For calculation of selection coefficients ( dN/dS ratio ) , we assumed that codon usages were identical to those in strain PAO1 , in which 25% of random mutations are synonymous [42] , and that the probability of the observed number of nonsynonymous SNPs , given the expected number of SNPs , is calculated from the Poisson distribution . Bayesian analysis of evolutionary rates was performed using BEAST V . 1 . 7 . 2 [46] . The BEAST program was run with the following user-determined settings: a lognormal relaxed molecular clock model , which allows rates of evolution to vary among the branches of the tree , and a HKY substitution model , which distinguishes between the rate of transitions and transversions and allows unequal base frequencies . Mutation rates were calculated from chains of 100 million steps , sampled every 5 , 000 steps . The first 10 million steps of each chain were discarded as a burn-in . To identify genes subject to convergent evolution , we picked out those that were mutated in at least half of the in parallel evolving sub-lineages CFA I–IV and CFD I–VI . The sub-lineages/clusters were defined as follows ( see Figure 2 ) . CFA lineage: Cluster I: branches Z , Y; Cluster II: branch A; Cluster III: branches C , D , E , F , G; Cluster IV: branches N , J , T , U , R , S , O , Q , K , I , M . CFD lineage: Cluster I: branch A2; Cluster II: branch C; Cluster III: branch E; Cluster IV: branches J , I , L , M , N , K , P , O; Cluster V: branches U , T , W , V , X; Cluster VI: branches R , G , Q . For example , gene aceE was hit by independent mutations in CFA branch J ( sub-lineage CFA-IV ) , CFA branch Y ( sub-lineage CFA-I ) , CFD branch C ( sub-lineage CFD-II ) , CFD branch E ( sub-lineage CFD-III ) , and CFD branch T ( sub-lineage CFD-V ) . To rule out the possibility that high accumulation of mutations resulted from particularly large gene sizes , mutations from large genes ( ≥10 kb ) were excluded from the analysis . MSTs for the set of mutated genes within each CFA and CFD lineage were obtained using Prim's algorithm ( Prim , 1957 ) as implemented in the Info-Gen program [89] . In this model , nodes represent an isolate's set of mutated genes , the distance between a pair of nodes is shown as the number of distinctive mutated genes , and nodes are connected in such a way that the sum of the distances is minimized ( Table S6 ) . The network connects each genotype to all other genotypes through a pathway of mutated genes . The CFA_2004/01 and CFD_1991/01 genomes were used as starting points of the network of the corresponding MSTs for each lineage . The number of mutated genes along the network was used as a measure of divergence between two given genotypes . Genomic DNA of P . aeruginosa CF isolates was extracted using a DNA Isolation Kit ( Qiagen ) . Primers used for PCR amplification and DNA sequencing are listed in Table S5 . PCR amplifications were performed with the following conditions: 8 min at 95°C , 33 cycles of 1 min at 94°C , 1 min 20 sec at 60°C , 2 min at 72°C , and a final extension of 10 min at 72°C . PCR products were cleaned with a Gel Purification Kit ( Qiagen ) , and both strands were sequenced directly using the same PCR primers ( DNA Sequencing Facility , Univ . of Chicago , IL , USA ) . To score mutations within the gene , sequencing results were compared with the corresponding gene sequence of strain PAO1 ( www . pseudomonas . com ) using the BLAST program of the NCBI database ( www . ncbi . nlm . nih . gov/blast/ ) . Two constructs of plasmid pMC5-MutS [90] , which contains the full coding region of the mutS gene from strain PAO1 , were generated by introducing −CG1551 and +CC334−CG1551 mutations to produce plasmids pMC5-MutS−CG1551 and pMC5-MutS+CC344−CG1551 , respectively . To introduce these mutations , the mutS genes from CFD_2011/27 and CFD_2011/11 isolates were amplified by PCR using oligonucleotides mutS-for1 and mutS-rev4 ( Table S5 ) , ligated to the pGEM-T Easy vector ( Promega ) , and cloned in the NdeI/EcoRI restriction sites of pMC5-mutS , in which the mutS gene has been digested with NdeI/EcoRI . Both plasmids were propagated on E . coli DH5α ( Invitrogen ) transformed by heat shock by standard procedures . To evaluate the genetic basis of the mutator phenotype , plasmids pMC5-mutS [90] and pMC5-mutL [91] were successively transferred into the mutator CFA and CFD isolates . To analyze +CC344−CG1551 mutations in the mutS gene , plasmids pMCS-MutS−CG1551 and pMCS-MutS+CC344−CG1551 were independently transferred into reference strain MPAO1MS . All plasmids were transferred by electroporation as described by Choi and Schweizer [92] , and transformed strains were selected on LB agar plates supplemented with 100 mg/ml gentamicin . Complementation was checked by the rifampin test described above . For each strain , complementation was confirmed for three independent transformed colonies . Phenotype MicroArrays PM1 and PM3B ( Biolog; Hayward , CA , USA ) were performed according to the manufacturer's instructions [93] , [94] . Export of Omnilog data was performed using Omnilog OL-FM/Kin software V . 1 . 20 . 02 ( Biolog ) . Phenotypes were determined based on the parameter “ave area” ( the area beneath the respiration curve of reduced tetrazolium vs . time ) . For comparison of clinical isolates , data were exported after 72 h incubation . Data analysis and statistical analysis were performed using R Project V . 2 . 10 . 0 ( http://www . R-project . org ) . The R packages used for analysis were “bioDist” ( B . Ding , R . Gentleman , V . Carey ) . Total catabolic function was calculated as described previously [95] . Statistical analyses were performed using a two-tailed T-test adjusted by Bonferroni correction . Differences with P≤0 . 05 were considered statistically significant . The P . aeruginosa isolates were obtained from sputum samples from one CF patient at the Hospital de Niños Santísima Trinidad ( Córdoba , Argentina ) ( patient CFA ) and one CF patient at the Copenhagen CF Centre at Rigshospitalet ( Copenhagen , Denmark ) ( patient CFD ) , as byproducts of the routine established for bacterial typing and antimicrobial susceptibility testing . I . e . , sputum sampling was not performed for the purposes or intent of the present study; isolates recovered from these sputa were simply derivatives of routine CF patient therapeutic controls . The therapeutic treatments of the two patients were not modified in any way as a consequence of the results obtained in this study . The research protocols followed in this study were approved and reviewed by the Ethics Committee of the Hospital de Niños Santísima Trinidad , Córdoba , Argentina and the local ethics committee , Region Hovedstaden , Copenhagen , Denmark ( H-A-141 and H-1-2013-032 ) . Both of the patients gave informed consent . | Patients with cystic fibrosis ( CF ) are often colonized by a single clone of the common , widespread bacterium Pseudomonas aeruginosa , resulting in chronic airway infections . Long-term persistence of the bacteria involves the emergence and selection of multiple phenotypic variants . Among these are “mutator” variants characterized by increased mutation rates resulting from the inactivation of DNA repair systems . The genetic evolution of mutators during the course of chronic infection is poorly understood , and the effects of hypermutability on bacterial population structure have not been studied using genomic approaches . We evaluated the genomic changes undergone by mutator populations of P . aeruginosa obtained from single sputum samples from two chronically infected CF patients , and found that mutators completely dominated the infecting population in both patients . These populations displayed high genomic diversity based on vast accumulation of stochastic mutations . Our results are in contrast to the concept of a homogeneous population consisting of a single dominant clone; rather , they support a model of populations structured by diverse subpopulations that coexist within the patient . Certain genes involved in adaptation were highly and convergently mutated in both lineages , suggesting that these genes were beneficial and potentially responsible for the co-selection of mutator alleles . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"bacterial",
"genomics",
"organismal",
"evolution",
"microbial",
"mutation",
"microbial",
"evolution",
"genetics",
"biology",
"and",
"life",
"sciences",
"microbiology",
"genomics",
"evolutionary",
"biology",
"bacterial",
"evolution",
"microbial",
"genomics"
] | 2014 | Coexistence and Within-Host Evolution of Diversified Lineages of Hypermutable Pseudomonas aeruginosa in Long-term Cystic Fibrosis Infections |
The relative roles of the endosomal TLR3/7/8 versus the intracellular RNA helicases RIG-I and MDA5 in viral infection is much debated . We investigated the roles of each pattern recognition receptor in rhinovirus infection using primary bronchial epithelial cells . TLR3 was constitutively expressed; however , RIG-I and MDA5 were inducible by 8–12 h following rhinovirus infection . Bronchial epithelial tissue from normal volunteers challenged with rhinovirus in vivo exhibited low levels of RIG-I and MDA5 that were increased at day 4 post infection . Inhibition of TLR3 , RIG-I and MDA5 by siRNA reduced innate cytokine mRNA , and increased rhinovirus replication . Inhibition of TLR3 and TRIF using siRNA reduced rhinovirus induced RNA helicases . Furthermore , IFNAR1 deficient mice exhibited RIG-I and MDA5 induction early during RV1B infection in an interferon independent manner . Hence anti-viral defense within bronchial epithelium requires co-ordinated recognition of rhinovirus infection , initially via TLR3/TRIF and later via inducible RNA helicases .
Human rhinovirus ( RV ) belongs to the Picornaviridae family and are implicated in an extensive range of human respiratory disorders including the common cold , viral bronchiolitis , and exacerbations of asthma and chronic obstructive pulmonary disease [1]–[5] . RV are classified as major or minor group based on receptor usage , or RNA identity as RV-A and RV-B . RVs of both major and minor groups are associated with human disease . Recently , this phylogeny been changed to include the newly designated RV-C group which represent a distinct group of RV [6] . RV of all groups generally infect the epithelial cells of both the upper and lower airway , and are responsible for the induction of a range of mediators including pro-inflammatory cytokines and growth factors [7]–[11] , type I interferon ( IFN ) -β and type III IFN-λs [12] . Pro-inflammatory cytokines contribute to the duration and severity of RV induced illnesses [13]–[16] . Recently , primary human bronchial epithelial cells ( HBECs ) from asthmatics were found to be defective in IFN-β and IFN-λ mRNA and protein , [17] , [18] , providing a likely explanation for the increased vulnerability to virus induced asthma exacerbations and enhanced symptom severity observed [16] , [19] . Understanding the mechanisms responsible for these deficiencies in asthma , as well as identifying new anti-inflammatory therapies requires a detailed understanding of the innate reponses to RV infection . Much is now known about the signal transduction pathways utlised by viruses to induce cytokines and IFNs . RNA viruses are initially sensed through pattern recognition receptors ( PRRs ) , such as recognition of dsRNA by endosomal Toll-like receptor ( TLR ) -3 , [20] , [21] or ssRNA by endosomal TLR7/8 [22] , [23] . TLR3 utilises the adaptor TIR domain-containing adapter inducing IFN-β ( TRIF ) , to activate IκB kinase ( IKK-ι/ε ) and TANK binding kinase-1 ( TBK-1 ) , and IKK-β activating interferon regulatory factor ( IRF ) -3 and NF-κB , transcription factors required for IFN-β gene expression . Within the intracellular compartment , exists a second set of PRRs , the RNA helicases , including retinoic acid inducible gene ( RIG-I ) [24] , melanoma differentiation associated gene-5 ( MDA5 ) [25] , and the inhibitory protein LGP2 [26] , [27] . The helicases signal via their caspase recruitment domains ( CARD ) , to adaptor inducing interferon-β ( CARDIF ) [28] , ( also known as IPS-1 , MAVS and VISA , [29]–[31] ) , and activate TBK1 , IKK-ι/ε and IKK-β , and hence IRF3 and NF-κB . Both RIG-I and MDA5 have been implicated in IFN-α/β production in various model systems . MDA5 recognises high molecular weight dsRNA [32] , while the specificity of RIG-I has been marked with controversy . While originally identified as a dsRNA binding helicase [24] , RIG-I has recently been shown to bind low molecular weight dsRNA [32] and also 5′-triphosphorylated ssRNA [33] , [34] . The 5′-triphosphorylated ssRNA binding preferences of RIG-I suggest it is unable to recognize Picornavirus infections [33] , [35]; which do not synthesis 5′-triphosphorylated RNA molecules . The relative importance of TLR3 , MDA5 or RIG-I in viral infections has been partly defined by cells derived from TLR3−/− [21] , RIG-1−/− or MDA5−/− mice [35] , [36] , however the importance of each PRR , including their exclusive or redundant roles in various infection models , and their direct relevance to human disease remains a subject of much debate . In order to understand the recognition of RV infection , and the induction of both pro-inflammatory cytokines and IFNs , we investigated the role of TLR3/7/8 , RIG-I and MDA5 in the innate response to RV infection in primary HBECs , the target cell for RV infection within the lower airway in vivo . We found that HBECs did not respond to the ligand R-848 . Importantly , TLR3 , and RNA helicase mediated signaling was required for maximal IFN-β , IFN-λ and pro-inflammatory cytokine gene expression , showing that RIG-I is required for anti-viral defense against Picornaviruses . Furthermore , RIG-I and MDA5 were virus inducible genes , induced early via a TLR3/TRIF pathway , indicating that TLR3 acts as an initial endosomal sensor and must induce the RNA helicases for maximal anti-viral defense during the course of infection . Thus the innate response to RV infection requires co-ordinated endosomal , and cytoplasmic recognition pathways , both of which contribute to IFN and cytokine production .
We first sought to assess the relationship between TLR3/7/8 , and the RNA helicases in RV infection in bronchial epithelial cells . Initial studies showed that HBECs encoded mRNA and protein for TLR3 , but did not induce IFN-β , IFN-λ , RIG-I or MDA5 mRNA in response to the TLR7/8 ligand R848 ( Table S1 in Supporting Information S1 ) , consistent with other studies showing that lung bronchial epithelial , alveolar and airway smooth muscle cells do not respond to TLR7/8 agonists [9] , [37] , [38] . The presence and/or absence and virus induction of TLR3 and RNA helicases was then investigated . Time course analysis in HBECs demonstrated that IFN-β was induced by 8 h post infection , and 4 h for IFN-λs post RV1B infection . IFN-β peaked at 12 h and remained at high levels until 48 h , while the IFN-λs remained elevated from 12 h and peaked at 48 h ( Figure 1A–C ) . In the same experiments , RIG-I and MDA5 mRNA levels were also measured , and were induced by RV1B by 8 h , peaked by 18 h post infection and remained at high level until 48 h post infection ( Figure 1D–E ) . At the protein level , RIG-I and MDA5 protein were both observed by 8 h following RV1B infection , and showed maximal levels at 18–24 h ( Figure 1F ) . Uninfected cells and cells sampled at time 0 h exhibited almost non-detectable expression of RIG-I and MDA5 protein , suggesting that in the absence of active infection , RIG-I and MDA5 proteins are absent or expressed at very low level . TLR3 protein however was present in both infected and uninfected cells , and the levels did not change over the time course of RV1B infection . Similar data was observed for TLR3 mRNA ( data not shown ) . The cytoplasmic staining of RIG-I and MDA5 were confirmed using immunofluorescence , with RV1B and IFN-β inducing both RIG-I ( Figure 1G upper panel ) and MDA5 ( Figure 1G lower panel ) at 24 h post treatment in HBECs , compared to cells treated with medium , or RV1B infected cells stained with a secondary antibody only . In order assess the baseline expression and the RV mediated induction of RIG-I and MDA5 protein in bronchial epithelium in vivo , bronchial biopsies were taken from 15 normal adult volunteers before experimental RV16 infection ( baseline ) or at day 4 post infection and stained for RIG-I and MDA5 by immunohistochemistry . Representative staining of biopsy samples are presented in Figure 2A–E . The degree of epithelial staining for RIG-I and MDA5 protein were scored quantitatively , and presented in Figure 2F . RIG-I ( A ) and MDA5 ( C ) had little staining at baseline , and MDA5 was increased at day 4 post RV16 infection , ( Figure 2D ) . Scoring of columnar epithelial staining ( F ) showed that MDA5 at day 4 was significantly higher than baseline ( p<0 . 05 ) , however RIG-I levels were not significantly different at day 4 versus baseline ( Figure 2B ) . The PRRs involved in RV infection and induction of innate responses are largely unknown . In order to assess the role of TLR3 , RIG-I and MDA5 in RV induced IFNs we used RNA interference with specific small interfering RNA ( siRNA ) to knockdown each PRR in HBECs in vitro , prior to RV infection . Initial experiments demonstrated that siRNA generated >75% knockdown of target mRNA at 24 h post treatment , and this knockdown was evident until 48 h post treatment ( data not shown ) . Therefore , siRNA was delivered 24 h before infection , and total RNA harvested 24 h post RV1B infection , giving a total siRNA transfection time of 48 h . The knockdown of each target mRNA was confirmed in each experiment , and knockdown of target protein also confirmed at 48 h post transfection ( Figure 3A–H ) . Also , experiments were performed in the absence of RV infection to examine the effects of siRNA on endogenous IFN and pro-inflammatory cytokine gene expression . Each siRNA did not significantly induce any of the IFNs or pro-inflammatory cytokines studied ( Table S2 & S3 in Supporting Information S1 ) . Furthermore , RIG-I and MDA5 siRNA were highly specific , RIG-I siRNA did not affect endogenous MDA5 gene expression , and MDA5 siRNA did not affect endogenous RIG-I gene expression ( data not shown ) . TLR3 , RIG-I and MDA5 siRNA all reduced RV1B induced IFN-β compared to control siRNA ( Figure 4A , D , G respectively ) . In contrast , siRNA targeting RIG-I did not reduce IFN-λ1 mRNA , ( Figure 4E ) however siRNA specific for TLR3 and MDA5 reduced IFN-λ1 ( Figure 4B , H respectively ) . RIG-I siRNA enhanced RV induced IFN-λ2/3 mRNA ( Figure 4F ) , while both MDA5 and TLR3 reduced RV1B induced IFN-λ2/3 ( Figure 4C , I ) . These data suggest that TLR3 , RIG-I and MDA5 are all required for IFN-β , and TLR3 and MDA5 for IFN-λ , however the importance of RIG-I in IFN-λs is less clear . To confirm these findings , we next used siRNA specific for the TLR3 adaptor TRIF , and the RNA helicase adaptor Cardif . We found that RV1B induced IFN-β and IFN-λ1 mRNA expression was inhibited by both siRNA to TRIF and Cardif , while IFN-λ2/3 was inhibited by TRIF siRNA only compared to control siRNA ( Figure S1 ) . The role of TLR3 , RIG-I and MDA5 on the T cell chemokines rantes and IP-10 , and the neutrophil chemokines IL-8 and ENA-78 were also investigated . Figure 5 shows that TLR3 , RIG-I and MDA5 siRNA all reduced RV1B induced rantes compared to control siRNA ( Figure 5A , E , I ) . Likewise , TLR3 , RIG-I and MDA5 siRNA all reduced RV induced IP-10 ( Figure 5B , F , J ) . TLR3 , RIG-1 and MDA5 siRNA also reduced RV induced IL-8 mRNA ( Figure 5C , G , K ) . Finally TLR3 , RIG-I and MDA5 siRNA also reduced RV1B induced ENA-78 compared to control siRNA ( Figure 5D , H , L ) . In support of the above , siRNA specific for Cardif and TRIF reduced all RV1B induced pro-inflammatory cytokines compared to control siRNA ( Figure S2 ) . As RIG-I and MDA5 were required for RV1B induced IFN-β and both IFN-β and IFN-λ gene expression respectively , we next reasoned that their role was not redundant and that abrogation of either RNA helicase would result in increased RV replication in HBECs . Figure 6A demonstrates that after 24 h post infection , RV16 RNA levels were increased after transfection with siRNA specific for RIG-I compared to control siRNA . RV1B RNA levels were also increased . Also , using siRNA specific for MDA5 , an increase in RV1B RNA was observed , compared to cells transfected with control siRNA , and RV16 RNA levels were slightly increased compared to control siRNA . In the same experiments , virus release was also determined by titration assay , 48 h after infection . Figure 6B demonstrates that transfection with RIG-I specific siRNA resulted in increased RV16 release compared to control siRNA and slightly increased RV1B virus release . Conversely , transfection with MDA5 specific siRNA resulted in higher RV1B release , and a small increase in RV16 virus release compared to control siRNA . Furthermore , transfection of the bronchial epithelial cell line BEAS-2B , with dominant negative RIG-I ( RIG-IC ) , resulted in enhanced replication of RV16 RNA ( Figure 6C ) , at 24 h and RV16 virus release at 48 h post infection ( Figure 6D ) . Transfection of constitutively active RIG-I ( ΔRIG-I ) resulted in up to 60 fold suppression of RV1B and 47 fold suppression of RV16 virus release at 48 h ( Figure S3 ) . This data further implicates the RNA helicase RIG-I in RV recognition and induction of anti-viral activity . As RV1B infection induced both early RIG-I and early MDA5 protein and mRNA production , and both are required for maximal anti-rhinoviral activity , we sought to identify the receptor ( s ) responsible for RNA helicase induction . As RV enters via the endosome , we hypothesized that TLR3 was responsible for increased RIG-I and MDA5 gene expression . In order to investigate the relationship between TLR3/TRIF signaling and RIG-I and MDA5 gene expression in RV infection , we next assessed if specific knockdown of TLR3 and the adaptor TRIF affected RV1B induced RNA helicase gene expression . Figure 7 demonstrates that siRNA specific for TLR3 reduced RV1B induced RIG-I mRNA versus control siRNA , ( Figure 7A ) , and reduced RV1B induced MDA5 mRNA ( Figure 7B ) . Consistent with the sequential affect of TLR3 on RNA helicase induction , we found that RIG-I specific siRNA did not affect TLR3 mRNA levels compared to control siRNA ( Figure 7C ) , and MDA5 specific siRNA also did not affect TLR3 mRNA levels compared to control siRNA ( Figure 7D ) . Furthermore , siRNA specific for the TLR3 adaptor TRIF , reduced RV1B induced RIG-I mRNA ( Figure 7E ) and also RV1B induced MDA5 mRNA compared to control siRNA ( Figure 7F ) . In each experiment , knockdown of TRIF mRNA by TRIF specific siRNA was confirmed , and we also confirmed knockdown of TRIF protein at 48 h post transfection ( Figure 3G , H ) . Using a plasmid encoding constitutively active TRIF ( ΔTRIF ) , the reverse of the above results were obtained in HBECs ( Figure 7G , H ) . Transfection with ΔTRIF significantly increased both RIG-I and MDA5 gene expression , compared to empty vector control ( pUNO1 ) . Furthermore , incubation of HBECs with the TLR3 ligand polyIC induced RIG-I mRNA from 4–24 h post treatment ( Figure 7I ) and MDA5 from 8–24 h ( Figure 7J ) . PolyIC treatment also induced RIG-I and MDA5 protein by 4–12 h post treatment as shown by western blotting ( Figure 7K ) and immunofluorescence ( Figure 8A ) . Finally siRNA specific for TLR3 or TRIF reduced polyIC induced RIG-I and MDA5 protein ( Figure 8B ) , strongly implicating TLR3/TRIF mediated signal transduction in RV induced RNA helicase induction in HBECs . As RV induced RIG-I and MDA5 occurred early , and in a TLR3/TRIF dependent manner , and as the TLR3 and TRIF pathway leads to IRF3 activation and IFN-β and IFN-λ induction , we assessed the role of IFN signalling in RV1B induced RIG-I and MDA5 . In our in vitro experiments with HBECs , it was difficult to rule out the role of endogenous , or RV induced early IFN-β/λ inducing RIG-I and MDA5 . We therefore ultilised IFNAR1 deficient mice , in a mouse model of RV1B infection [39] . IFNAR1 deficient mice are devoid of IFN-α/β signaling , and spleen cells have recently been shown to have reduced IFN-λ ( IL-28B ) production in vitro and lack IFN-λ within the vagina in vivo [40] . Figure 9 shows that following RV1B infection , IFNAR1 deficient mice produced a low level of IFN-β within the lung but did not produce IFN-λ , however wildtype controls produce both IFN-β mRNA at 24 h ( Figure 9A ) and IFN-λ mRNA at 48 h ( Figure 9B ) . At 8 h post infection , RV1B infection of IFNAR1 deficient mice and wildtype mice both resulted in increased RIG-I ( Figure 9C ) and MDA5 gene expression compared to time 0 h , ( Figure 9D ) . The induction of RIG-I and MDA5 was biphasic for both IFNAR1 deficient and wildtype mice , wildtype mice produced higher levels of RIG-I and MDA5 mRNA at 16 h post infection , and at 48 h post infection for RIG-I only . Cytoplasmic protein from lung homogenates were extracted and probed for RIG-I and MDA5 protein by western blotting . Both IFNAR1−/− and wildtype mice exhibited increased RIG-I and MDA5 protein at 8 h compared to mock infected controls . At later time points ( 24 h and 48 h ) wildtype mice had more RIG-I and MDA5 protein compared to IFNAR1 deficient mice , likely caused by IFN signaling ( blots are shown in Figure 9E and densitometry compared to α-tubulin shown in Figure 9F&G ) . This data therefore shows that initial RV induced RIG-I and MDA5 is IFN independent , and that at later time points , further virus replication increases in RIG-I and MDA5 gene expression later on in IFNAR1−/− mice , and in wildtypes further virus replication and IFN signaling events contribute to later lung RNA helicase mRNA expression .
Viral dsRNA and ssRNA are recognised by at least two independent pattern recognition pathways , composed of TLR3 , and TLR7/8 in the endosome and the RNA helicases RIG-I , MDA5 in the cytoplasm . Previous studies have employed well established models of virus infection , however similar studies using viruses important to human disease in their natural host cell type in vivo are largely yet to be performed . In the present study , we describe the role of TLR3 and RNA helicases in the recognition of RV infection in primary HBECs , the main cell type infected in the lower respiratory tract in vivo . RV is an important human pathogen , responsible for a range of illnesses including asthma exacerbations . We have used both major group RV16 and minor group RV1B ( approximately 77% genome identity ) in our studies , as both these groups represent the majority of RVs involved in human disease . Understanding the basis of RV recognition and signalling leading to IFN and pro-inflammatory cytokines could potentially lead to new therapeutic targets for RV associated illnesses . We found that TLR3 , RIG-I and MDA5 were required for IFN-β while MDA5 and TLR3 were required for maximal IFN-λ1 and IFN-λ2/3 mRNA expression . We then hypothesized that the most likely explanation for the dual requirement of both endosomal and cytoplasmic recognition systems and the importance of both RIG-I and MDA5 for IFN-β expression was that the endosomal and cytoplasmic recognition pathways were in some way linked , and their sequential activation was required for maximal IFN-β and IFN-λ gene expression . This is in contrast to other studies , which have observed mostly cell type specific differences concerning RIG-I/MDA5 and the TLR family members , with murine embryonic fibroblasts ( MEFs ) and conventional dendtitic cells ( DCs ) utilizing RIG-I and MDA5 for virus induced IFN production [25] , [36] while plasmacytoid DCs use TLRs including TLR3 [35] , [36] . Despite these findings , a recent study has reported the induction of type I IFN after i . p . injection of polyIC , to be reduced in TRIF−/− and IPS-1−/− double deficient mice , compared to either IPS−/− or TRIF−/− mice [41] . A number of recent in vitro studies have also reported a role for both TLR and RNA helicase signalling in polyIC induced responses , [42]–[45] . In short , explanations for differences between these results are likely due to cell type dependence on one pattern recognition system versus another , and the models of virus infection employed . We argue that for structural cells , including epithelial cells at mucosal surfaces , these cells may have evolved a dual dependency on both endosomal and intracellular recognition systems , which may be dissimilar for leukocytes including DCs . Hence as a first line defense against viral infection , efficient IFN production is an outcome of both TLR and RNA helicase mediated signalling working together . The siRNA experiments we performed have demonstrated the requirement of TLR3 and MDA5 for IFN-λs , but TLR3 , RIG-I and MDA5 for IFN-β gene expression . TLR3 , RIG-I and MDA5 were also all important for a range of pro-inflammatory cytokines , including T cell chemokines rantes and IP-10 and neutrophil chemokines IL-8 and ENA-78 . These results confirm and expand on recent data by Wang et al [46] , using BEAS-2B cells and specific siRNA transfection , Wang et al showed that MDA5 , TRIF but not RIG-I was important for RV induced IFN-β , IFN-λs and several interferon stimulated genes ( ISGs ) . The role of MDA5 was confirmed in primary tracheobronchial epithelial cells . However why or how MDA5 and TLR3 are both required for RV infection in these studies was not analyzed in any depth . Our observation of RIG-I being important for IFN-β in HBECs could be due differences between BEAS-2B cells and HBECs , or differences in efficiency of siRNA knockdown of target mRNA . Our data also show that the RIG-IC DN also increases RV replication in BEAS-2B cells , again suggests a role for RIG-I in RV responses . Why RIG-I was not required for IFN-λ expression in our studies is unclear . Possible explanations for these results could be that MDA5 siRNA was more consistent at reducing the target mRNA compared to RIG-I siRNA , or that at the time point studied ( 24 h ) , MDA5 is more important than RIG-I in IFN-λ expression . IFN-β and the IFN-λs are both IRF3 responsive [47] , and it is possible that for RV infection , MDA5 is more efficiently activated , resulting in robust signalling and IRF3 activation . Recently , it has been shown that IRF3 activation is complex , requiring multiple kinases for maximal phosphorylation and activation [48] . It is possible that differences exist between RIG-I and MDA5 signaling to IRF3 , or IFN-β and IFN-λ promoters have different requirements for IRF3 activation . It has also previously been observed that siRNA transfection can interfere with endogenous RNA sensing molecules [49] , [50] and induce spontaneous IFN or cytokine production . All our siRNA were used to minimize potential off-target effects; they were used as pools of four individual siRNAs , and were designed with minimal known stimulatory sequences and also contained 3′ UU overhangs to minimize activation of RIG-I , which can be activated by siRNA [49] . We were careful to assess the likelihood of spontaneous induction of IFN or cytokines studied by all siRNA , and we did not observe significant induction for control or any specific siRNA . Therefore , we are confident that our results using siRNA are accurately describing the role of each molecule in RV dependent responses , and are not confounded due to siRNA recognition by endogenous processes or are the result of obvious off-target effects . Initial experiments into the specificity of RIG-I suggested it bound in vitro transcribed and/or 5′-triphosphorylated ssRNA [33]–[35] , and therefore could not recognize the RNA of picornaviruses such as encephalomyocarditis virus ( EMCV ) which do not synthesize 5′-triphosphosphorylated ssRNA . This has been questioned recently with the observation that RIG-I can bind low weight dsRNA [32] . Other than EMCV , little is known about the role of RIG-I in the infection of Picornaviruses . Our data provide definitive evidence that both RIG-I and MDA5 are important in innate responses to RV infection . Knockdown of both RIG-I and MDA5 produced higher viral loads of major group RV16 and minor group RV1B , again providing evidence that RIG-I can recognize the RNA of Picornaviruses . As the 7kB RV genome replicates in a RNA dependent manner , we argue that dsRNA molecules are present during replication in the cytoplasm . Hence , it is plausible that dsRNA of differing sizes could potentially ligate both RNA helicases . Future studies to confirm these interactions , such as immunoprecipitation studies are necessary to investigate the exact nature of the RNA species that RIG-I and MDA5 are binding to in the context of RV infection in HBECs . Having established that TLR3 , RIG-I and MDA5 were all required for innate responses and that RIG-I and MDA5 protein was not constitutively expressed in HBECs , whereas TLR3 was constitutively expressed , we hypothesized that RIG-I and MDA5 could be induced by TLR3 activation . While we did not study the early events of viral entry , RV has been used as a model Picornavirus and their biology has been extensively studied . RV enters via the endosome where the acidified environment is essential to viral uncoating and release of +ve sense ssRNA . [51] , [52] . Despite being a ssRNA virus , the 7kB ssRNA genome contains some secondary structures [53] , [54] , including the 5′ multiple stem loop structure , containing the ribosome entry site , which has been previously visualized in endosomes during virus uncoating [55] . As RV infection requires the formation of mature endosomes , and contains dsRNA structures , thus our results could be explained by TLR3 sensing these events during viral entry; and initiate signaling leading to IFN , and RIG-I and MDA5 gene expression during the first few hours of infection . Further experiments are required to prove or disprove this idea however . HBECs were unresponsive to the TLR7/8 ligand R848 , strongly suggesting this cell type lacks these TLRs , and that TLR3 rather than TLR7/8 is involved in RIG-I and MDA5 induction . TLR3 has previously been implicated in RV recognition in BEAS-2B cells [56] and HBECs [9] . Interestingly , unlike BEAS-2B cells , we found that TLR3 was not virus inducible in undifferentiated HBECs . In both models however , TLR3 is constitutively expressed , enforcing the hypothesis that TLR3 is an initial sensor of viral nucleic acid in this cell type . A range of experiments in HBECs demonstrated that RIG-I/MDA5 induction was TLR3/TRIF dependent . Very recently , co-operation between RIG-I/MDA5 and TLR3 has been suggested in a murine model of coxsackievirus infection [57] . TLR3 was absolutely required for defense against coxsackievirus infection , by inducing IFN-γ . The authors suggest that TLR3 mediated induction by IFN-γ may work in parallel with RIG-I/MDA5 inducing type I IFN , and further suggested that these responses may be coupled , although potential mechanisms for this were not explored . We believe however that our study is the first to provide definitive evidence of endosomal and intracellular PRRs working in concert . We argue that as RV enters via endosomes , yet most of the dsRNA load occurs in the cytoplasm , the idea of two related pattern recognition systems seem plausible . RV replicates in the cytoplasm , and produces multiple centers of RNA dependent RNA replication , therefore increases in the RNA helicases can be viewed as a mechanism to continually monitor the intracellular RNA load ( likely dsRNA and ssRNA ) , and through RNA helicases , induce IFN and cytokines consistently , during the course of infection . The upregulation of IFN-β , IFN-λ and T cell and neutrophil cytokines are highly important for the control of virus replication and acute inflammatory responses within the airway ( for a summary , see Figure S4 ) . It would be extremely interesting to study other common human respiratory viruses which preferentially infect bronchial epithelial cells , or viruses that infect other mucosal surfaces such as the gut , to investigate if maximal IFN and cytokine responses require both TLR3 , and RIG-I/MDA5 . Finally , we assessed the role of IFN-α/β signaling in RIG-I and MDA5 induction . RIG-I and MDA5 are ISGs , a consistent observation in many different cell types and models [24] , [57]–[60] . We have also observed that in HBECs , RIG-I and MDA5 are IFN-β and IFN-λ inducible ( data not shown ) . As RIG-I and MDA5 are IFN inducible , our in vitro experiments could not rule out the possibility of the effects of low level IFN inducing RIG-I and MDA5 . Our data in vivo , however clearly show that RIG-I and MDA5 can be induced in the absence of IFN-β and IFN-λ , IFNAR1−/− mice , which cannot respond to type I IFN and are unable to produce IFN-λ in the lung , still produced RIG-I and MDA5 mRNA and protein upon RV1B infection . Also , in HBECs with intact IFN responses , RIG-I/MDA5 mRNA was upregulated early , by 4–8 h , and we argue this quick response is likely IFN independent . We believe this is the first report of sequential involvement or collaboration of TLR and RNA helicase mediated pathways for innate defense against a virus infection . Our overall model is depicted in Figure S4 . The model argues that RIG-I and MDA5 that are not well expressed in uninfected cells are both virus inducible via the constitutively expressed TLR3/TRIF , and later IFN inducible . Upon RV infection , TLR3 signaling in the endosome gives quick induction of new RIG-I and MDA5 . Importantly , the model highlights the need for RNA helicase induction to be quick , in a few hours in infected cells . After several hours , as virus moves out of the endosome and into the cytoplasm , newly synthesized virus RNA is sensed by RIG-I/MDA5 . This is where the majority of viral nucleic acid will be for the rest of the infection cycle , and is likely key in the induction of innate immune response to RV infection . At later time points , the actions of IFN-β and IFN-λs may further induce RIG-I and MDA5 in infected and non-infected cells . In non-infected cells , the presence of IFN may “warn” neighboring cells about the presence of viruses , and prepare the epithelium through upregulation of interferon inducible genes including RIG-I and MDA5 . Our initial interest in PRRs important in RV infection and IFN expression came from studies in asthmatic individuals which showed bronchial epithelial cells from asthmatics had very low levels of IFN-β and IFN-λ expression and increased RV replication compared to non-asthmatic controls [17] , [18] . It is currently believed that IFN-β , or IFN-λs could contribute to the outcome of asthma exacerbations . While these studies implicate the bronchial epithelium as a key producer of IFN-β and IFN-λs in RV infections , it is unclear which IFN is more important in protection . Understanding the regulation of both type I and type III IFNs is therefore a research goal for identifying why asthmatics have deficient innate responses to RV infection . The results of the present study have identified the importance of RIG-I , MDA5 and TLR3 in RV induced IFN , therefore future studies should scrutinize these pathways in asthmatics and non-asthmatics to ascertain if asthmatics have defective signaling leading to decreased IFN production . In summary we provide evidence that the dsRNA receptor TLR3 acts to induce both RIG-I and MDA5 gene and protein expression in HBECs in a model of RV infection . Both RIG-I and MDA5 were required for maximal IFN and pro-inflammatory cytokine induction , and control of RV replication indicating that they have non-redundant roles in RV infection . The data support a model that in HBECs , TLR3 but not TLR7/8 operate in concert with RIG-I/MDA5 , and are together required for innate responses to RV infection . As asthmatic HBECs have reduced RV inducible IFN-β and IFN-λ expression compared to non-asthmatic cells , both the TLR3 and RNA helicase pathways warrant further exploration in order to ascertain why these cells produce reduced IFN expression .
HBECs from non-asthmatic , non-smoking individuals were obtained from a commercial source ( Clonetics , Wokingham , UK ) , and cultured in BEGM with supplements according to the suppliers recommended protocol ( Clonetics ) . Unless otherwise stated , all data was derived from experiments from 3 different HBEC sources . BEAS-2B cells ( European Collection of Cell Cultures ) were cultured in RPMI with 10% FCS ( Invitrogen , Paisley , UK ) . HeLa cells were grown in DMEM with 10% FCS ( Invitrogen ) , and used for RV titration assays . Major group RV16 and minor group RV1B were grown in HeLa cells , after three cycles of freeze and thawing , supernatant and cellular material where clarified by centrifugation at 4 , 000 rpm for 15 min , filter sterilized , aliquoted and frozen at -80°C . The serotype of all RV stocks was confirmed by titration with serotype specific anti-sera ( American Type Culture Collection ) , and all RV stocks and cells were confirmed to be free of Mycoplasma contamination using a commercially available detection kit ( Roche , Burgess Hill , UK ) . HBECs were cultured to 80% confluency in 12 well plates and transfected with 100 nM specific siRNA or control siRNA ( specific for luciferase , Dharmacon , Lafayette , CO , USA ) , for 24 h prior to infection with RV1B . Time courses and dose responses of siRNA were performed previously to determine optimum conditions for knockdown of target genes . All siRNA achieved at least a 75% knockdown of target mRNA , and each siRNA was assessed for the induction of IFN or pro-inflammatory cytokine mRNA in the absence of infection . HBECs were cultured to near confluency in 12 well plates and then transfected with 0 . 25 µg per well of ΔTRIF-pUNO1 ( a constitutive active TRIF cDNA , Invivogen , San Diego CA , USA ) , or pUNO1 control plasmid ( Invivogen ) or left untransfected for 5 h . All transfections were with Lipofectamine 2000 ( Qiagen , Crawley , UK ) according to the manufacturers recommended protocol . Complexes were removed , medium replaced and cells left for 24 h . RNA was then extracted and RIG-I , MDA5 mRNA measured . ΔRIG-I , RIG-IC and pEF-BOS control vector [24] were used to transfect BEAS-2B cells , at 0 . 25–0 . 5 µg per well , using Superfect ( Qiagen ) according to the manufacturer's recommended protocol . HBECs were cultured to 80% confluency in 12 well plates , using BEGM ( Clonetics ) and starved in BEBM ( no supplements ) overnight and infected with RV1B for 1 h with shaking at room temperature , and samples taken at appropriate time points . Alternatively , HBECs were transfected with siRNA ( Dharmacon ) , placed in BEGM without serum and then infected with RV1B . BEAS-2B cells were placed in 2% FCS containing RPMI medium overnight and infected with RV1B or RV16 for 1 h as above and placed in 2% FCS containing RPMI medium until required . For experiments involving enumerating virus replication , adhered virus was washed off by three additions of medium after the 1 h infection period . Total RNA was extracted from HBECs ( RNeasy kit , Qiagen ) , and 2 µg was used for cDNA synthesis ( Omniscript RT kit , Qiagen ) . Total RNA was also extracted from the upper left lobe of the mouse lung , and placed in RNA later ( Qiagen ) , prior to RNA extraction and cDNA synthesis ( as above ) . Quantitative PCR was carried out using specific primers and probes for each gene ( Table S4 in Supporting Information S1 ) . Reactions consisted of 12 . 5 µl of 2X Quanti-Tect Probe PCR Master Mix ( Qiagen ) in 25 µL . cDNA for 18S amplifications were diluted 1/100 in sterile water . Reactions were analyzed using an ABI 7000 TaqMan , ( ABI Foster City , CA , USA ) at 50°C for 2 min , 94°C for 10 min , and 45 cycles of 94°C for 15 s and 60°C for 15 s . Each gene was normalized to 18S rRNA , and for HBEC studies , presented as copies of each mRNA per 2×105 cells , and for mouse lung , per cDNA reaction using a standard curve based on amplification with plasmid DNA . For siRNA experiments , copy number was expressed as a % of copy number versus control siRNA . For western blotting , total protein lysates were run on 4–12% Bis-Tris polyacrylamide gels , and transferred onto nitrocellulose membranes ( Invitrogen ) , blocked in 5% skim milk , and probed with antibodies specific for mouse and human RIG-I ( Cell Signaling , Danvers , MA , USA ) , diluted to 0 . 083 µg/mL , MDA5 1 µg/mL ( Santa Cruz Biotechnology Inc , CA , USA ) , α-tubulin 0 . 2 µg/mL ( Santa Cruz Biotechnology Inc ) , or β-actin , 1 µg/mL ( Biovision , Mountain View CA , USA ) . Secondary antibodies used were goat anti-mouse HRP , 0 . 08 µg/mL , sheep anti-rabbit HRP , 2 µg/mL ( AbD Serotec , Oxford , UK ) and swine anti-goat HRP 1 . 4 µg/mL ( Invitrogen ) . Blots were developed using ECL ( GE Healthcare , Chalfont St Giles , UK ) . For immunohistochemistry , HBECs were grown on 8 well chamber slides ( Nunc , Rochester NY , USA ) , infected with RV or treated with 10 ng/mL IFN-β , 5 µg/mL polyIC or medium for 8 or 24 h , and washed with PBS , fixed in 4% paraformaldehyde at room temperature for 5–7 min , washed once with PBS , and permeabilized with 0 . 2% Triton X-100 for 5 min at room temperature , and washed again in PBS . Slides were then blocked with a 1% BSA , 10% FCS-PBS overnight at 4°C . Cells were then stained with either anti-RIG-I , 2 µg/mL ( Santa Cruz Biotechnology , Inc ) or anti-MDA5 , 2 . 7 µg/mL ( Santa Cruz Biotechnology Inc ) for 1 h at room temperature . Slides were washed three times with PBS , and stained with for 1 h at room temperature with donkey anti-goat Alexa Fluor 488 , 6 . 7 µg/mL , and washed as above . Slides were then mounted with 4 , 6-diamidino-2-phenyindole dilactate containing mounting medium , and analysed using a colour CCD camera microscope ( Zeiss , Rugby , UK ) . Paraffin embedded bronchial biopsies were obtained from 15 non-asthmatic non-smoking individuals used in a previous in vivo challenge study with RV16 [16] . Samples were coded , and analyzed blind according to infection status , for RIG-I and MDA5 prior to infection ( baseline ) or at day 4 after infection . Goat antibodies to RIG-1 and MDA5 ( Santa Cruz Biotechnology , Inc ) at 2 µg/mL and 1 µg/mL were used respectively . Swine anti-goat LSAM-HRP reagents ( DakoCytomation , Ely , UK ) , were used as per the manufacturer's recommended protocol , and antibody binding visualized using perioxidase staining . Staining intensity on surface epithelium was scored accordingly as 0–3 , with no staining scored as 0 and intense staining scored as 3 . Female IFNAR1−/− and 129/SvJ control mice aged 6–9 weeks were inoculated intranasally with 5×106 TCID50 of RV1B , essentially as previously described [39] , and culled humanely by lethal injection at various time points . The human experimental challenge study was approved by St Mary's NHS Trust . All volunteers gave informed , written consent . All animal work was in accordance with Project License PPL70/6387 , and performed according to regulations outlined by the Home Office , UK , in agreement with the Animals ( Scientific Procedures ) Act 1986 . All in vitro experiments were performed 5–6 times , Figures 1 , 3–7 used 3 independent HBEC donors , and data generated in the supporting information file also utilized 3 independent HBEC donors . For siRNA experiments , data from each independent experiment was converted to a % of the control siRNA + RV data , and mean ± SEM generated . All other data were expressed as mean ± SEM . Experiments using siRNA or transfection with plasmids were analyzed by one ANOVA and Bonferroni's multiple comparison test , and time course data using two-way ANOVA and Bonferroni's multiple comparison test , using GraphPad Prism software with p<0 . 05 taken as significant . For differences between two groups , a student's t-test was employed with p<0 . 05 taken as significant . Experiments in the mouse model involved 3–4 animals per group , in two independent experiments , ( total of 6–8 animals ) data were analyzed using two-way ANOVA and Bonferroni's multiple comparison test in GraphPad Prism . Staining of human bronchial epithelium for RIG-I and MDA5 was analyzed by using the paired Mann-Whitney U test , p<0 . 05 taken as significant . | Host-pathogen interactions are mediated by pattern recognition receptors that identify conserved structures of micro-organisms that are distinct from self . During a viral infection , important pattern recognition receptors include the endosomal Toll-like receptors ( TLRs ) , and a second set of cytoplasmic pattern recognition receptors known as the RNA helicases . Many studies have highlighted the importance of TLR3 , TLR7/8 and the RNA helicases in providing robust anti-viral immunity via interferon induction and inflammation . Both endosomal TLR and cytoplasmic RNA helicase mediated pathways are believed to exist as separate yet non-redundant entities; however , little thought is given to why both systems exist , and few studies also consider how both pathways together contribute to anti-viral immunity . Using models of rhinovirus infection in primary bronchial epithelial cell culture in vitro and experimental infection in mouse and human models in vivo , we show that the RNA helicases are preferentially induced early in the infection cycle via TLR3 mediated signaling events , and work in a co-ordinated , systematic manner . The results help understand the complex events that determine effective innate immunity to rhinovirus infection and how these processes contribute to virus induced exacerbations of asthma and chronic obstructive pulmonary disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunology/cellular",
"microbiology",
"and",
"pathogenesis",
"immunology/innate",
"immunity",
"infectious",
"diseases/viral",
"infections",
"infectious",
"diseases/respiratory",
"infections",
"immunology/immunity",
"to",
"infections"
] | 2010 | Co-ordinated Role of TLR3, RIG-I and MDA5 in the Innate Response to Rhinovirus in Bronchial Epithelium |
Herpes simplex encephalitis ( HSE ) is a fatal infection of the central nervous system ( CNS ) predominantly caused by Herpes simplex virus type 1 . Factors regulating the susceptibility to HSE are still largely unknown . To identify host gene ( s ) regulating HSE susceptibility we performed a genome-wide linkage scan in an intercross between the susceptible DA and the resistant PVG rat . We found one major quantitative trait locus ( QTL ) , Hse1 , on rat chromosome 4 ( confidence interval 24 . 3–31 Mb; LOD score 29 . 5 ) governing disease susceptibility . Fine mapping of Hse1 using recombinants , haplotype mapping and sequencing , as well as expression analysis of all genes in the interval identified the calcitonin receptor gene ( Calcr ) as the main candidate , which also is supported by functional studies . Thus , using unbiased genetic approach variability in Calcr was identified as potentially critical for infection and viral spread to the CNS and subsequent HSE development .
Herpes simplex type 1 virus ( HSV-1 ) is a member of the Herpesviridae family ( Alphaherpesvirinae subfamily ) that infects a large fraction of humans resulting in transient cold sores or non-symptomatic infection that persists lifelong in the sensory ganglia . Recurrent herpetic disease results from reactivation of HSV-1 in the sensory ganglia subsequently leading to axonal transport of the virus to the periphery where it causes skin lesions , cold sores , often located around the mouth . However , HSV-1 can also cause a much more severe condition , Herpes simplex encephalitis ( HSE ) , an acute inflammatory condition of the brain . Even though Herpes simplex is a neurotropic virus , HSE occurs in only 2–3 previously healthy individuals/million/year in all age groups [1] . In more than ninety percent of the cases , HSE is caused by HSV type 1 and in the remaining by HSV type 2 [2] . HSE is characterized by acute onset of focal infection , inflammation and necrosis , mostly starting unilaterally in the fronto-medio-basal temporal lobe . The disease has a tendency to relapse or to have a progressive course [3] . The mortality is high and there is significant morbidity among the survivors . Host factors contributing to susceptibility or resistance to HSE are still largely unknown . Genetic analysis is one approach to identify these factors . Polymorphisms in the UNC-93B and TLR3 genes were shown to regulate susceptibility to HSE in small human pedigrees , in which the production of IFN-α/β and -λ dependent on UNC-93B protein expression controls HSV-1 by TLR3-dependent and/or TLR-independent pathways [4] [5] . In addition , recently , autosomal dominant and recessive deficiencies in TRIF , an adaptor molecule involved in downstream signaling of TLRs , have been reported in a few children with HSE [6] . However , in an experimental mouse model for HSE , a natural killer ( NK ) complex-linked locus , Rhs1 ( resistance to Herpes simplex virus 1 ) , on chromosome 6 has been identified to control resistance to acute and latent HSV-1 infections resulting in HSE [7] . In 2003 Lundberg and colleagues identified in another mouse model of corneal HSV-1 infection an additional locus on chromosome 6 , Hrl ( Herpes resistance locus ) influencing survival after HSV-1 infection in C57BL/6J mice and the HSE development in 129S6SvEv/Tac mice [8] . Up to date , no HSE susceptibility genes have been identified by positional cloning in mice . Several mouse knock-out studies have shown the complex immune control of HSE , with excessive infiltration of leukocytes leading to the release of cytokines into the CNS suggested to be a major determinant of brain damage after infection , in turn regulating outcome [9] . The aim of the present study was to identify additional host factors determining HSE receptiveness by genetic dissection of the previously characterized discordant HSE susceptibility pattern in the inbred Dark Agouti ( DA ) and Piebald Virol Glaxo ( PVG ) rat strains [10] . This in vivo model for HSE in DA rats resembles in some aspects the viral spread seen in human HSE , where the virus starts spreading from the whiskers area of the rats ( the labio-facial area in humans ) , through the trigeminal nerve to the ipsilateral side of the brain stem dispersing both to the contralateral side and towards the thalamus , causing immune activation and lethal encephalitis 5 days post-infection ( dpi ) [10] . Interestingly , in our previous study , PVG rats were found to be completely resistant to development of clinical symptoms and did not show evidence of penetration of virus into the brain . Although both strains showed similar virus presence in the whiskers area [10] , as well as in the immediate proximity of nerve endings and small nerve fascicles [11] , spread of HSV-1 to the brain was only seen in DA rats . The major histocompatibility complex ( MHC ) of DA or PVG did not regulate the resistance to HSE , since also MHC congenic PVG . A ( RT1 . AV1 ) rats carrying the same MHC haplotype as the DA were protected [11] . Similar results have been shown previously in studies with inbred and congenic mice where genes within the H-2 ( major histocompatibility complex ) did not influence resistance or susceptibility to HSV-1 infection [12] [13] . To identify gene loci critical for the strain dependent difference in HSE susceptibility , we performed a genome-wide linkage study in a large F2 ( DAxPVG . A ) cohort in which all HSV-1 infected F2 rats were phenotyped for disease symptoms . The study identified Hse1 on chromosome 4 as the single strong quantitative trait locus ( QTL ) . To find the critical gene variant within Hse1 we performed further analyses using recombinants , haplotype mapping and gene sequencing , as well as expression analysis , visualization of the tissue localization and receptor modulation experiments .
HSV-1 was inoculated into the right whiskers' pad of 45 days old DA rats . However , while DA rats developed a severe , lethal HSE with nearly 100% incidence at 5 days post infection ( dpi ) , both PVG and the PVG . A rats remained completely asymptomatic even after several weeks post infection [10] . To identify gene regions causing differences in HSE susceptibility , we crossed DA with PVG . A rats for two generations to produce an F2 population . All rats of the F1 generation were resistant to the disease . In a cohort of 239 F2 ( DAxPVG . A ) rats infected with HSV-1 the total incidence of HSE was 15% . The remaining rats did not show signs/symptoms of disease or were only affected by minimal weight loss around 5 dpi . Stratification for gender showed 20% HSE incidence in F2 males , with mean onset of disease at 6 dpi and 10% HSE incidence in F2 females , with a slightly delayed onset of disease at 8 dpi . The studied phenotypes are summarized in Table 1 . A genome-wide linkage scan was performed using 127 microsatellite markers on 180 F2 ( DAxPVG . A ) rats to determine the genomic region ( s ) influencing susceptibility to HSE . We found a very strong linkage to a region on chromosome 4 , designated Hse1 which regulated the incidence of encephalitis with a logarithm of odds ( LOD ) score of 29 . 5 at the D4Kini3 marker located at 27 . 8 Mb ( Figure 1 , Table 2 ) . The confidence interval ( CI ) of Hse1 is between D4Kini1 and D4Arb25 ( 24 . 3–31 . 1 Mb ) ( Table 2 ) . The linkage strength was reduced when analyzing females only , with a LOD score of 12 . 1 . In males the LOD score was 26 . 4 ( Figure 1C ) . In the entire cohort , 46 rats out of 55 being homozygous for DA alleles at the peak marker developed HSE ( 84% incidence ) . In contrast , only 4 rats out of 91 homozygous for PVG alleles in Hse1 developed HSE ( 4% incidence ) , while none of the 34 heterozygous rats developed HSE ( Figure 1D ) . Another QTL on chromosome 3 ( Hse2 ) was found to be significant for HSE incidence in females ( Figure 2A , Table 2 ) . In addition to HSE incidence , the onset of disease and the body weight change were recorded as phenotypes . Body weight change was determined as the difference between the weight at the day of infection and weight at 5 dpi ( d0–d5 ) ( Table 1 ) . The locus Hse1 was the main regulator of the onset of disease , which also correlated to body weight change between d0–d5 ( Figure 2B and 2C ) . Hse4 and Hse5 located on chromosomes 6 and 10 , showed suggestive linkage to disease onset and weight change between d0–d5 in females , respectively ( Figure 2B and 2C ) . Notably , for all these QTLs linkage was stronger in males and the effect plots showed disease susceptibility associated with DA alleles . These QTLs are summarized in Table 2 . A major host genetic contribution to HSE susceptibility was provided by the Hse1 locus , contributing with 51% in females and 50% in males to the variance of incidence and further contributing with 54% in females and 64% in males to the variance of onset . Additional interactive QTLs specific for each sex were identified in a number of chromosomes collectively contributing to more than 90% of the variance in most phenotypes in females and less in males , indicating a more complex regulation of HSE in females ( Table S1 ) . To obtain direct experimental proof that gene ( s ) in the region indeed regulate HSE susceptibility , and for further genetic dissection we bred a set of congenic lines . These lines included DA . PVGc4-Hse1 and DA . PVGc4-Hse1-R1 , which carries the PVG fragment between the microsatellite markers D4Kini3 – D4Rat177 and D4Kini3 – D4Mgh14 , respectively , transferred onto DA background . In order to test if PVG alleles in these fragments confer resistance to HSE , 5 homozygous rats from each congenic strain were infected . None of the rats developed symptoms , demonstrating the protective role of the PVG alleles in the Hse1 region ( Figure 3 ) . For both mapping and control purposes , we furthermore tested HSE susceptibility in a set of congenic lines with different chromosome 4 congenic PVG fragments on the DA background; R2:DA . PVG ( D4Rat23–D4Rat108 ) , R11:DA . PVG ( D4Rat103 – D4Mit12 ) , R21:DA . PVG ( OT40 . 07 – D4Mit12 ) covering the experimental autoimmune encephalomyelitis ( EAE ) QTLs Eae24–Eae27 [14] and R17 ( D4Kiru12–D4Kiru55 ) covering the APLEC genes region that are associated with arthritis and autoimmunity in rats and humans [15] ( Figure 3 ) . All of these lines are homozygous for DA alleles in Hse1 . All four lines developed disease with a clinical phenotype inseparable from that of DA rats , providing further support for the disease regulatory effect to be located within Hse1 . Notably , the non-overlapping fragment between on the one hand DA . PVGc4-Hse1 and on the other hand the R2 , R11 and R21 congenic lines delineates the disease regulatory effect of Hse1 to a region between D4Kini3 – D4Mgh14 ( Figure 3 ) . To fine map the regulatory effect within the large CI of Hse1 obtained from the initial F2 linkage analysis ( 24 . 3–31 . 1 Mb∼6 . 8 Mb region ) , the entire initial F2 population together with additional F2 rats recombinant in the CI from a parallel intercross were genotyped with more microsatellite markers within the CI ( D4Kini2 , D4Kini4–D4Kini16 ) ( Figure 4A , Table S2 ) . By defining the recombination positions within the CI region in recombining F2 rats , we were able to narrow the CI into a smaller region where resistant and susceptible rats have different genotypes . The fine mapping in F2 rats narrowed the CI of Hse1 to 1 . 01 Mb ( CI D4Kini3–D4Kini8; 27 . 81–28 . 82 Mb ) ( Figure 4A ) . To further define the CI and the exact region governing the disease susceptibility , we determined the development of HSE symptoms in a panel of other inbred rat strains and correlated their susceptibility to the allelic variation in this region ( Figure 4B ) . We found that the Lewis ( LEW ) , Fisher 344 ( F344 ) and Spontaneously Hypertensive Rat ( SHR ) were susceptible to HSE and developed a similar disease phenotype as DA , while the Bio Breeding type 1 diabetic rat ( BB ) and Brown Norway ( BN ) in similar to PVG were resistant . Other inbred strains developed a different disease pattern from DA , where some rats developed HSE symptoms while others displayed only mild or no symptoms of disease . These strains were considered to develop an intermediate phenotype , and included the August Copenhagen Irish ( ACI ) , Wistar Furth ( WF ) and Fawn-Hooded ( E3 ) strains ( Figure 4B ) . Based on the haplotype map in this larger set of inbred strains , the disease regulatory effect was suggested to be located within a smaller , 0 . 17 Mb region ( CI D4Kini5–D4Kini7; 28 . 37–28 . 54 Mb ) ( Figure 4B , Table S2 ) . However , the haplotype map was based on the susceptibility to HSE and the development of clinical symptoms in each strain after infection and not the viral presence in the CNS . The haplotype map also suggests a more complex genetic regulation of susceptibility vs . resistance to HSE , since the resistant BN strain carries the same genotype in this region as the susceptible strains ( Figure 4B ) . This supports the notion that different genes might be involved in regulating the clinical phenotype of HSE as well as pattern of virus spread across different strains . In the Ensembl release ( version 66; http://www . ensembl . org/ ) the genomic region defined by F2 recombination ( CI D4Kini3–D4Kini8; 27 . 81–28 . 82 Mb ) contains the genes cyclin-dependent kinase 6 ( F1MA87_RAT ) , hepacam family member 2 ( Hepacam2 ) , calcitonin receptor ( Calcr ) and tissue factor protein inhibitor 2 ( Tfpi2 ) , as well as 11 additional poorly annotated genes . The region also includes 3 microRNAs; novel miRNA ( ENSRNOG00000043729 ) and known miRNAs ( rno-miR-489 and rno-miR653 ) which are located within the Calcr gene . However , the region narrowed by the haplotype map ( CI D4Kini5–D4Kini7; 28 . 37–28 . 54 Mb ) includes only the three genes Ccdc132 , Calcr and Tfpi2 ( Figure 4B ) . All genes within the recombinant CI region were sequenced from genomic DNA of DA and PVG . A rats . Comparison of the DA and PVG sequences to the BN reference sequence revealed about 1470 SNP variations in the PVG sequence in Hse1 . The majority of the sequence polymorphisms were detected within the Ccdc132 , Calcr and Tfpi2 genes . These variations were mostly insertions/deletions in the intronic regions , SNPs in 5′ untranslated regions and intergenic regions . Only 4 synonymous SNPs in exons were detected ( Figure 5 ) . In addition , no SNPs were found in pre-miRNA sequences of miR-489 and miR-653 when comparing the DA and PVG with the reference genome ( data not shown ) . In order to study if these polymorphisms might affect translational stability , splicing or transcriptional control , we determined the mRNA expression pattern for Ccdc132 , Calcr and Tfpi2 in the whiskers area , the trigeminal ganglia and the brain stem using qRT-PCR ( Figure 6A–D ) . The most conspicuous finding was that PVG . A rats displayed higher Calcr expression in the whiskers area both in all controls as well as after HSV-1 infection compared to DA ( Figure 6B and 6D ) . Little or no expression of Calcr was detected in the trigeminal ganglia in both strains , indicating that the influence of Calcr is located in the periphery . Calcr has three splice variants; the primer used in Figure 6 covered exons common for all the variants . No differences in expression between the strains were detected when measuring the splice form of Calcr . 1b , however it was expressed more in the brain , with no or low expression in the trigeminal ganglia and the whiskers area , respectively ( data not shown ) . The expression of Tfpi2 was significantly higher only after infection in the trigeminal ganglia ( 2 dpi ) and the brain stem ( 4 dpi ) in DA rats compared to PVG rats . This could partly be explained by the presence of the virus in these compartments , since no difference was detected in the periphery . Primers used for qRT-PCR are listed in Table S3 . As for the two miRNAs encoded in one of the introns of Calcr , mir-489 and mir-653 , the expression of both the mature and the mature* miRNA was measured in the whiskers area of naïve and 5 dpi rats using qRT-PCR . The expression of both the mature mir-489 and mir-653 was significantly higher in the PVG rats at 5 dpi compared to the DA rats ( Figure 6E and 6F ) . As expected , the expression of both mature* miRNAs was very low , showing a similar expression pattern to the mature miRNAs , with significantly higher expression of mir-653* in PVG rats at 5 dpi as the only detected strain difference ( data not shown ) . Notably , the expression of the two miRNAs followed the mRNA expression pattern of Calcr in the whiskers area , suggesting that they are not expressed independently but rather as a result of Calcr expression . To visualize differences in viral spread and tissue immune response compared to the tissue-location of CalcR , the whiskers area was dissected and subsequently processed for staining using immunohistochemistry . Tissues from DA and PVG . A strains were dissected from naïve , Hank's solution-injected controls and infected rats after 5 dpi . Results are summarized in Table 3 . In a previous study [11] , we have shown that HSV-1 staining was similarly distributed in the whiskers area after infection in PVG and DA rats , but subsequently it became strongly increased , mostly in the perineurium , in DA rats ( Figure 7A ) , i . e . the layer of connective tissue surrounding the nerve fascicles/bundles of peripheral nerves . In contrast , in PVG . A rats replication of virus decreased in the whiskers area after 2 dpi ( Figure 7B ) . HSV-1 staining was only found in DA rats in the trigeminal ganglia and the brain stem , while in PVG . A rats these compartments were completely free of HSV-1 . In sections from the whiskers area at 5 dpi ( Figure 7 ) , virus labeling was present in the epineurium , i . e . the outermost layer of connective tissue surrounding several nerve fascicules/bundles , in both DA ( Figure 7D and 7E ) and PVG . A rats ( Figure 7G and 7I ) . However , DA rats also displayed positive staining within the peri- and endoneurium , i . e . in the layer of connective tissue surrounding each nerve fascicle and nerve fiber inside the fascicles , respectively . CalcR staining was stronger in the whiskers area of naïve and infected PVG . A rats compared to DA rats corroborating the qRT-PCR findings ( Table 3 ) . Interestingly , the tissue distribution of CalcR staining differed between the infected DA and PVG . A rats . In DA rats ( Figure 7C , 7E , 7I and 7K ) , CalcR staining was present mainly in perineurial cells and in the endoneurium , while in PVG . A ( Figure 7F , 7H , 7L and 7N ) rats it was expressed more in the outer part of the epineurium . Thus , the distribution pattern of CalcR staining resembles that of HSV-1 in both DA and PVG . A rats at 5 dpi . The macrophage marker Iba1 ( Figure 7J , 7K , 7M and 7N ) also followed the distribution pattern of HSV-1 and CalcR in each strain . Larger numbers of natural killer ( NK ) cells and CD8+ T cells were present in the whiskers area of DA compared to PVG . A rats . Infiltrating NK cells outnumbered the CD8+ T cells in the DA rats at 5 dpi ( Table 3 ) . To investigate the influence of Calcr on HSV-1 entry and spread to the CNS in vivo in DA rats , we modulated the calcitonin receptor by injecting rat Amylin , an agonist for calcitonin receptor , or Calcitonin ( 8–32 ) ( Salmon I ) , a potent and selective antagonist [16] , into the whiskers area prior to the infection . A scrambled peptide ( similar to the amino acid composition of rat Amylin , but in reverse order ) was used as control to the rat Amylin . Treatment was performed in 44 days old DA rats , 18 hours before the infection with HSV-1 . The rats were examined and weighed every day from the day of the injection with the modifying substances until 11 dpi , to check for signs and symptoms of disease development . Treatment with rat Amylin significantly improved the survival of the DA rats from 0 to 75% after infection compared to the control group , which was only HSV-1 infected . The survival rate in rats treated with Calcitonin ( 8–32 ) ( Salmon I ) group was 62% . In contrast , only 20% of rats injected with the scrambled control peptide survived until day 11 after infection ( Figure 8 ) and all the control rats died . Collectively , these findings suggest that in vivo modulation of the Calcr in the whiskers area affects viral entry to the CNS and progression to HSE in DA rats . To study susceptibility mechanisms at the molecular level , we performed infections of primary neuronal cell cultures from DA and PVG rats' dorsal root ganglia . No differences in susceptibility to infection were detected in these cells in vitro ( data not shown ) , indicating that in vivo , other features of the natural environment of the neurons contribute to the susceptibility or tolerance to infection . In addition to assess whether CalcR expression could modulate infectivity in vitro , HEK293T cells , which are semi-permissive to HSV-1 infection , were transfected with a plasmid encoding for a myc-tagged version of RAMP1 , a receptor activity-modifying intracellular protein which is transported to the cell surface by CalcR and is necessary for the responsiveness of CalcR to amylin , alone or in combination with plasmids encoding for the human or the rat CalcR ( Figure 9 ) . Staining for myc-tag and CalcR showed that in co-transfection experiments only RAMP1 expressing cells also co-expressed the CalcR ( Figure 9A ) , even though detection of the human CalcR was weaker , possibly due to lower reactivity of the antibody with the receptor form found in humans ( Figure 9B ) . Transfected cells were then infected with HSV-1 for 24 hours and HSV-1 infection and replication were assessed by intracellular staining and flow cytometry . RAMP1 or RAMP1/Human CalcR transfected cells were analyzed ( Figure 9A–9C ) for the percentage of HSV-1infected cells , which indicates how permissive the cells were to the virus , as well as for the mean fluorescence intensity ( MFI ) of HSV-1 staining in the infected population , which reflects the level of virus replication in these cells ( Figure 9D ) . As shown in ( Figure 9E–9H ) , no change in the proportion of HSV-1 infected cells could be seen between RAMP1 only or RAMP1/human CalcR transfected cells , indicating that in these semi-permissive cells the expression of CalcR did not alter virus uptake and internalization . However , the addition of rat Amylin ( calcitonin agonist ) reduced both the number of cells infected by HSV-1 ( Figure 9E ) as well as the MFI of staining for structural viral components in the infected cells population ( Figure 9F ) . The experiment was repeated with transfection of the rat CalcR yielding similar results ( data not shown ) . To assess whether the kinetics of CalcR triggering could affect virus replication , transfected cell cultures were incubated with rat Amylin at different time points ( before , during and after infection , or only after infection ) and the same analysis was performed ( Figure 9G and 9H ) . As expected from the previous experiment , rat Amylin significantly decreased the infectivity as well as the intensity of staining for virus structural components in cells that were co-transfected with RAMP1 and CalcR , but not in those transfected with the receptor modifier RAMP1 alone . In addition , the effect was slightly more pronounced in the cultures that were pretreated with rat Amylin prior to infection . Experiments involving the CalcR antagonist Calcitonin ( 8–32 ) ( Salmon I ) were performed in transfected HEK293T cells as described for rat Amylin but showed no modulator effect in vitro suggesting that the antagonist has an alternative mode of action in vivo ( data not shown ) . Finally , to assess whether the CalcR could serve as a receptor or co-receptor for virus entry in a non-permissive in vitro system , we utilized a rat adeno-carcinoma cell line ( CRL-1666 ) that cannot be readily infected with HSV1 and transfected and infected it in the same way as described above for HEK293T cells . Regardless of CalcR expression , these cells remained non-permissive , ruling out a role of CalcR as a receptor for virus infection per se ( data not shown ) .
We here demonstrate a potent genetically regulated host factor critical for Herpes simplex virus type-1 neuro-invasion mapping to a small genome fragment on rat chromosome 4 . Though very strong circumstantial evidence suggest genetic variants of the Calcr gene to be responsible , ultimate proof will require further experiments . HSV-1 infects the majority of the population inducing cold sores in affected individuals . Human necropsy studies suggest that viral DNA can be isolated from nearly all post-mortem brains , implying that with time most individuals are infected . However , in younger people ( between 20 to 49 years of age ) HSV-1 serology suggests that only 50–60% have been infected , as reported by independent studies in different populations [17] , [18] , [19] . These studies also demonstrate an increasing prevalence of HSV-1 infection with age . Thus , different susceptibility patterns to establish HSV-1 infection are possible in the human population . Therefore , our findings in the rat are not necessarily in contradiction to the situation in humans . Alternatively , the differences in susceptibility through peripheral nervous system ( PNS ) uptake and transport of the virus , as here documented in the rat , species are not relevant for human herpes infection . A third possibility is that the molecule is not subject to genetic variation in humans , but still being important mechanistically for viral-host interactions . On the other hand , in two to three individuals per million per year , the virus infection leads to a much more devastating condition with invasion and replication of virus in the CNS , causing focal necrotizing plaques affecting primarily the temporal and inferior frontal lobes of the brain . Although rare , HSE remains the most common cause of acute , sporadic viral encephalitis in the Western world [20] . The underlying host determinants regulating HSE susceptibility are largely unknown . The entry of HSV-1 into the host cells is known to depend on the interaction of several glycoproteins on the surface of the enveloped virus with receptors on the surface of the host cell . These entry receptors include Herpes virus entry mediator ( HVEM ) , nectin-1 and 3-O sulfated heparan sulfate [21] . Through these known entry receptors the virus remains latent in sensory neurons . However , still little is known about the host factors that influence reactivation and the different entry ports of HSV-1 into the CNS . The in vivo model used in this study in 45 days old rats simulates the human infection in a number of aspects . It starts from the whiskers area corresponding to the labio-facial area in humans . Interestingly , DA rats at an age of over 60 days at infection were resistant to HSE disease phenotype development , while PVG rats below an age of 30 days were susceptible ( unpublished observation ) . Age dependent effects on susceptibility to HSE have been described previously and it has been recognized in many viral infections of man and experimental animal species . Metabolic and hormonal changes , antibody responses , inhibitory substances , anatomical characteristics , and interferon production have all been suggested to explain this development of resistance [22] , [23] . The virus in the DA susceptibility model penetrated the trigeminal nerve and the ipsilateral side of the brain stem after infection , subsequently spreading in contralateral and cranial direction within the CNS . While high virus titers were observed in the DA rats , no traces of HSV-1 could be detected in the resistant PVG , nor live virus was retrieved from the trigeminal ganglia or the brain stem using qRT-PCR [10] or immunohistochemistry [11] . These findings support the notion of an underlying genetic difference affecting the ability of HSV-1 to enter the nervous system of PVG rats . Given the dichotomous difference in HSE susceptibility between the DA and PVG strains , HSV-1 infection of an F2 ( DAxPVG . A ) intercross was performed in order to identify the underlying genetic determinants in an unbiased fashion . The main finding of the linkage analysis was the identification of a new QTL on rat chromosome 4 , Hse1 . This QTL was the main regulator of disease both in males and females . In female rats the regulation of HSE seems to be more complicated as linkage analysis identified three additional smaller QTLs regulating different disease phenotypes suggesting the influence of other genes in disease development . The observed sex difference is in concordance with previous findings in the mouse , where sex dependent differences have been shown for the Hrl ( Herpes resistance locus ) locus and the Sml ( sex modifier locus ) locus which enhance resistance in females [8] . These two QTLs identified on mouse chromosome 6 correspond to a region towards the end of rat chromosome 4 , outside the identified QTLs found in this study . However , Hse1 was confirmed as the main disease regulating region by infecting congenic lines . The antigen-presenting lectin-like receptor gene complex ( APLEC ) located towards the end of chromosome 4 was previously reported to have a disease regulatory effect in a different model of HSV-1-induced encephalitis in the DA strain [15] . However , in this present study by infecting R17 congenic [24] rats with HSV-1 , we could not confirm any HSE regulation by the APLEC genes region . The use of additional microsatellite markers , inclusion of more F2 rats with allelic recombinations and haplotype mapping in a set of inbred strains within the Hse1 region made it possible to narrow down the confidence interval to a region of 3 genes . However , the finding that HSE resistant BN rats carry the same genotype as the DA rats in this region , suggests the existence of additional gene regions regulating HSE resistance in the BN rat . This demonstrates the significance of studying the genetic regulation in several inbred strains to enable the identification of all genes influencing the complexity of disease susceptibility . Sequencing of the genes in the Hse1 region identified a number of SNP variants mainly in the Ccdc132 , Calcr and Tfpi2 genes of the PVG . A strain , all of which were silent SNP variations . These silent variants do not change the amino acid sequence of the proteins; nevertheless these variants could possibly affect the translational stability , splicing or transcriptional control of these genes in PVG rats . Interestingly , we found that the mRNA expression of Calcr was significantly higher in the whiskers area both in the naïve and the infected PVG . A rats . Two microRNAs ( miRNAs ) are present within the Calcr , rno-mir-489 and rno-mir-653 . MiRNAs are short ( 22±3 nucleotides ) RNA molecules that post-transcriptionally regulate gene expression by binding to 3′-untranslated regions ( 3′UTR ) of target mRNAs , thereby inducing translational silencing and/or transcript degradation [25] . Both rno-miR-489 and rno-miR-653 are predicted to regulate a vast numbers of genes , making speculation on miRNA function difficult ( http://www . microrna . org/microrna/home . do; http://www . targetscan . org/; http://www . mirdb . org/ ) . Notably , there were no SNP variations in the sequence of mir-489 and mir-653 in DA and PVG . The levels of expression of the mature miR-489 and miR-653 were significantly higher at 5 dpi in PVG rats , resembling Calcr expression pattern in the whiskers area . However , both mature and mature* miR-489 and miR-653 were less abundant in the whiskers tissue , arguing against an influence on HSE susceptibility . Calcitonin receptor ( Calcr ) is a seven-transmembrane G protein-coupled receptor which binds the peptide hormone calcitonin ( 32 amino acid residue ) , secreted by the parafollicular cells of the thyroid gland and is involved in the maintenance of the calcium homeostasis , particularly with respect to bone formation and metabolism . The ‘calcitonin family’ is a group of peptide hormones that share structural similarities with calcitonin and includes calcitonin gene-related peptide ( CGRP ) , amylin , adrenomedullin and adrenomedullin 2 ( intermedin ) . Heterodimerization of CalcR with any of the three receptor activity modifying proteins ( RAMPs ) forms the multimeric amylin receptors AMY1 ( CT+RAMP1 ) , AMY2 ( CT+RAMP2 ) , and AMY3 ( CT+RAMP3 ) [26] , [27] . The CalcR is expressed in a variety of tissues and cell types including the CNS , which also differs according to the developmental stage [28] , [29] . In bone , it is restricted to osteoclasts , where it regulates their activity [30] . Little is known about the role of the CalcR in other tissues and it has not been previously implicated in infectious conditions . Of great interest is the amylin hormone ( also known as islet amyloid polypeptide ( IAPP ) ; 37 amino acid residues ) , which is similar in structure to calcitonin hormone and signals through Calcr . It is secreted by pancreatic β-cells parallel to insulin and is associated with type 2 diabetes development [31] . The in vivo use of CalcR agonist rat Amylin to modulate CalcR significantly enhanced the survival of DA rats after HSV-1 infection compared to controls . However , modulating an in vitro system using CalcR transfection of cell lines suggested that the presence of CalcR does not directly influence the infectivity of cells . Nevertheless , amylin signaling through CalcR could decrease the viral infection and/or replication inside cells through a more complex mode of action . In the same way the differences in expression levels together with the tissue localization of CalcR in DA and PVG rats could play a role in the degree of signaling through the receptor and thereby affect further viral spread . In addition , the high CalcR protein expression in the peri- and endoneurium layers in DA rats , together with the low Calcr expression in the trigeminal ganglia suggest a possible alternative route of axonal transport to the CNS causing encephalitis . In conclusion , we here demonstrate that Hse1 is the main genetic determinant for the susceptibility of DA rats to HSE and that it co-regulates differences in expression and tissue localization of Calcr . In addition , a direct clinical effect is evident by in vivo modulation of CalcR signaling . In vitro experiments , however , do not support a role of CalcR simply as a regulator of viral entry into cells , but rather to modulate infectivity and replication in a more complex fashion . Further studies are needed to define the contribution of the Calcr gene to HSE susceptibility , which may define novel mechanistic pathways involved in HSV-1 pathogenesis .
This study was carried out in accordance with the guidelines from the Swedish National Board for Laboratory Animals and the European Community Council Directive ( 86/609/EEC ) and approved by the Swedish ethical committee ( Stockholm's North Ethical Committee - Stockholms Norra Djurförsöksetiska Nämd ) ( ethical permits N128/04 , N340/08 , N32/11 ) . The inbred rat strains Dark Agouti-RT1av1 ( DA ) and MHC- ( RT1 . AV1 ) congenic strain on Piebald Viral Glaxo-RT1av1 background ( PVG . A ) were obtained from in-house breeding at the Animal Facility of Center for Molecular Medicine , Karolinska Institutet , Sweden . All rats used in experiments were 45 days old when infected with 2×106 PFU of neurovirulent HSV-1 ( strain I-2762 ) in the whiskers area . Susceptible inbred strains debut with severe clinical HSE symptoms including coordination/balance disturbance , paralysis and/or die at 5 dpi . All rats were monitored for clinical HSE symptoms and weighed daily for 11 days , the set end time-point of the experiment . This was done considering that F2 animals possess different genetic composition compared to parental inbred strains and could present a different disease course . The HSE phenotypes followed in this study include; 1 ) Incidence: diseased rats were defined by the detection of clinical symptoms of HSE such as difficulties with coordination and balance , paralysis , weight loss >20% and death before day 10; whereas not diseased rats were defined as rats not showing any HSE symptoms . 2 ) Onset: defined by the first day of two consecutive days of weight loss or death . 3 ) Body weight change: measured by the differences in body weight at day 0 , the start of the experiment and weight at 5 dpi . If animals showed body weight loss >20% , ataxia or paralysis the rats were euthanized and considered diseased . HSV-1 virus strain I-2762 was used as described in our previous studies [10] , [11] . After being thawed to room temperature , 100 µl virus suspension , containing 2×106 PFU HSV-1 was injected instantaneously subcutaneously ( s . c . ) into the area of the whiskers' base unilaterally , on the right side , under 2% Isoflurane ( Baxter ) anesthesia . Genomic DNA was extracted from tail tips using a standard protocol [33] . Polymorphic microsatellite markers were selected from available Internet databases: Rat Genome Database ( http://rgd . mcw . edu ) , Center for Genomic Research , Whitehead Institute/MIT ( http://www-genome . wi . mit . edu/rat/public/ ) , Ensembl ( http://www . ensembl . org/ ) and The National Center for Biotechnology Information is available at ( http://www . ncbi . nlm . nih . gov/ ) . Oligo 6 . 0 software ( National Bioscience ) was used to design new microsatellite markers on rat chromosome 4 ( D4Kini1–D4Kini16 ) from generated sequences available in Ensembl . Genotyping was performed using both fluorescent and radioactive methods . Flourophore-conjugated primers were purchased from Applied Biosystems ( Carlsbad , CA , USA ) . PCR amplification was performed using a standard protocol and PCR products were separated using the electrophoresis capillary sequencer ( ABI3730 ) and analyzed in the GeneMapper v3 . 7 software ( Applied Biosystems ) . Radioactive PCR amplification was performed as previously described [34] with [γ-33P] ATP end-labeled forward primers ( PROLIGO , France ) . The PCR products were size fractioned on 6% polyacrylamide gels and visualized by autoradiography . All genotypes were evaluated manually by two independent observers . Linkage analysis was performed using the statistical software R 2 . 8 . 0 ( http://www . r-project . org ) with the R/qtl package version 1 . 05–2 [35] and the marker map was obtained from Ensembl , version 45–2007 . 180 individual rats from the F2 generation ( total: 239 , Table 1 ) were included in the linkage analysis , 74 females and 106 males . All rats were genotyped with 127 evenly spaced microsatellite markers providing 97% and 91% genome coverage with 25 cM to 20 cM spacing . Single-QTL genome scans were implemented by using the “Scanone” function of R/qtl with imputation method ( step = 2 . 5 , n . draws = 64 ) [36] for the following phenotypes: incidence , onset and body weight loss . The phenotype disease was also scanned using the binary model ( step = 2 . 5 ) and similar results were obtained . The logarithm of odds ( LOD ) thresholds for a significant QTL [37] was obtained by performing permutations using 1000 simulations at the 95% and 63% confidence intervals ( CI ) , respectively [38] . The threshold level of 95% was considered as significant linkage , whereas threshold level of 63% were considered as suggestive linkage at different given levels . Peak markers were in Hardy-Weinberg equilibrium . A confidence interval for linkage was defined by the utmost closest microsatellite marker after a LOD drop of 1 . 5 [39] . All traits were analyzed in the complete set including both males and females but also re-analyzed in each gender subgroup . To identify polygenic influence on HSE , we used forward selection to a model of 10 additive/interactive QTLs followed by backward elimination to the null model to identify a multiple-QTL model . The fit to a multiple-QTL model was used to statistically validate the independent effect of each identified QTL and percentage of phenotypic variance explained by identified multiple-QTL models . Allelic effects of QTLs identified in the multiple-QTL model and significance levels for phenotypic differences between parental strains were calculated using two-sided Student's t test using GraphPad Prism 5 . 0 ( San Diego , CA , USA ) . A value of P≤0 . 05 was considered statistically significant . All sequences were generated from our genome-wide sequencing ( Diana Ekman et al , manuscript in preparation ) . Genomic DNA from DA and PVG rats was used to construct 3 mate-pair libraries . The libraries were sequenced in the Uppsala Genome Center ( Sweden ) with SOLiD next generation sequencing version 2 and 3 machines . The sequences of the following genes: F1LVX5_RAT ( ENSRNOG00000009135 ) ; Fam133b ( ENSRNOG00000009163 ) ; F1MA87_RAT ( ENSRNOG00000009258 ) ; F1M477_RAT ( ENSRNOG00000039809 ) ; F1LZE6_RAT ( ENSRNOG00000033874 ) ; F1M482_RAT ( ENSRNOG00000026450 ) ; D4A1X5_RAT ( ENSRNOG00000039801 ) ; D4A1X6_RAT ( ENSRNOG00000039800; D4A1X7_RAT ( ENSRNOG00000039799 ) ; D4A1Y1_RAT ( ENSRNOG00000039798 ) ; E9PTD6_RAT ( ENSRNOG00000009841 ) ; Hepacam2 ( ENSRNOG00000009711 ) ; Ccdc132 ( ENSRNOG00000009894 ) ; Calcr ( ENSRNOG00000010053 ) and Tfpi2 ( ENSRNOG00000010513 ) were mapped to reference genome ( BN ) from Ensembl database release 66 , with a >80% SNPs detection . Twenty rats were used for immunohistochemistry , 5 DA and 5 PVG . A HSV-1 infected and dissected at 5 dpi , as well as 5 naïve DA and 5 naïve PVG . A . The procedure used and antibodies were described in a previous study [11] . In addition mouse anti-calcitonin receptor ( 1∶100 ) ( Dako , Denmark ) was used to stain expression of Calcitonin receptor in the different compartments . Fifty-two male DA rats were used for the in vivo receptor modulating experiment . All rats were 44 days old when pre-treated 18 hours prior to HSV-1 infection . Under isoflurane anesthesia stimulation substances were injected unilaterally into the right whiskers' pad , where the HSV-1 was also injected the day after , under anesthesia . Sixteen rats were pre-treated with rat Amylin ( 0 . 05 mg/rat∼0 . 25 mg/kg body weight ) ( Bachem , Switzerland ) , 16 rats with Calcitonin ( 8–32 ) Salmon I ( 0 . 05 mg/rat∼0 . 25 mg/kg body weight ) ( Bachem , Switzerland ) , 10 rats with specially ordered scrambled peptide ( 0 . 05 mg/rat∼0 . 25 mg/kg body weight ) ( CASLO Laboratory , Lyngby , Denmark ) consisting of the same amino acids as the rat Amylin , however in a reverse order and 10 rats were not pre-treated and used as infected controls . Rats were weighed and observed daily for disease symptoms until 11 dpi . Transfections were performed with plasmids encoding for a myc-tagged version of the receptor activity-modifying protein 1 ( RAMP1 ) , necessary for modulating the CalcR signaling towards amylin , and with either the human CalcR or the rat CalcR ( pcDNA3 . 1 and pcDNA1 , respectively , all kind gifts from Professor Patrick Sexton , Monash University , Victoria , Australia ) . Briefly , HEK293T cells ( human embryonic kidney cell line ) , which are semi-permissive to HSV-1 infection but do not express the CalcR , were plated in 24 wells plates at a density of 0 , 2×106 cells/ml . Transfections were performed 24 hours later with 150 µg/well of all plasmids with Effectene transfection reagent ( Qiagen ) according to the manufacturer's instructions . Infections were performed 24 hours after transfection with the HSV-1 virus strain I-2762 at 3×105 PFU/ml and 500 µl/well for 30 minutes at 37°C . Cells were subsequently washed once and rested for additionally 24 hours before staining . For the experiments with in vitro modulation of CalcR activity , the CalcR agonist rat Amylin ( Bachem , Switzerland ) was added either 4 hours prior to , and left during infection , being replenished after washing the infected cells; or was added only after infection . Experiments on the HSV-1 non-permissive rat mammary adeno-carcinoma cell line 13762 MAT B III ( ATCC CRL-1666 ) were performed under the same conditions as described above . Stainings for CalcR expression were performed on cells transfected for 24 hours . Briefly , cells were collected by pipetting and then fixed in Cytofix/cytoperm ( BD Biosciences ) for 20 minutes , washed and subsequently incubated with the Alexa Fluor 647 Conjugated mouse anti Myc-tag antibody ( 9B11 ) from Cell Signaling for the detection of RAMP1 , and the rabbit polyclonal to CalcR ( Ab11042 ) from Abcam followed by Alexa Fluor 488 Conjugated goat anti rabbit IgG ( Molecular Probes ) for the detection of both human and rat CalcR . For the assessment of virus infections , 24 hours infected cultures were collected as above , fixed and stained for RAMP1 as well as with the rabbit anti-HSV1 antibody B0114 ( Dako ) followed by an anti-rabbit secondary antibody as specified above . RAMP1 transfected cells ( either alone or co-transfected with the human or rat CalcR plasmids ) were gated and analyzed for the percentage of HSV-1 positive cells in the RAMP1-positive gate as well as for the mean fluorescence intensity ( MFI ) of the HSV-1 infected cell population . All samples were acquired by a Gallios flow cytometer and analyzed using Kaluza software ( Beckam Coulter ) . | Herpes simplex encephalitis ( HSE ) is a rare , but severe infection of the central nervous system ( CNS ) caused by Herpes simplex virus type 1 . We have previously characterized a model for HSE in the inbred DA rat which resembles human HSE . Interestingly the inbred PVG rat is completely resistant to the disease and displays reduced or no uptake of viral particles into the peripheral and central nerve compartments respectively . To identify the gene ( s ) regulating HSE pathogenesis , we crossed the susceptible DA and the resistant PVG . A rats for two generations and infected 239 rats of the F2 ( DAxPVG . A ) cohort with HSV-1 . A genome-wide linkage scan demonstrated one strong quantitative trait locus ( QTL ) , Hse1 , on rat chromosome 4 regulating disease susceptibility . Fine mapping , haplotype mapping , sequencing and expression analysis of the genes in the Hse1 interval collectively support the underlying genetic variation to be located in , or adjacent to the calcitonin receptor gene ( Calcr ) . Further support for a role of CalcR in regulating HSV-1 replication and propagation is provided by strain-dependent differences in the calcitonin receptor protein tissue localization and in functional studies . Using an unbiased genetic mapping approach this study identifies Calcr as a candidate for regulating susceptibility to HSE . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"herpes",
"simplex",
"neurology",
"neurological",
"disorders",
"viral",
"diseases",
"encephalitis"
] | 2012 | The Calcitonin Receptor Gene Is a Candidate for Regulation of Susceptibility to Herpes simplex Type 1 Neuronal Infection Leading to Encephalitis in Rat |
Non-retroviral RNA virus sequences ( NRVSs ) have been found in the chromosomes of vertebrates and fungi , but not plants . Here we report similarly endogenized NRVSs derived from plus- , negative- , and double-stranded RNA viruses in plant chromosomes . These sequences were found by searching public genomic sequence databases , and , importantly , most NRVSs were subsequently detected by direct molecular analyses of plant DNAs . The most widespread NRVSs were related to the coat protein ( CP ) genes of the family Partitiviridae which have bisegmented dsRNA genomes , and included plant- and fungus-infecting members . The CP of a novel fungal virus ( Rosellinia necatrix partitivirus 2 , RnPV2 ) had the greatest sequence similarity to Arabidopsis thaliana ILR2 , which is thought to regulate the activities of the phytohormone auxin , indole-3-acetic acid ( IAA ) . Furthermore , partitivirus CP-like sequences much more closely related to plant partitiviruses than to RnPV2 were identified in a wide range of plant species . In addition , the nucleocapsid protein genes of cytorhabdoviruses and varicosaviruses were found in species of over 9 plant families , including Brassicaceae and Solanaceae . A replicase-like sequence of a betaflexivirus was identified in the cucumber genome . The pattern of occurrence of NRVSs and the phylogenetic analyses of NRVSs and related viruses indicate that multiple independent integrations into many plant lineages may have occurred . For example , one of the NRVSs was retained in Ar . thaliana but not in Ar . lyrata or other related Camelina species , whereas another NRVS displayed the reverse pattern . Our study has shown that single- and double-stranded RNA viral sequences are widespread in plant genomes , and shows the potential of genome integrated NRVSs to contribute to resolve unclear phylogenetic relationships of plant species .
Events of horizontal gene transfer ( HGT ) have been identified between various combinations of viruses and their eukaryotic hosts . HGT can occur during evolution in 2 inverse directions: “from host to virus” or “from virus to host . ” In the host to virus direction , viral acquisition of host genes is observed as insertion of cellular genes for proteases ( see [1] for review ) , ubiquitin [2] , chloroplast protein [3] and heat-shock proteins [4] , [5] into viral genomes . The virus to host direction involves endogenization of viral genes . Fossil sequences of viral origin , mostly from retroviruses , have been detected in many animal genomes . However , retrovirus sequences have not been identified in plants; instead , reverse-transcribing DNA viruses ( pararetroviruses ) have been identified . Although pararetroviral sequences have been found in some plant nuclear genomes [6] , [7] , [8] , [9] , only a limited number of integrated sequences are exogenized to launch virus infection; however , their cellular functions remain unclear in other examples . In contrast , the sequences of non-retroviral RNA viruses were considered not to integrate into host chromosomes . However , recent reports identified endogenized genes of non-retroviral elements in mammals [10] , [11] , [12] , [13] . Examples include the nucleocapsid protein ( N ) and nucleoprotein ( NP ) genes of bornaviruses and filoviruses , members of the negative-strand RNA virus group in the order Mononegavirales [11] , [12] , [14] . While some integrated N genes are expressed , their biological significance is unclear . Identification of these sequences contrasts with the lack of evidence for negative-strand RNA virus genome integration into plant genomes . Furthermore , RNA-dependent RNA polymerase ( RdRp ) and capsid protein ( CP ) coding domains from a group of monopartite dsRNA viruses have been identified in yeast chromosomes , and while some of these viruses appear to be expressed , their biological significance has not been explored [15] , [16] , [17] . The white root rot fungus Rosellinia necatrix is a soil-borne phytopathogenic ascomycetous fungus that causes damages to perennial crops . An extensive search of a large collection of field fungal isolates ( over 1 , 000 ) was conducted to identify dsRNA ( mycoviruses ) that may serve as virocontrol ( biological control ) agents . Approximately 20% of field isolates were infected with known or unknown viral strains [18] , [19] , [20] . During molecular characterization of these viruses , we identified a novel partitivirus termed Rosellinia necatrix partitivirus 2 ( RnPV2 ) in an ill-defined R . necatrix strain . The family Partitiviridae contains members with small bi-segmented dsRNA genomes [21] that infect plants , fungi or protozoa . They are thought to replicate using virion-associated RdRp in the host cytoplasm , which are phylogenetically related to those from the picorna-like superfamily [22] . Surprisingly , the RnPV2 CP showed the highest level of sequence identity to an Arabidopsis thaliana gene , IAA/LEU resistant 2 ( ILR2 ) , which was previously shown to regulate the activity of the phytohormone auxin [23] . Combined with information regarding integrated mononegaviral sequences in animals , this finding generated significant interest in searching currently available genome sequence data for not only dsRNA but also negative-strand viral sequences . In October 2010 , Liu et al . [24] reported similar results based on an extensive search conducted in 2009 . This group identified sequences in the chromosomes of diverse organisms that may have been acquired from monopartite ( totiviruses and related unclassified viruses ) and bipartite dsRNA viruses ( partitiviruses ) . We further examined plant genome sequences available as of December 10 , 2010 for integrated sequences of not only partitivirus genomes but also negative- , and positive-strand RNA viruses ( Table S1 ) . Combining database searches and molecular analyses led to the identification of multiple endogenized sequences related to partitiviruses , cytorhabdoviruses , varicosaviruses and betaflexiviruses in the genomes of a variety of plants including those from the families Solanaceae and Brassicaceae . For example , while some partitivirus-related sequences are conserved on the orthologous locus across some genera , e . g . , Arabidopsis , Capsella , Turritis , and Olimarabidopsis within the family Brassicaceae , others are retained in only a few species within a single genus , Arabidopsis . A similar integration pattern was observed for a rhabdovirus-related sequence in the family Solanaceae . These profiles of occurrence can potentially resolve unclear phylogenetic relationships between plants . Our study demonstrates widespread endogenization of non-retroviral RNA virus sequences ( NRVSs ) including sequences of plant positive- and negative-strand RNA viruses for the first time . We have proposed a model of viral gene transfer , in which NRVSs are suggested to be a factor constituting plant genomes .
We determined the complete nucleotide ( nt ) sequence of the genome segments ( dsRNA1 and dsRNA2 ) of a novel partitivirus , RnPV2 , from the white root rot fungus Rosellinia necatrix , a soil-borne phytopathogenic ascomycetous fungus . DsRNA2 was found to be 1828 nt long , encoding a polypeptide of 483 amino acids ( aa ) ( CP , 54 kDa ) . Low-level sequence similarities among CPs from Partitiviridae family members were observed using a BLASTP search with RnPV2 CP against non-redundant sequences available in the NCBI database ( http://www . ncbi . nlm . nih . gov/ ) . Surprisingly , RnPV2 CP showed the highest degree of sequence similarity to ILR2 from Ar . thaliana . Notably , sequence similarities between RnPV2 CP and ILR2 were greater than those between the CP sequence from another mycovirus , Sclerotinia sclerotiorum partitivirus S ( SsPV-S ) and ILR2 noted previously [24] . ILR2 is known to regulate indole-3-acetic acid ( IAA ) -amino acid conjugate sensitivity and metal transport . An Ar . thaliana mutant with a single amino acid substitution in ILR2 , known as ilr2-1 , was shown to exhibit normal root elongation in the presence of a high concentration of exogenous IAA-leucine conjugates , which represses root elongation in wild-type lines [23] . Magidin et al . [23] identified 2 alleles of ILR2 in Ar . thaliana accessions ( a long and a short allele ) ( Figure 1A ) . Although the authors confirmed ILR2 expression for only the WS ecotype ( short allele ) , they determined that both short and long versions of ILR2 were functional . Given the similarity between ILR2 and RnPV2 CP sequences , we hypothesized that HGT occurred between the 2 organisms . Therefore , we assessed the extent to which ILR2 is conserved in plants . We used 3 approaches: BLAST search , genomic PCR , and Southern blot analyses . We first conducted an exhaustive BLAST ( tblastn ) search against genome sequence databases as described in the Materials and Methods . This search identified ILR2 homologs in Ar . lyrata and Mimulus guttatus ( yellow monkey flower ) , which included both short and long versions of ILR2 homologs with modest levels of aa sequence identities ( over 20% ) to RnPV2 CP ( Table S2 , Figure 1A ) . Furthermore , a variety of partitivirus CP-related sequences with low-levels of aa sequence identities ( approximately 20% ) to RnPV2 CP were also detectable from genome sequences from other 17 plant species ( Table 1 ) . These sequences were classified into a total of 8 subgroups based on relatedness to best matched extant partitiviruses ( Table 1 ) . Their nomenclature is: AtPCLS1 ( ILR2 ) is from Arabidopsis thaliana partitivirus CP-like sequence ( PCLS ) 1 . Differently numbered PCLSs , referring to proteins potentially encoded by PCLSs , show the highest level of aa sequence identities to CPs encoded by different partitiviruses . Genomic PCR analysis with primers corresponding to highly conserved 240-bp portions revealed that ILR2 homologs were retained in genera closely related to Arabidopsis , such as Capsella , Turritis , and Olimarabidopsis , but not in members of distantly-related genera , Brassica , Thellungiella , Crucihimalaya , Sisymbrium , and Thlaspi within the Brassicaceae family ( Figure 1B ) . Genomic PCR fragments covering the entire ILR2-like domains of the plants shown in Table S4 were sequenced directly or after cloning into a plasmid . It should be noted that PCLS1s of closely related genera reside in an orthologous position [25] , i . e . , in a convergent configuration with the gene for the transmembrane Golgi matrix protein AtCASP , which shares a high degree of sequence similarity across kingdoms [26] . This notion was confirmed by genomic PCR in which a primer pair allowed detection of 0 . 75- to 1-kb fragments spanning the CASP gene . Previous comparative genomics studies proposed a hypothesis that the Brassicaceae genomes consist of 24 ( A to X ) conserved genome blocks [27] . The ILR2 locus is on block F which is considered to be duplicated in B . rapa . A search against the Brassica database ( BRAD ) confirmed the absence of a PCLS1 on the 2 B . rapa loci that flank the CASP gene . Southern blotting with members of the Brassicaceae , Cucurbitaceae , Solanaceae , and Leguminosae families indicated that PCLS1 ( ILR2 ) is present in Ar . thaliana and Cap . bursa-pastoris , but absent in the other plants ( Figure 1C ) , consistent with BLAST results and genomic PCR analyses . Furthermore , the absence of ILR2 in Crucihimalaya lasiocarpa , Sisymbrium irio and B . rapa was confirmed by sequence analysis of genomic PCR fragments covering the entire ILR2 region and its flanking regions ( Figure 1D ) . Genome sequences with low levels of similarities to RnPV2 CP included a number of PCLSs from various plants spanning more than 17 species from 8 families ( Table 1 ) . Most PCLSs confirmed to be present on their chromosomes of these organisms were identified by genomic PCR and/or Southern blotting and sequencing ( Tables 1 , S4 ) . For instance , AtPCLS2 and Ar . lyrata PCLS3 ( AlPCLS3 ) are retained on non-orthologous loci of ILR2s of Ar . thaliana and Ar . lyrata , respectively ( Figure 2A ) . AtPCLS2 ( At4g14104 ) resides between the genes for COP9 ( constitutive photo-morphogenic-9 , COP9 ) and an F-box protein , while AlPCLS3 is between 2 coding sequences for F-box domains corresponding to At4g02760 and At4g02740 [25] . AtPCLS2 and AlPCLS3 from 2 closely related plant species show the highest sequence identities to the CPs from 2 different partitiviruses: Raphanus sativus cryptic virus 2 ( RSCV2 ) and Fragaria chiloensis cryptic virus ( FCCV ) ( dsRNA2 ) [28] . The PCLS retention profile was revealed by genomic PCR using 2 primer sets . A primer set designed to amplify internal AtPCLS2 sequences provided DNA fragments of an expected size of 470 bp in Ar . thaliana accessions Col-0 , Ler , and Shokei , but not in Ar . lyrata , Ar . Arenosa , or Cap . rubella ( Figure 2B , top panel ) . A different primer set specific for AtPCLS2 and the F-box protein gene ( At4g14103 ) gave the same amplification pattern ( Figure 2B , second panel ) as shown in the top panel . Using the same approach with 2 sets of primers , PCLS3 was detected by genomic PCR in Ar . lyrata and Ar . arenosa , while no such sequence was observed in Ar . thaliana ecotypes or Cap . rubella ( Figure 2B , third and fourth panels ) . Although the COP9 and the F-box protein genes are conserved on the corresponding loci of Ar . lyrata , no counterpart of AtPCLS2 was identified between the genes ( Phytozome ) . Similarly , no AlPCLS3 homolog was observed on the corresponding chromosomal position of Ar . thaliana [25] . PCLS4 and PCLS5 were found in the genome sequence databases of B . rapa ( BrPCLS4 and 5 ) , Solanum phureja ( wild species of potato ) ( SpPCLS5 ) ( Figure 3A , S2 ) , and Nicotiana tabacum ( NtPCLS5-1 and -2 ) ( Figure S1A ) . These sequences commonly exhibited greater sequence similarity to CPs of previously reported plant partitiviruses than to RnPV2 CP ( Tables 1 ) . The 3 PCLS5s from the Solanaceae family were very similar to each other ( approximately 60% aa sequence identity ) , and showed high sequence identity ( over 45% ) ( Table S2 ) to CP of Raphanus sativus cryptic virus 1 ( RSCV1 , plant partitivirus ) [29] . Two PCLSs , BrPCLS4 ( Bra021820 ) and BrPCLS5 ( Bra020160 ) , which are detected on different scaffolds , were determined to not flank the CASP gene of B . rapa as AtPCLS1 ( ILR2 ) does . BrPCLS4 and 5 show much greater aa sequence identities to CPs of RSCV1 and carrot cryptic virus 1 ( CaCV1 , plant partitivirus ) [30] than it does to RnPV2 CP ( Table S2 ) . Molecular analyses were performed to determine how widely these PCLS4 and PCLS5 are conserved . Genomic PCR using a primer set specific for BrPCLS4 detected related sequences in all Brassica species tested , but not in other plants including members of the family Solanaceae or genera other than Brassica in Brassicaceae , such as Ar . thaliana , Cru . lasiocarpa , Thellungiella parvula , Thl . arvense and Sis . irio , and Raphanus sativus ( Figure 3B , top panel ) . For BrPCLS5 , the primer set , PC5a-1 and PC5a-2 enabled detection of expected PCR fragments in all Brassica plants in addition to R . sativus , while no PCR fragments were amplified in the other plant species ( Figure 3B , second panels ) . A different detection profile was obtained by genomic PCR with a primer set specific for SpPCLS5 in which PCLS5-related sequences were detectable only in Sol . tuberosum and Sol . lycopersicum ( Figure 3B , third and fourth panels ) . We failed to yield amplification from all other tested plants in the families Brassicaceae and Solanaceae including Sol . melongena . Interestingly , PCLS5 , but not PCLS4 fragments , were detected in R . sativus . Moreover , the presence or absence of PCLSs was confirmed by genomic Southern analysis . As expected from the genomic PCR results , hybridization signals were detected with a BrPCLS4- or a BrPCLS5-specific probe in the Brassica species such as B . rapa and B . oleracea ( Figure 3C , top and second panels ) ; however , the numbers and signal positions differed between the 2 blots . The StPCLS4-specific probe allowed detection of 2 and 1 hybridization signals in Sol . tuberosum and Sol . lycopersicum , respectively , but not in any other plants examined in this study ( Figure 3C , fourth panel ) . In addition to PCLS1 to PCLS5 , 2 other subgroups of PCLSs ( PCLS6 and PCLS7 ) were observed in the GSS database of N . tabacum and showed an interesting detection pattern in Nicotiana species ( Figure S1 ) . NtPCLS6 and NtPCLS7 showed moderate aa sequence identities to CPs encoded by FCCV dsRNA3 ( 38% ) [28] and RSCV3 dsRNA2 ( 30% ) [29] , respectively . Sequencing of genomic PCR fragments and Southern blotting ( Figure S1B , E ) suggested that NtPCLS5-1 and NtPCLS5-2 are retained only in N . tabacum , but not in other Nicotiana species examined , such as N . benthamiana and N . megalosiphon , whereas PCLS6 was detected in both N . tabacum and N . megalosiphon ( Figure S1B ) . In contrast , PCLS7 is conserved in all 4 Nicotiana plants tested , although sequence divergence was observed among the PCLS7s . Other PCLSs from 2 legume plants , MtPCLS7 and LjPCLS8 were identified on their nuclear genomes by PCR ( Figures S1A , C , D ) . An expanded BLAST ( tblastn ) search against the EST sequence libraries ( in NCBI ) helped detect many related sequences of possible plant partitiviruses that shared moderate levels of sequence similarity . Some representative EST sequences , PCLSs and partitivirus CPs , whose entire sequences are available , were aligned using the MAFFT program . Three relatively well-conserved motifs are located on the N- terminal , central , and C-terminal regions of partitivirus CPs and PCLSs , and are represented by PGPLxxxF [31] , F/WxGSxxL and GpfW domains ( Figure S2 ) . As expected from sequence similarities , phylogenetic analysis of partitivirus CPs and PCLSs identified in plant genomes clearly show that members of each PCLSs subgroup ( PCLS1 , 2 , 4 , 5 , 7 , 8 ) clusters together with the CP of the respective partitivirus that shows the highest sequence similarities ( Figure 4 , Table 1 ) . For example , RnPV2 CP ( in red ) , MgPCLS1 , and ILR2 homologs ( PCLS1s ) from Arabidopsis-related genera ( in green ) constitute one group in the tree . The MgPCLS1 clade includes an assembled sequence in the EST database from meadow fescue ( Festuca pratensis ) ( in purple ) believed to be from a plant partitivirus . Another group includes PCLS5s from the families Brassicaceae and Solanaceae ( in green ) , CPs of fungal ( in red ) and plant partitiviruses ( in blue ) are grouped together . Within this group , PCLSs from the families Brassicaceae ( BrPCLS5 , BoPCLS5 , and BnPCLS5 ) and Solanaceae ( StPCLS5 , SpPCLS5 , SlPCLS5 , and NtPCLS5-1 ) comprised 2 subgroups that included CPs encoded by RSCV1 ( CP ) and RSCV1 dsRNA3 ( Figure 4 ) , respectively , which are considered to be from two different partitiviruses . PCLS4s from members of the genus Brassica clustered together with CPs of other plant partitiviruses including white clover cryptic virus 1 ( WCCV1 ) [32] , CaCV1 , beet cryptic virus 1 ( BCV1 ) [33] , and vicia cryptic virus ( VCV ) [34] . The tree topology shown in Figure 4 was similar to that reported by Liu et al . [24] . The current study used more PCLSs detected in various plants but not partial PCLSs such as PCLS3 and NtPCLS5-2 , 6 and 7 ( Tobacco Contig-2 , -3 and -4 ) analyzed phylogenetically by Liu et al . [24] . Because negative-strand RNA viral sequences are found in animal chromosomes , we searched for negative-strand RNA viral sequences ( Table S1 ) in plant genomes as described in the Materials and Methods . This search identified sequences related to the N protein in members of the genus Cytorhabdovirus ( Lettuce necrotic yellows virus , LNYV , Lettuce yellow mottle virus , LYMoV , and northern cereal mosaic virus , NCMV ) and a CP of the genus Varicosavirus ( Lettuce big-vein associated virus , LBVaV ) in the genomes of a variety of plants such as Populus trichocarpa , N . tabacum , and B . rapa ( Figures 5 , S3 , Table 2 ) . While varicosaviruses have bipartite genomes replicated in the cytoplasm of infected plant cells , they are phylogenetically closely related to cytorhabdoviruses with monopartite genomes [35] , [36] . Varicosavirus CP is phylogenetically and functionally equivalent to rhabdovirus N . Thus , these plant nuclear sequences were designated as rhabdovirus N-like sequences ( RNLSs ) and classified into 4 subgroups ( RNLS1 to RNLS4 ) based on the sequences of presently existing viruses with the highest levels of sequence similarities ( Table 2 ) . Their potentially encoding proteins were designated as RNLSs as in the case for PCLSs . To confirm the presence of the RNLSs in plant chromosomes , we conducted genomic PCR and Southern blot analyses . Interestingly genomic PCR with primers specific for an RNLS1 from B . rapa ( BrRNLS1 ) detected RNLS1s in R . sativus and all tested plants within the Brassica genus , but not in members in other genera ( Figure 5C ) , in a pattern similar to that of PCLS5s from the family Brassicaceae ( Figure 3B ) . Consistent with these results , Southern blotting detected hybridization signals in 3 Brassica plants ( Figure 5D ) with a probe specific for BrRNLS1 . The NtRNLS2 sequence was detected in N . tabacum , while no fragments were generated from other Nicotiana species using genomic PCR ( Figure 5E ) . Southern blotting results supported this detection profile ( Figure 5F ) ; N . tabacum , but not N . benthamiana , was shown to carry an NtRNLS2-related sequence ( Figure 5F , left panel ) . All other RNLSs discovered through the similarity search of genome sequence databanks ( Table 2 ) , except for PtRNLS4 from Pop . trichocarpa and TcRNLS1 from Theobroma cacao , were shown to be retained on respective plant genomes by genomic PCR and subsequent sequencing ( Figure S3 ) . RNLS1s molecularly analyzed included those from Aquilegia flabellata ( a close relative of Aq . coerulea ) ( AfRNLS1 ) , Lotus japonicus ( LjRNLS1 ) , Malus x domestica ( MdRNLS1 ) and Cucumis sativus ( CsRNLS1 ) ( Figure S3B–H ) . The AqfRNLS1 sequence defined in this article showed approximately 98% nt sequence identity to AcRNLS1 whose sequence is available in the database ( Phytozome ) . LjRNLS1-1 from L . japonicus line B129 and CsRNLS1 from 3 cucumber varieties ( Hokushin , Suyo , and ‘Borszcagowski’ line B10 ) were identical to the reported RNLS1 sequences for line MG-20 ( Kazusa DNA Research Institute ) and ‘Chinese long’ line 9930 [37] , respectively . Approximately 97% nucleotide sequence identity was found between MdRNLS1s of cultivars ‘Sun-Fuji’ and ‘Golden Delicious . ’ ‘Golden Delicious’ is currently used in the apple genome sequence project [38] ( http://www . rosaceae . org/projects/apple_genome ) . These examined RNLS sequences are listed in Table S5 . Several sequences found through searching plant EST databases ( Table S6 , Figure S4 ) were included in our phylogenetic analysis . Deduced amino acid sequences of plant RNLSs , the N ( CP ) proteins of negative-strand RNA viruses , and related EST entries were aligned using the MAFFT program ( Figure S5 ) . Pair-wise similarities between selected RNLSs and viral N ( CP ) sequences are shown in Table S7 . Two amino acid segments , GmH and YaRifdxxxfxxLQtkxC are relatively well-conserved among these sequences . A dendrogram generated on the basis of alignment showed 4 major groups containing plant RNLSs ( Figure 6 ) . RNLS1s are separated into two major groups . The first group includes varicosavirus CPs and RNLS1s from apple , cucumber and Brassica plants ( MdRNLS1 , CsRNLS1 , BoRNLS1 , and BrRNLS1 ) in addition to a few ESTs . The second group accommodates RNLS1s from Aquilegia and Lotus ( AqfRNLS1 , AqcRNLS1 , LjRNLS1 ) , together with an RNLS2 from Mim . guttatus ( MgRNLS2 ) and EST sequences from Cichorium intybus and B . oleracea . The placement of MgRNLS2 in this group may be explained by low-level sequence identity to its most closely related extant varicosavirus , LNYV ( Table 2 ) . NtRNLS3 , PtRNLS4 , and Ns of cytorhabdoviruses ( LNYV , LYMoV , and NCMV ) form the third group ( Figure 6 ) . A dichorhabdovirus ( orchid fleck virus , OFV ) and nucleorhabdoviruses ( PYDV and SYNV ) , replicating in the nuclei of host plants , are placed into an independent clade . Whether most of the analyzed ESTs originated from viruses or plant chromosomes is unknown . However , an EST from F . pratensis is presumed to originate from a plant virus in our preliminary experiment not only because the N ( CP ) - but also the L ( RdRp ) -derived ESTs were detected in the same EST library of F . pratensis . This suggests a presently existing virus more closely related to RNLSs of the genus Brassica than LBVaV , because both N- and L-related sequences are rarely found in a single plant genome ( Table 2 ) . Extensive searches of genome sequence databases for plant plus-strand RNA viral sequences were conducted using genome sequences of various plus-strand RNA viruses representing the major virus genera and families Potyviridae , Luteoviridae , Tombusviridae , and Bromoviridae ( Table S1 ) . Compared to searches for double- or negative-strand RNA viral sequences , the search for plus-strand RNA virus sequences yielded a much smaller number of hits . The Medicago truncatula database ( HTGS ) contains sequences of 320 and 475 nts with over 98% sequence identity to the capsid and movement protein genes of cucumber mosaic virus , a member of the family Bromoviridae . However , this sequence was not amplified in Med . truncatula line A17 used in the genome sequence project by genomic PCR with different sets of internal and external primers . A sequence similar to replication-related genes of citrus leaf blotch virus ( CLBV ) [39] belonging to the family Betaflexiviridae , is identified in the complete genome databases for the cucumber ‘Chinese long’ line 9930 [37] and termed Cucumis sativus flexivirus replicase-like sequence 1 , CsFRLS1 ( Figure 7A ) . The GSS database of cucumber ‘Borszczagowski’ line B10 also contains CsFRLS1 ( http://csgenome . sggw . pl/ ) , but its available sequence is fragmented ( Figure 7A , dashed purple bar ) and shorter than that in the complete genome sequence data base . Two independent cucumber genome databases for 2 different lines strongly suggest the presence of CsFRLS1 in the cucumber chromosome . We confirmed this by genomic PCR using different sets of primers corresponding to methyltransferase ( Met ) and RNA helicase ( Hel ) domains , the inter-domain region ( FR1-3 and FR1-4 ) and the entire CsFRLS1 region ( Figure 7B ) . DNA fragments of expected sizes were amplified on genomic DNA from the ‘Borszczagowski’ line B10 , but not from watermelon , Citrullus lanatus ( Figure 7B ) . Furthermore , genomic PCR fragments covering FRLS1 and its flanking putative open reading frames ( ORFs ) were amplified , strongly suggesting that FRLS1 resides on the nuclear genome as shown in Figure 7A and B . The phylogenetic tree containing CsFRLS1 potentially encoded by CsFRLS1 and its counterparts from related viruses shows that CsFRLS1 is closely related to the genus Citrivirus within the family Betaflexiviridae ( Figure 7C ) . The distance between CsFRLS1 and citriviruses are similar to intra-genus distances in the genera Carla- , Fovea- , Viti- and Potexviruses .
The finding that the CP of a novel partitivirus , RnPV2 from a fungal phytopathogen matched a plant gene product , ILR2 from Ar . thaliana initiated a comprehensive search of the plant genomic sequence data available as of December 10 , 2010 for non-retroviral RNA virus sequences ( NRVSs ) in plant genomes . While this study showed a variety of sequences related to the N ( CP ) genes of negative-stranded RNA viruses ( cytorhabdoviruses and varicosaviruses ) in members in the plant families including Solanaceae , Leguminosae , Brassicaceae and Phrymaceae , only one plus-sense RNA virus-related sequence ( betaflexivirus replication-related gene ) was found to be present in the cucumber genome . Furthermore , this survey detected sequences related to CP from dsRNA viruses ( partitiviruses ) ( PCLSs ) in various plants in addition to PCLSs reported by Liu et al . [24] . These authors performed a thorough search of eukaryotic genomic sequences available as of September 2009 for NRVSs and showed multiple dsRNA virus-related sequences not only in plants but also animals . Importantly , many of the NRVSs revealed by BLAST searches in this study were subsequently identified in plant genomes by Southern blotting , genomic PCR and sequence analyses ( Figures 1–3 , 5 , 7 , S1 , S3 ) . These findings provide interesting insights into plant nuclear genome evolution , plant phylogeny and virus/host interactions . Horizontal gene transfer , HGT , can occur “from virus to plant” or “from plant to virus . ” A retention profile of PCLS1 among plants strongly suggests that HGT may have involved the former direction . The family Brassicaceae of the order Brassicales includes the genus Arabidopsis , which is believed to have diverged after the split of the families Phrymaceae and Solanaceae , accommodates the genera Mimulus and Solanum and belong to different orders , Lamiales and Solanales , respectively ( Figure 8 ) . No PCLS1 homologs are found in Vitis vinifera or Carica papaya , and that this gene resides on non-orthologous chromosomal positions of Mim . guttatus ( data not shown ) and Arabidopsis-related species ( Figure 1A ) . This strongly suggests that independent HGT events from virus to the Arabidopsis and Mim . guttatus lineages may have occurred ( Figure 8 ) . This observation is also true for other PCLSs . The families Solanaceae and Brassicaceae contain PCLS5s , while their counterparts are not found in other plants whose complete genome sequences are available ( Figure 8 ) . The observation that a relatively widely conserved gene PUX_4 is disrupted in Sol . phureja by SpPCLS5 ( Figure 3A ) provides additional evidence for its insertion into the PUX_4 locus . The HGT direction “from virus to plant” was further confirmed by phylogenetic analysis showing that plant PCLSs and partitivirus CPs are placed in a mixed way ( Figure 4 ) . Viral sequences are basal in each of the three major clades , supporting the direction of transfers from virus to plant . The divergence time of plant lineages is estimated through a classical approach using fossils and mutations rates of some particular genes . Alternatively , if we assume that cellular genes evolve at a constant rate , their divergence time can be calculated from the genome-wide , spontaneous mutation rate determined on a generation basis in the laboratory [40] . Together with the patterns of occurrence of the non-retroviral integrated RNA virus sequences , these values allow us to estimate time of some , if not all , HGTs identified in this study . For example , the integration of PCLS1 ( ILR2 ) may have post-dated the split of the lineages containing the genera Arabidopsis and Brassica ( 16 . 0–24 . 1 million years ago ) and pre-dated the speciation of Arabidopsis spp . , or more accurately the divergence of Arabidopsis and its closely related genera ( Figure 8 ) ( 10–14 million years ago ) [40] , [41] , [42] . The phylogenetic relation among PCLS1s from Arabidopsis and its close relatives within the tribe Camelina ( Capsella , Olimarabidopsis , and Turritis ) agrees with the phylogeny of the family Brassicaceae deduced from systematic analyses [43] . Moreover , assuming that the Ar . thaliana and Ar . arenosa separated 10 million years ago , the mutation rates calculated for PCLS1s between the 2 plants are estimated to be 6 . 8×10−9 base substitutions per site per year , a value close to the genome-wide base substitution rate , 7×10−9 , reported for Ar . thaliana by Ossowski et al . [40] . These observations suggest that endogenized PCLS1s accumulated mutations in a manner similar to those of other nuclear sequences during the course of evolution after a single HGT event in an ancestral Arabidopsis plant . The genome of B . rapa in the family Brassicaceae retained 2 PCLSs ( BrPCLS4 and BrPCLS5 ) with low-level similarities to RnPV2 CP on chromosomal positions different from each other and from that of the PCLS1 ( ILR2 ) homologs of Arabidopsis-related genera . No PCLS1 homolog was identified on the orthologous positions of the B . rapa genome , and no BrPCLS4 or BrPCLS5 homologs were found on the corresponding locus of the Ar . thaliana or Ar . lyrata genome . Therefore , BrPCLS4 and 5 may have been introduced into the B . rapa genome separately from each other and from PCLS1 ( ILR2 ) after the divergence of the Brassica and Arabidopsis lineages ( Figure 8 ) . Similarly , the detection profile of AtPCLS2 and AlPCLS3 ( Figure 2 ) shows that they may have been introduced into Ar . thaliana and Ar . lyrata chromosomes independently after the separation of 2 plant species ( 3 . 0–5 . 8 million years ago ) ( Figure 8 ) ; these are more recent HGT events than the PCLS1 integration into the Arabidopsis lineage . PCLS integrations into the Solanaceae lineage were slightly complex . Relatively high or moderate levels of aa sequence identities ( 47–68% ) are shared within the PCLS5s from the family Solanaceae . However , a lack of information regarding genome sequences flanking the PCLS5s caused difficulty in determining whether a single event or multiple HGT events may have occurred within the lineage ( Figure 8 ) . Gene sequences related to rhabdovirus or varicosavirus N ( CP ) genes ( RNLSs ) are detected in many genera including Brassica , Raphanus , Mimulus , Nicotiana , Lotus , Malus , Cucumis , Populus , Theobroma , and Aquilegia ( Figures 5 , 8 , S3 ) . Using similar rationale for the HGT of PCLSs , multiple integrations of RNLSs into plant chromosomes are thought to have occurred ( Figure 8 ) . RNLSs are distributed in an irregular manner in the plant lineage , while rhabdovirus N proteins show similar tree topology to that exhibited by corresponding RdRps . This is consistent with the hypothesis that HGT occurred “from virus to plants . ” RNLS2 was detected in a very narrow range of plants , i . e . , detectable only in N . tabacum but not other Nicotiana species ( Figure 5 ) . RNLS1 was detected in all tested Brassica species , R . sativus and Aq . coerulea , while it was not detected in the genomes of Ar . thaliana [25] or Ar . lyrata ( Phytozome ) , which are much closely related to Brassica than Aq . coerulea to Brassica . If these sequences were of plant origin , homologous sequences are expected to be retained at least within some members of the families Brassicaceae and Solanaceae . However , Southern blotting and genomic PCR analyses with NtRNLS2- and BrRNLS1-specific probes and primers failed to detect their related sequences in plants other than N . tabacum , and Brassica species and R . sativus , respectively ( Figure 5C–F ) . A search using NtRNLS2 and BrRNL1 against the genome sequences of Ar . thaliana and Ar . lyrata did not yield any hits . This indicates that multiple HGTs of RNLSs occurred from “virus to plant . ” While the BrRNLS1 integration may have postdated the split of the Arabidopsis and Brassica lineages ( 43 . 2–18 . 5 million years ago ) , NtRNLS2 and NtRNLS3 integration may have occurred after the divergence of N . tabacum ( allotetraploid ) and its maternal parent N . sylvestris ( diploid ) ( 0 . 2 million years ago ) [44] . This hypothesis must be verified by sequence analysis of the corresponding regions of N . tabacum and other Nicotiana species . The detection pattern of PCLSs within the family Brassicaceae provided an interesting insight into the phylogenetic relationship of some genera in the family . The family Brassicaceae is one of the largest families comprising over 300 genera and approximately 3 , 300 species that include an important plant biology model plant , Ar . thaliana , and agriculturally important Brassica species . Their phylogenetic relationships have been extensively studied and are occasionally controversial , because they rely on data sets and methods exploited for analyses . For example , placement of the genus Crucihimalaya is interesting to note in relation to this study . The genus is placed into a clade containing the genus Boechera , and is assumed to have separated from an ancestor common to the genus Capsella after the divergence of the Arabidopsis lineage based on phylogenetic analyses with a single nuclear gene ( chalcone synthase gene ) [45] or multiple data sets containing plastid and nuclear genes [45] , [46] , [47] , [48] . However , utilization of different data sets shows different tree topologies , suggesting that the Crucihimalaya genus may have diverged before the split between Arabidopsis and Capsella [45] , [49] . PCLS1s ( ILR2 homologs ) were detected in relatives of Arabidopsis but not in Cru . lasiocarpa ( Figure 1B , D ) , strongly supporting the phylogenetic relation proposed by Lysak et al . [49] . The absence of the PCLS1 in a homologous position of the Cru . lasiocarpa chromosome was confirmed by sequencing of genomic PCR fragments generated with a specific primer set ( Figure 1D ) . Therefore , these results clearly indicate that PCLSs have the potential to supplement phylogenetic estimates by serving as molecular markers . Furthermore , a similar insight into phylogenetic relations among Nicotiana species may be gained from data regarding 4 PCLSs identified in N . tabacum as more data in the genome and PCLS sequences of the genus Nicotiana become available . Many examples of HGT from minus-sense RNA and dsRNA viruses , particularly from partitiviruses , have been found in plant nuclear genomes . Endogenization of NRVSs required 3 events to occur: ( 1 ) replication of the ancestral viral genome in the germ lines of host plants , ( 2 ) reverse transcription of genomic RNA , and ( 3 ) its subsequent integration into plant chromosomes . Many plant viruses are reported to be transmitted through pollens and seeds [50] , while their transmission rates depended on virus/host combinations . Seed-transmitted viruses include positive-strand and negative-strand RNA viruses and partitiviruses with dsRNA genomes . The family Partitiviridae accommodates members that infect plants or fungi , and some plant and fungal partitiviruses are phylogenetically closely related ( [21]; Figure 4 ) . PCLS1 is most closely related to a novel fungal partitivirus , RnPV2 , but the other PCLSs show the closest resemblance to plant partitiviruses ( Table 1 , Figure 4 ) . Therefore , PCLS1 integration occurred when an ancestor of RnPV2 acquired the ability to infect an ancestral plant during endosymbiotic [51] or parasitic interactions between its host fungus and the plant , a host of the fungus , and to invade the plant germ cells . In support of this hypothesis , an assembled EST sequence is present in F . pratensis that is more closely related to PCLS1 than the RnPV2 CP gene and considered to have originated in a plant partitivirus ( Figure 4 ) . Such a virus may have been a direct source of plant PCLS1 . Alternatively some fungal partitiviruses may be intrinsically able to infect plant cells . The expected capability of plant partitiviruses to replicate in host germ cells may be associated with their high rates ( ∼100% ) of seed transmission via ovule and/or pollen [21] , an uncommon phenomenon for plant viruses . Although germ lines are hypothesized to have the ability to eliminate virus infection , partitivirus may be able to overcome such a host mechanism . It is also likely that ancestral negative-strand RNA viruses may have invaded germ cells of host plants . For the second required event , integration of NRVSs likely involved reverse transcription that may have been mediated by reverse transcriptase encoded by retrotransposons or pararetroviruses . However , the mechanism by which the viral RNA sequences were converted to DNA and introduced into plant genomes remains unknown . Interestingly LjPCLS8 harbors an unrelated sequence of 1 . 3-kb sequence in its central region ( Figure S1A , D ) , suggesting a recombination event of during reverse transcription or a 2-step integration of 2 distinct molecules , PCLS8 and a sequence of an unknown origin . For the third event , as suggested by Liu et al . [24] , transposon-mediated integration [52] and/or double-strand-break repair ( non-homologous recombination ) [53] may be involved . Flanking regions of some plant genome-integrated NRVSs ( e . g . , RNLS1s and CsFRLS1 , see Figures 7 , S3 ) carried transposable elements or multiple repeat sequences , supporting the first type of integration . Vertebrate cultured cells are useful for experimentally monitoring de novo integrations of negative-strand RNA viral sequences [11]; however , the agents that facilitate the reverse transcription and integration steps remain unknown . In contrast to the nuclear integrations of partitivirus CP sequences and negative-strand RNA virus N sequences , plus-strand RNA virus endogenizations were observed much less frequently . A level of viral transcripts in germ cells may be one of factors governing the frequency of NRVSs . This is supported by the observation ( data not shown ) that , whereas we searched for integrated partitiviral RdRp sequences or other non-N sequences of rhabdoviruses , we could seldom detect them . Partitivirus CP and rhabdovirus N coding transcripts are highly likely to be produced in cells infected by the respective viruses more than other viral transcripts . Plus-strand RNA viruses , are believed to accumulate in infected plant cells much more than plant partitiviruses . However , plus-strand RNA viruses may generally be more able to be detected by a surveillance system of host germ cells and/or less competent to escape from their defense system . A smaller number of FRLS integrations observed in this study ( Figure 7 ) may be associated with a lower ability of ancestral plus-strand RNA viruses to invade host germ cells , as predicted from the low seed transmissibility of CLBV [54] . Alternatively , plus-strand RNA virus sequences are disfavored by reverse transcriptase and agents that facilitate integration of their complementary DNA in the second and third events , respectively , although this possibility may be low .
A virus-infected fungal strain of R . necatrix , W57 , was isolated in the Iwate Prefecture , Japan . Molecular characterization of genomic dsRNAs were performed according to the methods described by Chiba et al . [55] , unless otherwise mentioned . Seeds for members of the Brassicaceae family , L . japonicus , Med . truncatula and Cuc . sativus cv . Borszczagowski B10 line were provided by the Arabidopsis Biological Resource Center of The Ohio State University , the Frontier Science Research Center , University of Miyazaki , and Drs . Kazuhiro Toyoda , Douglas Cook , and Grzegorz Bartoszewski , respectively . Seeds for members of the genus Nicotiana were originally obtained from Nihon Tabako , Inc ( Tokyo , Japan ) and maintained at Okayama University . Dr . Takashi Enomoto of Okayama University provided the remaining plants . Plant genomic DNA was isolated from seeds or fresh leaf materials and used in genomic PCR and Southern blot analyses as described by Miura et al . [56] . Sequences of ILR2 homologs ( PCLS1s ) from members of the family Brassicaceae , except for Ar . thaliana accessions Col-0 and WS , and Ar . lyrata , were obtained by sequencing genomic PCR fragments . Genomic PCR fragments or clones were used to determine the sequences of other selected PCLSs , RNLSs and FRLSs . Digoxigenin ( DIG ) -labeled DNA , prepared as described by Chiba et al . [55] , was used as probes in Southern blotting analyses as described by Faruk et al . [57] . Table S3 includes sequences of primers used in this study . BLAST ( tblastn ) searches [58] were conducted against genome sequence databases available from the NCBI ( nucleotide collection , nr/nt; genome survey sequences , GSS; high-throughput genomic sequence , HTGS; whole-genome shotgun reads , WGS; non-human , non-mouse ESTs , est others ) ( http://www . ncbi . nlm . nih . gov/ ) , Phytozome v6 . 0 ( http://www . phytozome . net/ ) , Brassica database ( BRAD ) ( http://brassicadb . org/brad/ ) , Potato Genome Sequencing Consortium ( http://potatogenomics . plantbiology . msu . edu/ ) , and Kazusa DNA Research Institute ( http://www . kazusa . or . jp/e/index . html ) . The databanks covered the complete and partial genome sequences of 20 plant species . Transposable element sequences were identified using the Censor ( http://www . girinst . org/censor/index . php ) [59] . Obtained non-retroviral integrated sequences were translated to amino acid sequences and aligned with MAFFT version 6 under the default parameters [60] ( http://mafft . cbrc . jp/alignment/server ) . For some non-retroviral integrated sequences with interrupted ORFs , frames were restored by adding Ns as unknown sequences to obtain continuous aa sequences ( edited residues are shown as Xs ) . Alignments were edited by using MEGA version 4 . 02 software [61] . To obtain appropriate substitution models for the maximum likelihood ( ML ) analyses , each data set was subjected to the Akaike information criterion ( AIC ) calculated using ProtTest server [62] ( http://darwin . uvigo . es/software/prottest_server . html ) . According to ProtTest results , WAG+I+G+F , LG+I+G , and LG+I+G+F were selected for PCLSs and partitiviruses , for RNLSs , plant rhabdoviruses and varicosaviruses , and for FRLS and flexiviruses , respectively . Phylogenetic trees were generated using the appropriate substitution model in PhyML 3 . 0 [63] ( http://www . atgc-montpellier . fr/phyml/ ) . In each analysis , four categories of rate variation were used . The starting tree was a BIONJ tree and the type of tree improvement was subtree pruning and regrafting ( SPR ) [64] . Branch support was calculated using the approximate likelihood ratio test ( aLRT ) with a Shimodaira–Hasegawa–like ( SH-like ) procedure [65] . The tree was midpoint-rooted using FigTree version 1 . 3 . 1 software ( http://tree . bio . ed . ac . uk/software/ ) . Two mycoviral genome sequences and a total of 73 non-retroviral integrated RNA virus sequences were analyzed . Sequence data ( 1 of the 2 genome segments of RnPV2 , 21 PCLSs , 12 RNLSs and 1 FRLS ) used for phylogenetic analyses in this article have been deposited into the EMBL/GenBank/DDBJ Data Library under the following accession numbers: AB569998 ( RnPV2 dsRNA2 ) , AB576168–AB576175 , AB609326–AB609329 ( ILR2-like sequences: PCLS1s ) , AB609330–AB609338 ( PCLS2–PCLS8 ) , AB9339–AB609350 ( RNLSs ) , and AB610884 ( CsFRLS1 ) ( Tables S4 and S5 ) . Other non-retroviral integrated RNA virus elements whose sequences were partially determined and analyzed in this study are available upon request . | Eukaryotic genomes contain sequences that have originated from DNA viruses and reverse-transcribing viruses , i . e . , retroviruses , pararetroviruses ( DNA viruses ) , and transposons . However , the sequences of non-retroviral RNA viruses , which are unable to convert their genomes to DNA , were until recently considered not to be integrated into eukaryotic nuclear genomes . We present evidence for multiple independent events of horizontal gene transfer from a wide range of RNA viruses , including plus-sense , minus-sense , and double-stranded RNA viruses , into the genomes of distantly related plant lineages . Some non-retroviral integrated RNA viral sequences are conserved across genera within a plant family , whereas others are retained only in a limited number of species in a genus . Integration profiles of non-retroviral integrated RNA viral sequences demonstrate the potential of these sequences to serve as powerful molecular tools for deciphering phylogenetic relationships among related plants . Moreover , this study highlights plants co-opting non-retroviral RNA virus sequences , and provides insights into plant genome evolution and interplay between non-reverse-transcribing RNA viruses and their hosts . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"plant",
"science",
"plant",
"evolution",
"virology",
"plant",
"pathogens",
"plant",
"biology",
"plant",
"pathology",
"biology",
"microbiology",
"viral",
"evolution"
] | 2011 | Widespread Endogenization of Genome Sequences of Non-Retroviral RNA Viruses into Plant Genomes |
The actin family of cytoskeletal proteins is essential to the physiology of virtually all archaea , bacteria , and eukaryotes . While X-ray crystallography and electron microscopy have revealed structural homologies among actin-family proteins , these techniques cannot probe molecular-scale conformational dynamics . Here , we use all-atom molecular dynamic simulations to reveal conserved dynamical behaviors in four prokaryotic actin homologs: MreB , FtsA , ParM , and crenactin . We demonstrate that the majority of the conformational dynamics of prokaryotic actins can be explained by treating the four subdomains as rigid bodies . MreB , ParM , and FtsA monomers exhibited nucleotide-dependent dihedral and opening angles , while crenactin monomer dynamics were nucleotide-independent . We further show that the opening angle of ParM is sensitive to a specific interaction between subdomains . Steered molecular dynamics simulations of MreB , FtsA , and crenactin dimers revealed that changes in subunit dihedral angle lead to intersubunit bending or twist , suggesting a conserved mechanism for regulating filament structure . Taken together , our results provide molecular-scale insights into the nucleotide and polymerization dependencies of the structure of prokaryotic actins , suggesting mechanisms for how these structural features are linked to their diverse functions .
The eukaryotic cytoskeleton , which is critical for many cellular functions such as cargo transport and morphogenesis , is comprised of three major elements: actin , tubulin , and intermediate filaments . These proteins bind nucleotides and form highly dynamic polymers [1] . Each of these proteins has numerous homologs across the bacterial and archaeal kingdoms that dictate cell shape and various intracellular behaviors [1 , 2] . However , relatively little is known about the structural dynamics of these prokaryotic homologs and whether dynamical behaviors are conserved . Among bacterial cytoskeletal proteins , actin homologs are the most structurally and functionally diverse class identified thus far . Although sequence homology to eukaryotic actin is generally low , prokaryotic actins have been identified via X-ray crystallography based on their structural homology to eukaryotic actin [3–6] , which has a U-shaped four-domain substructure , with two beta domains and a nucleotide binding pocket between two alpha domains [7] . Among the actin homologs , one of the best studied is MreB , which forms filaments that coordinate cell-wall synthesis in many rod-shaped bacteria and is essential for maintaining cell shape in these species [8 , 9] . FtsA is an actin homolog with a unique structural domain swap that is essential for anchoring the key cell-division protein and tubulin homolog FtsZ to the membrane [5 , 10] . The actin homolog ParM forms filaments that move R1 plasmids to opposite ends of rod-shaped bacteria prior to cytokinesis [11] . Crenactin forms part of the archaeal cytoskeleton [12]; its biological function is currently unknown , but it is hypothesized to be involved in DNA segregation and/or cell-shape control [12] . Given the common structural features of prokaryotic actins , it is unknown how they exert such a wide variety of functions . Features such as the domain swap in FtsA suggest that some proteins may have the capacity for unique intramonomeric conformational changes [13] . Another possibility is that functional differences emerge at the filament level: a wide variety of double-protofilament bacterial-actin filament structures have been observed [14 , 15] . The extent to which lessons about structure-function relationships are general across the diverse actin family can be informed by understanding commonalities and distinctions in their structural dynamics . While X-ray crystallography and cryo-electron microscopy ( cryo-EM ) have proven critical for elucidating the structures of monomers and filaments of prokaryotic actins , understanding the mechanisms by which these proteins exert their functions , particularly their mechanical roles , requires integration with other experimental and computational techniques . Microscopy has revealed that most actin homologs can form long filaments within cells [3 , 4 , 16–19] . In vitro , ParM filaments exhibit dynamic instability [20] , and all actin homologs except FtsA have been observed to undergo nucleotide hydrolysis [12 , 21 , 22] . However , these experimental techniques lack the spatial and temporal resolution necessary to understand how these filament properties are linked to changes in structure . Various mechanistic models of cytoskeletal function have focused on nucleotide hydrolysis as a key determinant of filament mechanics [23–25] . Understanding how nucleotide hydrolysis and polymerization affect structural transitions in prokaryotic actins requires a method that can interrogate molecular behaviors with atomic resolution . All-atom molecular dynamics ( MD ) simulations have been successfully employed to probe the effects of perturbations on prokaryotic and eukaryotic cytoskeletal proteins . MD simulations of eukaryotic actin monomers have uncovered nucleotide-dependent changes in the structure of the nucleotide-binding pocket [26] , and simulations of actin filaments showed nucleotide-dependent changes to filament bending [27] . MD simulations predicted that GTP hydrolysis of the tubulin homolog FtsZ can result in substantial polymer bending [28] , which was subsequently verified through X-ray crystallography [29] . MD simulations of MreB and FtsA filaments also revealed intra- and inter-subunit changes with important implications for their respective cellular functions [13 , 17] . In sum , structural changes to cytoskeletal filaments are generally observable within the time frame accessible to MD simulations , potentiating a systematic survey of general and specific connections among bound nucleotide , polymerization , and subunit conformations across the prokaryotic actin family . Here , we used MD simulations to probe the conformational dynamics of monomers and filaments of MreB , FtsA , ParM , and crenactin ( Fig 1 ) . We found that these proteins exhibit a wide range of intrasubunit motions that are generally well described by the centers-of-mass of their four subdomains , and hence the majority of monomer dynamics can be explained by changes in opening and dihedral angles formed by the subdomain centers . Our results predict that some proteins exhibit strong dependence on the bound nucleotide , while others are unaffected by hydrolysis . In ParM , opening is inhibited by interactions between two subdomains . As with MreB , changes in the dihedral angle of FtsA and crenactin subunits generally impact the bending or twisting of polymers . This work provides insight into how molecular-scale perturbations of these proteins contribute to their diverse roles in cell-shape regulation and intracellular organization across bacteria and archaea .
In a previous study , we performed all-atom MD simulations on unconstrained MreB monomers using CHARMM27 force fields and found that ATP-bound monomers had larger opening and dihedral angle than ADP-bound monomers [13] . For our study of prokaryotic actins , we first sought to interrogate the robustness of these findings with respect to the force field used and the dimensional reduction to the centers-of-mass of subdomains IA , IB , IIA , and IIB of actin family members . While simulations using different force fields mostly preserve large-scale motions , distinct behaviors can emerge at finer levels of detail [30] . Thus , we performed all-atom MD simulations on Thermatoga maritima MreB ( PDB ID: 1JCG ) [4] using CHARMM36 force fields [31] . As done previously for actin [32] and MreB [28] , we quantified conformational changes by calculating two opening angles ( Fig 2A ) and a dihedral angle ( Fig 2C ) from the center of mass of each of the four subdomains ( Methods ) . We carried out all simulations until the opening and dihedral angles adopted distributions that were well fit by Gaussians ( we define this state as “equilibrated” ) , and compared mean values across independent replicate simulations . MreB monomers reached stability within ~55 ns of simulation in our previous study using CHARMM27 force fields [13] , and we observed similar time scales using CHARMM36 . We extended one MreB-ATP simulation to 80 ns , and the opening angle remained similar ( S1A Fig ) , suggesting the open conformation of MreB is stable on our simulation time scale . While the opening angle was 5–10° smaller with CHARMM36 than with CHARMM27 [13] ( Figs 2B and S1A ) , in both sets of simulations subdomains IB and IIB of ATP-bound monomers rapidly hinged apart to form stable , open conformations . Additionally , using CHARMM36 , the opening angle stabilized at smaller angles for ADP- than ATP-bound MreB ( S1B Fig ) , as expected from our previous study [13] . ATP-bound MreB monomers also adopted a larger dihedral angle than ADP-bound monomers using CHARMM36 , similar to CHARMM27 ( Fig 2C and 2D , S1C Fig ) . Thus , despite small differences , a similar nucleotide dependence in the conformation of MreB monomers was observed using both CHARMM27 and CHARMM36 force fields , supporting our use of CHARMM36 going forward . While previous studies used the centers-of-mass of the four subdomains of actin-family proteins to dramatically reduce the dimensionality of the protein structure [4 , 13 , 32] , it is also possible for conformational changes to arise within subdomains in addition to the hinges between them . To distinguish between these scenarios , we calculated the root mean square deviation ( RMSD ) of the Cα atoms from the energetically minimized structure for each subdomain separately , and also for the entire protein , at each time point of our simulations . In the CHARMM27 ATP-bound simulation , the RMSD of the entire protein increased past 5 Å as the opening angle increased . However , the RMSD of each subdomain remained at ~2 Å ( S1D Fig ) , suggesting that most conformational changes were inter-subdomain . Unsurprisingly , since the CHARMM36 simulation adopted a smaller opening angle than the CHARMM27 simulation , the RMSD of the protein was smaller as well ( Fig 2E ) . Nonetheless , consistent with the CHARMM27 simulation , the RMSD of each subdomain was smaller than the RMSD of the whole protein ( Fig 2E ) . To determine whether subdomain structure was consistent between distinct MreB monomer conformations , we computed RMSDs between the CHARMM36 equilibrium structure and the CHARMM27 simulation at each time point . Since the CHARMM27 simulation adopted a larger opening angle than the CHARMM36 simulation , the RMSD of the whole protein increased relative to the CHARMM36 equilibrium structure . Still , the subdomain RMSDs remained at ~2 Å ( S1E Fig ) . Thus , the structure of each subdomain was largely maintained as the whole protein underwent large conformational changes . To further quantify the stability of the MreB subdomains , we performed a principal component analysis ( PCA ) on the simulation trajectory of an ATP-bound MreB . The eigenvector with the largest eigenvalue explained 70 . 3% of the variance in the whole protein . We then performed PCA on each of the four subdomains separately , and found that the eigenvector with the largest eigenvalue explained only 25 . 5% to 45 . 9% of the variance . These results indicate that the trajectory of the protein can be largely explained through global , systematic changes , while the subdomains exhibit more randomness in their motion ( S1F Fig ) . We perturbed the mean MreB structure along the first eigenvector , and identified that movement along the first eigenvector corresponds to changes in the opening angle ( S1G Fig ) . To complement these findings , we performed a cross-correlation analysis on MreB using the Bio3D package [33] , and showed that subdomains IA/IB exhibited highly correlated , collective motions . The same was true for subdomains IIA/IIB , and the two groupings ( IA/IB and IIA/IIB ) had opposing movements . These results support our conclusion that the intramonomeric dynamics of MreB can mostly be represented by the relative motions of its subdomains ( S1H Fig ) . Simulations at longer time scales have the potential to reveal dynamics within MreB subdomains and increase our understanding of how intra-subdomain changes fine-tune MreB functions . We next investigated FtsA ( PDB ID: 4A2B ) , an essential protein involved in tethering the key division protein FtsZ to the membrane [5 , 10] . FtsA has a four-subdomain architecture similar to those of actin and MreB , but subdomain IB is replaced by a new subdomain ( IC ) located on the opposite side of subdomain IA ( Fig 1 ) that has no structural similarity to the actin subdomains [5] . To determine whether this domain swap impacts the conformational dynamics around the nucleotide-binding pocket and alters the coupling of dihedral/opening angles to nucleotide hydrolysis , we first carried out all-atom unconstrained MD simulations on ATP- and ADP-bound FtsA monomers . While FtsA monomers showed little conformational flexibility , they still exhibited distinct ATP- and ADP-bound states with respect to opening and dihedral angles ( Fig 3A and 3B , Methods ) . In all simulations , the RMSD of each subdomain as well as the entire protein remained <2 Å ( S2 Fig ) , and the opening angle exhibited very little variation . Compared to an ATP-bound MreB monomer , whose opening angle reached a different equilibrium value ( 102 . 1±2 . 4° and 93 . 2±1 . 0° , mean ± standard deviation ( s . d . ) measured over the final 30 ns of simulation ) in replicate simulations , the opening angle of an ATP-bound FtsA monomer was much more constrained and was highly reproducible ( 110 . 1±0 . 7° and 109 . 6±0 . 8° in two replicates; Fig 3A , 3C and 3D ) . The FtsA equilibrium opening angle exhibited slight , but highly reproducible , nucleotide dependence: the opening angle for ADP-bound FtsA equilibrated at 112 . 4±1 . 0º and 111 . 5±0 . 7° . In ATP- and ADP-bound FtsA , the dihedral angle equilibrated at 20 . 6±1 . 9° for ATP and 20 . 3±2 . 5° for ADP , respectively ( Fig 3B , 3E and 3F ) , with a highly reproducible mean value across simulations ( Fig 3F ) . To address whether the limited duration of our simulations ( ~200 ns ) precluded access to other states of FtsA , we steered the opening angle of an equilibrated ATP-bound monomer up to the equilibrium value in our simulations of an ADP-bound monomer . This conformation was unstable; the opening angle rapidly decreased back to the value in our simulations of an ATP-bound monomer ( S3A Fig ) . Similarly , steering the opening angle of an equilibrated ADP-bound monomer down to the equilibrium value in our simulations of an ATP-bound monomer was followed by rapid re-opening ( S3B Fig ) . Longer time-scale simulations will help to quantify the stability of the ATP- and ADP-bound conformations . Nonetheless , as with MreB and actin , FtsA likely has two distinct states dependent on the bound nucleotide that are stable on the time scale of our simulations . We next used all-atom MD simulations to investigate ParM , which forms filaments that push apart plasmids to segregate them into daughter cells [6 , 18] . ParM monomers exhibited large , nucleotide-dependent conformational changes , with substantial variability across replicate simulations , suggesting the possibility of multiple conformational states rather than a single equilibrium state . In all simulations of ATP-bound ParM , the opening angle rapidly increased from 97° in the crystal structure to over 100° ( Fig 4A ) . In one simulation ( ParM-ATP-1 ) , subdomains IB and IIB continued to hinge apart to 109 . 0±2 . 0° after 100 ns , then reverted back to 103 . 3±2 . 9° in the last 30 ns of the 200-ns simulation ( S4A Fig ) . While this simulation appeared not to converge to a single open state , it potentially revealed the presence of two distinct states of ATP-bound ParM , both with opening angles larger than the crystal structure . In the other two simulations , the opening angle stabilized at 101 . 6±1 . 3° and 102 . 2±1 . 7° , but did not further open up . ADP-bound monomers were less open , equilibrating between 97° and 99° ( Fig 4A ) . Unlike MreB , we did not observe consistent nucleotide dependencies on the dihedral angle of ParM monomers ( S4B Fig ) . To identify whether certain intrasubunit motions of ParM contributed to the transient period of opening angle >105° in our first ATP-bound simulation , we calculated the RMSD of each subdomain and the whole protein relative to the minimized structure in that simulation . Subdomains IB and IIB exhibited large conformational variability , similar to the protein as a whole ( Fig 4B ) . We identified residues 35–45 and residues 58–67 on subdomain IB and residues 211–216 on subdomain IIB as having the greatest root mean square fluctuation ( RMSF ) ( Fig 4C ) , a measure of the positional variability of specific residues . The subdomain RMSDs calculated after excluding these high-RMSF residues decreased to <2 Å , suggesting a stable core within each subdomain of ParM ( Fig 4D ) . We re-measured opening and dihedral angles excluding these high-RMSF residues , and found that while the overall values changed , the same nucleotide dependencies relating to dihedral and opening angle were observed ( S5 Fig ) . The high degree of variability in opening angle across replicate simulations suggested the opportunity to identify the structural elements that underlie ParM opening . In the crystal structure , the high RMSF loop of residues 58–67 interacts strongly ( defined as a Cα-Cα distance <5 Å ) with residues 173–174 , which lie near the ATP binding pocket , as well as with residues 200–202 ( Fig 4E ) . In the ParM-ATP-1 simulation with the most variable opening angle , these interactions were largely abolished within 40 ns , but eventually re-established after ~150 ns ( Fig 4E ) . The presence and absence of these interactions largely coincided with the large shifts in opening angle ( S4A Fig ) . Throughout this simulation , the opening angle was highly correlated with the distance between the center of mass of residues 65–67 and the center of mass of 173–174 ( Fig 4F ) . By contrast , in the other two ParM-ATP simulations with smaller opening angles , the interaction between residues 58–67 and 173–174 persisted throughout the simulation ( S6A and S6B Fig ) . In the ParM-ATP-2 simulation , the interaction between residues 58–67 and 173–174 was initially disrupted but quickly recovered ( S6A Fig ) , consistent with the smaller increase in opening angle in this simulation ( S4A Fig ) . To determine whether disrupting the interaction between residues 173–174 and 58–67 would cause ParM to open , we steered the center-of-mass distance between residues 173–174 and 65–67 from the crystal structure value of 9 . 3 Å to various larger values . In a steered simulation in which we steered the distance between residues 173–174 and 65–67 to 19 . 3 Å , the opening angle increased to 104 . 0±1 . 4° ( Fig 4G ) , suggesting that breaking this interaction directly changes the ParM protein conformation . Steering the distance between residues 173–174 and 65–67 ( Fig 4G ) to 19 . 0 Å and 14 . 0 Å resulted in opening angles of 101 . 9±2 . 0° and 98 . 6±1 . 0° , respectively , indicating that the distance between residues 173–174 and 65–67 tunes the opening angle of ParM monomers . Longer time-scale simulations , along with more replicates , will be necessary to quantify the frequency of opening-angle transitions , and to determine whether they are always coupled to interactions between the 65–67 and 173–174 residue groups . The dihedral angles of ParM in a monomer crystal structure [6] and in a cryo-EM filament structure [18] were 26 . 7° and 7 . 54° , respectively ( Fig 4H ) , suggesting that polymerization impacts ParM conformations . ParM forms left-handed double-helical filaments that make MD simulations infeasible due to the large number of subunits required to mimic a biologically relevant system . To overcome this challenge and to glean information about whether ParM filaments flatten upon polymerization , we steered the dihedral angle of an ATP-bound ParM monomer to 7° to match that of the cryo-EM filament structure . Upon release , the monomer rapidly unflattened to 20° ( Fig 4I ) , similar to the stabilized values of our ParM-ATP monomer simulations ( S4B Fig ) , suggesting that ParM monomers , like MreB [13] , flatten upon polymerization . Thus , ParM likely has some similar conformational properties as MreB , even though the interactions between the flexible regions of subdomains IB and IIB unique to ParM provide tunability to its opening angle . For MreB , we previously discovered that the dihedral angle of the bottom subunit in a dimer simulation was directly coupled to dimer bending [13] . In particular , the intersubunit bending of ATP-bound MreB was correlated to the dihedral angle throughout each simulation , and steering the dihedral angle to a flatter conformation reduced the bending of a dimer structure [13] . We confirmed these findings for the CHARMM36 force field by steering the dihedral angle of the bottom subunit of an MreB-ATP dimer to 23 . 1° , 28 . 3° , and 33 . 0° , and observed the expected inverse relationship between dihedral angle and filament bending ( S7 Fig ) . Given the similarities between the dynamics of MreB and other bacterial actin homologs at the monomeric level , we hypothesized that other actin-like filaments also exhibit intersubunit behaviors coupled to intrasubunit changes . We performed MD simulations of dimers of FtsA ( PDB ID: 4A2B ) and Pyrobaculum calidifontis crenactin ( PDB ID: 4CJ7 ) ; crenactin is an archaeal actin homolog for which our MD simulations of ATP- and ADP-bound monomers exhibited similar conformations ( S8A and S8B Fig ) . Dimer structures were initialized from repeated subunits of the appropriate crystal structure . Due to ParM’s complicated filament structure , which requires four points of contact per monomer , we were unable to construct biologically relevant ParM dimers with a stable interface in silico [34] . For each time step of dimer simulations , we measured two bending angles and one twisting angle between the subunits ( Fig 5A and 5D; Methods ) . We did not observe any significant nucleotide-dependent changes in bending or twisting angles for FtsA and crenactin dimers ( S9 Fig ) , likely because there was little or no nucleotide dependence in monomer conformations of FtsA ( Fig 3 ) and crenactin ( S8 Fig ) , respectively , although it is possible that differences could emerge at longer time scales . Similar to MreB , the dihedral angle of the bottom subunit of an FtsA dimer was correlated with filament bending along the second bending axis ( Fig 5A and 5B ) . To test whether coupling between the dihedral angle and filament bending was direct , we steered the dihedral angle of the bottom subunit to average values of 16 . 4° , 20 . 8° , 25 . 5° , and 29 . 6° ( measured over the last 20 ns of steered simulations; S10A Fig ) . The resulting bending angles of the dimer shifted systematically with the dihedral angle ( Fig 5C ) , indicating that subunit dihedral changes drive bending of the FtsA filament . Interestingly , the bending angle flipped from positive to negative ( Fig 5C ) ; this flexibility could play a key role in regulating the transition of the division machinery from assembly to constriction . In the crenactin filament crystal structure ( PDB: 4CJ7 ) , subunits have a large twisting angle of -47 . 3° ( negative indicates a right-handed filament ) . In our simulations , ATP- and ADP-bound dimers equilibrated between -45° and -53° ( S9D Fig ) , suggesting that the large twisting angle is not a result of strained crystal contacts . By contrast to MreB and FtsA , the dihedral angle of the bottom subunit of crenactin was not correlated with filament bending , but rather with filament twist ( Fig 5D and 5E ) . To test causality , we steered the dihedral angle of the bottom subunit to 22 . 6° , 23 . 4° , and 26 . 7° ( S10B Fig ) , and observed progressive increases in twist magnitude ( Fig 5F ) . In sum , coupling of filament degrees of freedom to subunit conformational changes is generalizable across at least some bacterial actin-family members .
Through all-atom MD simulations of four actin-family proteins , we identified both conserved and specific dynamical behaviors across the actin family . First , we confirmed that the dihedral and opening angles between the centers-of-mass of the four subdomains represent the majority of conformational changes . In all simulated prokaryotic actins , the four subdomains exhibited high stability throughout the simulation , even as the whole protein changed conformation ( Figs 2E , 4B , S2 and S8C ) . This analysis supports the model used by previous MD studies that measured dihedral and opening angles of actins [4 , 13 , 32] , and provides a verified metric for future MD simulations of actin-family proteins . Based on our findings , we propose a general model of the regulation of the structure of actin-family filaments in which the intra-subunit dihedral or opening angle of an actin monomer regulates filament bending and/or twisting angles . The model suggests a mechanistic explanation for previous experimental results that have revealed variable filament structures for actin homologs . Electron microscopy of MreB , for instance , revealed straight filaments and arc-like filaments [6 , 8] , and cryo-EM of crenactin filaments showed highly variable twists ranging from 32° to 56° [35] . Our simulations suggest that changes to bound-nucleotide state explain some of the variability in bend and twist for these dimers by tuning the dihedral angles of each subunit , highlighting a conserved mechanism by which actin homologs can adopt a range of filament conformations . Additionally , our finding that dihedral angle changes drive bending in FtsA and MreB but twisting in crenactin ( Figs 5 and S7 ) indicate that the mechanism is not a trivial mechanical consequence of the four-subdomain structure of actin homologs . Instead , the coupling between dihedral angle and key filament angles has likely been tuned for alternative filament behaviors over evolutionary time scales . For actin , previous studies have shown that its double-protofilament , helical structure leads to twist-bend coupling , as the double-protofilament interface constrains the allowed conformational space [36] . Similar behaviors are likely in the recently identified double-protofilament conformation of Caulobacter crescentus MreB [15] , for which our simulations suggest that the nucleotide-dependent bending in single protofilaments would be translated to nucleotide-dependent twisting of double protofilaments due to conformational constraints . For filaments such as crenactin that exhibit substantial twisting , the twisting can couple strongly to bending through membrane binding [37] , through balancing between the favorable energies of membrane binding and the required energy cost of untwisting the polymer along the membrane surface to expose the membrane-binding interface . Filament bending and twisting can thus be regulated by many interacting factors , calling for simulations of larger biomolecular systems and longer simulation times . Longer and more replicate simulations can also provide more information about the degree to which states truly represent an equilibrium , as opposed to transitions between multiple states . In cases where simulations are currently prohibitive for all-atom MD simulations , course-grained models similar to those previously utilized for actin may prove informative [38] . We observed distinct behaviors across actin homologs in terms of nucleotide dependence that may provide insight regarding the biochemical activities of each protein . MreB and ParM monomers exhibited distinct nucleotide-dependent states ( Figs 2A–2D and 4A ) . These monomers have been shown to have ATPase activity [21 , 39] , suggesting that structural changes occur during the hydrolysis of ATP . Our results are also synergistic with efforts to translate the conformational variability of bacterial actin homologs for engineered purposes , including using ParM as a biosensor for ADP [40] . Numerous studies have attempted to detect ATPase activity in FtsA , but have found little or no activity [22 , 41 , 42] . Our simulations visualized distinct and reproducible nucleotide-dependent states ( Fig 3 ) , albeit with smaller differences than MreB or ParM . Similar to our previous observation that the bending axis of an FtsA dimer rapidly changes upon release from crystal contacts [17] , there is likely flexibility in the conformation of FtsA subunits that is masked in X-ray crystallography by symmetry requirements . For crenactin , we did not observe nucleotide dependence in monomer conformation in our simulations , all of which were carried out at 37°C ( S8 Fig ) . Crenactin has little ATPase activity at 37°C , with maximum ATPase activity at 90°C , which is far outside the temperature range for simulations with CHARMM force fields [12] . Thus , it remains to be seen whether crenactin behaves more like MreB/ParM or FtsA in its native environmental conditions of thermophilic temperatures . Hsp70 , which forms a superfamily with actin based on a common fold , also exhibits nucleotide-dependent allostery [43] , indicating that these intramonomeric changes may be general to a larger group of proteins . This basis for the large intramonomeric conformational changes in proteins such as MreB and ParM also suggests a strategy for the future design of proteins with similar flexibility and for the design of antibiotics that inhibit or disrupt these motions . For prokaryotic actins , small perturbations in the protein’s environment can vastly impact structure . Many prokaryotic actins require binding proteins to confer their function in vivo , such as RodZ binding to MreB [44 , 45] . Furthermore , simulations of FtsA-FtsZ complexes could reveal why cell division relies upon the correct ratio of FtsA and FtsZ [46] . Crystal structures of FtsA-FtsZ complexes exist , but as we have shown with FtsA , crystal structures do not necessarily capture the relevant physiological state [5] , motivating the use of complementary techniques such as MD . In addition , genetic perturbations to prokaryotic actins can significantly impact cellular phenotypes . For example , mutations in MreB can have large effects on cell size and shape as well as MreB’s ability to sense curvature [47 , 48] . Certain ParM mutations restrict the formation of helical filaments [49] , and a variety of FtsA mutations restore viability after zipA deletion and alter cell shape [50–52] . Ultimately , crystallography , cryo-EM , in vivo light microscopy , and MD should prove a powerful combination for understanding and exploiting the numerous functions of cytoskeletal proteins .
All simulations ( S1 Table ) were performed using the molecular dynamics package NAMD v . 2 . 10 [53] with the CHARMM36 force field [31] , except where otherwise noted , including CMAP corrections [54] . Water molecules were described with the TIP3P model [55] . Long-range electrostatic forces were evaluated by means of the particle-mesh Ewald summation approach with a grid spacing of <1 Å . An integration time step of 2 fs was used [56] . Bonded terms and short-range , non-bonded terms were evaluated every time step , and long-range electrostatics were evaluated every other time step . Constant temperature ( T = 310 K ) was maintained using Langevin dynamics [57] , with a damping coefficient of 1 . 0 ps−1 . A constant pressure of 1 atm was enforced using the Langevin piston algorithm [58] with a decay period of 200 fs and a time constant of 50 fs . Setup , analysis , and rendering of the simulation systems were performed with the software VMD v . 1 . 9 . 2 [59] . Steering of the dihedral angle and of distances between residues was achieved by introducing collective forces to constrain angles and distances to defined values through the collective variable functionality of NAMD [53] . MD simulations performed in this study are described in S1 Table . For simulated systems initialized from a MreB crystal structure , the crystallographic structure of T . maratima MreB bound to AMP-PMP ( PDB ID: 1JCG ) [4] was used; for FtsA , the crystallographic structure of T . maratima FtsA bound to ATP gamma A ( PDB ID: 4A2B ) [5] was used; for ParM , the crystallographic structure of E . coli ParM ( PDB ID: 1MWM ) [6] bound to ADP was used; for crenactin , the crystallographic structure of P . calidifontis crenactin bound to ADP ( PDB ID: 4CJ7 ) [12] was used . The bound nucleotide was replaced by both ATP and ADP for all simulated systems , and Mg2+-chelating ions were added for stability . Water and neutralizing ions were added around each monomer or dimer , resulting in final simulation sizes of up to 157 , 000 atoms . All unconstrained simulations were run for 58–220 ns . All steered simulations were run until equilibrium was reached . For mean values and distributions of measurements , unless otherwise noted , only the last 30 ns of unconstrained simulations or the last 20 ns of steered simulations were used . To test whether simulations had potentially reached equilibrium , measurement distributions were fit to a Gaussian , and mean values were compared across replicates . The centers-of-mass of the four subdomains of each protein were obtained using VMD . For each time step , we calculated one opening angle from the dot product between the vector defined by the centers-of-mass of subdomains IIA and IIB and the vector defined by the centers-of-mass of subdomains IIA and IA . Similarly , we calculated a second opening angle from the dot products between the vectors defined by the centers-of-mass of subdomains IA and IB ( or IA and IC for FtsA ) and of subdomains IA and IIA . The opening angles we report are the average of these two opening angles . The dihedral angle was defined as the angle between the vector normal to a plane defined by subdomains IA , IB , and IIA and the vector normal to a plane defined by subdomains IIB , IIA , and IA . Subdomain definitions were obtained by mapping the boundaries based on a structure-based sequence alignment with eukaryotic actin , and are provided for each protein in S2 Table . At each time step of a dimer simulation , the coordinate system of the bottom and top monomers was defined using three unit vectors {d1 , d2 , d3} . d1 approximately aligns to the center-of-mass between the two subunits , and bending around d3 is defined to be zero at the start of the simulation . Rotation around d1 represents twist between the bottom and top subunits . Since d3 is defined to be zero at the start of the simulation , d2 represents the ideal bending axis . d3 represents bending in a direction orthogonal to d2 . Linear regressions were performed using the LinearModel class in Matlab . The reported p-values of linear regressions are for the F-statistic , where the null hypothesis is a zero coefficient of regression . To take into account typical correlation time scales , p-values were adjusted to represent a sample size corresponding to a 1-ns or 3-ns interval between independent states in the simulation [60] . | Simulations are a critical tool for uncovering the molecular mechanisms underlying biological form and function . Here , we use molecular-dynamics simulations to identify common and specific dynamical behaviors in four prokaryotic homologs of actin , a cytoskeletal protein that plays important roles in cellular structure and division in eukaryotes . The four actin homologs have diverse functions including cell division , cell shape maintenance , and DNA segmentation . Dihedral angles and opening angles in monomers of bacterial MreB , FtsA , and ParM were all sensitive to whether the subunit was bound to ATP or ADP , unlike in the archaeal homolog crenactin . In simulations of MreB , FtsA , and crenactin dimers , changes in subunit dihedral angle led to bending or twisting in filaments of these proteins , suggesting a mechanism for regulating the properties of large filaments . Taken together , our simulations set the stage for understanding and exploiting structure-function relationships of prokaryotic cytoskeletons . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"crystal",
"structure",
"monomers",
"condensed",
"matter",
"physics",
"geometry",
"mathematics",
"materials",
"science",
"protein",
"structure",
"crystallography",
"oligomers",
"polymer",
"chemistry",
"contractile",
"proteins",
"actins",
"solid",
"state",
"physics",
"proteins",
"dimers",
"chemistry",
"biophysics",
"molecular",
"biology",
"dihedral",
"angles",
"physics",
"biochemistry",
"biochemical",
"simulations",
"cytoskeletal",
"proteins",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"materials",
"macromolecular",
"structure",
"analysis",
"biophysical",
"simulations"
] | 2019 | Conservation of conformational dynamics across prokaryotic actins |
Cutaneous leishmaniasis is caused by several Leishmania species that are associated with variable outcomes before and after therapy . Optimal treatment decision is based on an accurate identification of the infecting species but current methods to type Leishmania isolates are relatively complex and/or slow . Therefore , the initial treatment decision is generally presumptive , the infecting species being suspected on epidemiological and clinical grounds . A simple method to type cultured isolates would facilitate disease management . We analyzed MALDI-TOF spectra of promastigote pellets from 46 strains cultured in monophasic medium , including 20 short-term cultured isolates from French travelers ( 19 with CL , 1 with VL ) . As per routine procedure , clinical isolates were analyzed in parallel with Multilocus Sequence Typing ( MLST ) at the National Reference Center for Leishmania . Automatic dendrogram analysis generated a classification of isolates consistent with reference determination of species based on MLST or hsp70 sequencing . A minute analysis of spectra based on a very simple , database-independent analysis of spectra based on the algorithm showed that the mutually exclusive presence of two pairs of peaks discriminated isolates considered by reference methods to belong either to the Viannia or Leishmania subgenus , and that within each subgenus presence or absence of a few peaks allowed discrimination to species complexes level . Analysis of cultured Leishmania isolates using mass spectrometry allows a rapid and simple classification to the species complex level consistent with reference methods , a potentially useful method to guide treatment decision in patients with cutaneous leishmaniasis .
Cutaneous leishmaniasis ( CL ) affects 1 . 5 million patients each year and displays a wide spectrum of clinical forms from small self-resolving papules to severe destructive mucosal lesions . The infecting Leishmania species influence the clinical presentation of CL [1] but lesion characteristics are not specific enough for a robust species determination in a given patient [2]–[4] . While 2 species of the Leishmania subgenus - L . major and L . mexicana - are associated with frequent spontaneous cure within a few months [3] , the 2 main species of the Viannia subgenus – Leishmania braziliensis and L . panamensis/guyanensis are associated with a 1–15% risk of delayed mucosal metastasis [5] . Considering the variable severity of CL , recent guidelines recommend using local therapy whenever possible and systemic therapy if local therapy fails or cannot be performed [3] , [6] , [7] . This step-wise decision process integrates not only lesion number and size , patients status ( age and co-morbidities ) , but also the infecting species [8] . The influence of the infecting Leishmania species on treatment outcome is well established [4] , [9] , [10] . Thus , species identification is important to determine the clinical prognosis and to select the most appropriate therapeutic regimen . In current clinical practice , treatment decision is generally presumptive , the infecting species being suspected on epidemiological and clinical grounds [3] but this approach requires a specific clinical expertise and frequently updated knowledge of the geographic distribution of Leishmania species [3] . A simple , rapid method to type cultured isolates would facilitate an easier and more robust treatment decision based on confirmed species identification . Available methods to type Leishmania in cultured isolates or directly in lesions are still complex and poorly standardized . At present , isolation of the parasite in culture is necessary for identification by multilocus enzyme electrophoresis ( MLEE ) , which has long been the reference for Leishmania species identification [11] [12] [13] . Only a few specialized centers currently perform MLEE , the result of which is available several weeks after the isolation of the parasite in culture . These difficulties have led to the development of molecular methods for species identification , generally based on DNA amplification by PCR , followed by single or multilocus sequencing ( MLST ) or restriction fragment length polymorphism analysis [14] or single strand conformation polymorphism or sequencing of different targets including the heat shock protein 70 ( hsp70 ) gene [15] , [16] . Some of these methods can be applied directly to biological samples avoiding the parasite culture step [17] . However , these molecular methods lack inter-laboratories standardization and require the use of expensive reagents . Matrix-assisted laser desorption ionization–time-of-flight ( MALDI-TOF ) mass spectrometry ( MS ) has emerged as a powerful tool for the identification of microorganisms . Using MALDI-TOF MS , the protein spectral “fingerprint” of an isolate is compared to a reference spectral database , yielding results within 1 hour [18] . Although spectrometers are still relatively expensive , the initial investment is justified by a broad use spanning a wide diversity of microbiological samples [19] , and the cost of reagents is very limited . This approach has been applied with success to bacteria , yeasts and filamentous fungi , but to our knowledge , no study on direct identification of protozoans has been published yet [20]–[23] . We have investigated the value of mass spectrometry MALDI-TOF for the identification of Leishmania species in patients with CL .
From 2011 through 2013 , data and samples were collected each time treatment advice was sought from an expert at our hospital for patients with CL . Diagnosis procedures were not modified by the process , expert treatment advice was part of normal medical care , data and sample collection was in the context of national health surveillance . Patients were informed of the process by their attending physician using a procedure common to all French National Reference Centers ( NRC ) ( http://www . parasitologie . univ-montp1 . fr/doc/Declaration_pub_2011 . pdf ) . Data were obtained through the standard reporting form of the NRCL . This form is available online and is anonymous and the anonymisation process is irreversible . The following characteristics are provided on the form: age ( children defined as <16 years ) , sex , clinical form , and for CL or MCL: number of lesions , the presumed place of infection . The collection of parasite isolates was performed in the context of this surveillance program . Parasitological diagnosis was performed and analyzed as previously described by lesion scraping , biopsy or aspirate followed by direct examination of Giemsa-stained smears , histological analysis of HES- or Giemsa-stained tissue sections , culture or PCR [24] . To increase the robustness of the analysis , 10 New World isolates were obtained from the Tropical institute in Antwerpen ( ITM , Belgium ) , all strains were re-suspended in 10% glycerol and stored in liquid nitrogen until use . Needle aspirate of skin lesions was performed under local anesthesia then cultured in both Nicolle-McNeal-Novy ( NNN ) medium and Schneider medium supplemented with 20% fetal calf serum , penicillin and streptomycin [25] . Cultures were kept at 25°C and observed under an inverted microscope in search for motile promastigotes , twice a week for 1 month . Each week positive culture was expanded in 20 mL glass bottles with Schneider medium ( 20% fetal calf serum , penicillin and streptomycin ) , one part was frozen in liquid nitrogen −80° , and the other was used for species identification . Aliquots were thawed immediately at 37° , re-suspended in 5 ml of Schneider medium ( 20% FCS , penicillin streptomycin ) and incubated à 27°C . Subcultures were counted daily and analysed at the end of 3-day growth period ( growth period being defined à> = ×3 fold multiplication over 3 day ) , a growth curve was established for each isolate to perform the proteomic analysis during the exponential growth or early stationary phase . For Leishmania strains isolated only in NNN medium , promastigotes were concentrated by centrifugation ( 2500 rpm×10 minutes ) and resuspended in 20% HS Schneider medium 24–72 hours before proteomic analysis . Positive cultures were sent to the NRCL for confirmation and species identification using a multilocus sequence typing ( MLST ) approach based on the analysis of seven single copy coding DNA sequences [26] . Isolates from ITM had already been typed by hsp 70 sequencing [15] , [16] . Promastigote suspensions from the expanded cultures were centrifuged 3000 g for 3 minutes and the supernatant removed before the pellet was washed twice in pure water , the pellet was then re-suspended in 300 µL of pure water before adding 900 µL of ethanol . After another round of centrifugation , 10 µL of 70% formic acid and 10 µL pure acetonitrile were added to the residual pellet and the subsequent solution was repeatedly and thoroughly vortexed before a final centrifugation . Each centrifugation step was performed at 10 000 g for 2 min at room temperature . The supernatant was distributed ( 0 . 5 µl droplet ) in duplicates on a MALDI ground steel sample slide ( Bruker-Daltonics , Bremen , Germany ) then air-dried . The α-cyano-4-hydroxy-cinnamic acid ( CHCA ) matrix ( Bruker-Daltonics ) , prepared at a concentration of 50 mg/ml in 50% acetonitrile and 50% water with 2 . 5% TFA , was sonicated for 5 min before being spotted ( 0 . 5 µl ) over the dried sample . A DH5-alpha Escherichia coli protein extract ( Bruker-Daltonics ) was deposited on the calibration spot of the sample slide for external calibration . MALDI analysis were performed on a BrukerAutoflex I MALDI TOF mass spectrometer with a nitrogen laser ( 337 nm ) operating in linear mode with delayed extraction ( 260 ns ) at 20 kV accelerating voltage . Each spectrum was automatically collected in the positive ion mode as an average of 500 laser shots ( 50 laser shots at 10 different spot positions ) . Laser energy was set just above the threshold for ion production . A mass range between 3 , 000 and 20 , 000 m/z ( ratio mass/charge ) was selected to collect the signals with the AutoXecute tool of flexcontrol acquisition software ( Version 2 . 4; Bruker-Daltonics ) . Only peaks with a signal/noise ratio >3 were considered . Spectra were eligible for further analysis when the peaks had a resolution better than 600 . For each cultivation condition , we collected mass spectra from 2 biological replicates and 4 technical replicates . Data were processed with Biotyper version 1 . 1 ( Bruker-Daltonics ) and ClinProTools 3 . 0 ( bruker-Daltonics ) as described [18] . ClinProTools software was used to visualize all spectra as virtual gels and to calculate variability for each of the defined markers . The Biotyper software performs smoothing , normalisation , baseline subtraction , and peak picking using default parameters . Strain comparison was done by principal component analysis ( PCA ) [27] . Distance values were calculated using Biotyper to build score-oriented dendrograms . Based on these distance values , a dendrogram was generated using the according function of the statistical toolbox of Matlab 7 . 1 ( The MathWorks Inc . , USA ) , which was integrated into Biotyper 1 . 1 . The clustering approach is based on similarity scores implemented in the software . Reproducibility was evaluated by comparing spectra obtained from two independent experiments for each strain . The repeatability and stability of the profiles over generations was tested using a series of extracts obtained from subcultures . One strain was maintained in 2 separate cultures then analyzed in duplicate every 72 h over 5 weeks .
Preliminary tests showed that stable spectra were obtained with 106 promastigotes but that an optimal discrimination of peaks was achieved with 107 promastigotes . Because the culture medium influences spectra ( not shown ) , isolates growing better in NNN medium were transferred for one cycle ( i . e . , 48–96 hours ) in Schneider medium before analysis . The growth kinetics of 2 L . tropica isolates was established over two consecutive cycles and spectra were obtained at three stages: exponential ( 24–72 hours ) , stationary ( 72–148 hours ) and decay , showing that spectra were reproducible at the exponential and early stationary stages ( data not shown ) . All spectra were then obtained from late exponential/early stationary stages . The reproducibility was further established by analyzing 16 replicates of the same samples for a L . ( L ) infantum and a L . ( V ) braziliensis isolate ( Suppl Fig . S2 & 3 ) , by analyzing samples from the same isolates of L . ( L ) infantum maintained in culture for several days ( Suppl Fig . S4 ) , by analyzing samples from the same isolates of L . ( V ) braziliensis analyzed at day 0 and a second culture of the same isolate frozen and thawed for subculture 6 months later ( Suppl Fig . S5 ) and by analyzing 3 L . ( L ) infantum and 5 L . ( V ) braziliensis isolates ( Suppl Fig . S6 & 7 ) . Identification was accurate in all cases . The same approach was also performed with a L . major isolate maintained in culture in duplicate for 5 weeks . Spectra lead to the same species identification at all points ( not shown ) . Peaks 11121 ( +/−7 ) and 7114 ( +/−4 ) were both present in all 18 isolates belonging to the Viannia subgenus - 13 L . braziliensis , 5 L . guyanensis/L . panamensis –and absent in all 28 isolates of the Leishmania subgenus - 3 L . mexicana/L . amazonensis , 16 L . major , 5 L . donovani/L . infantum , 4 L . tropica/L . killicki ( Table 1 , Fig . 1 ) . Conversely , peaks 6153 ( +/−3 ) and 7187 ( +/−5 ) were present in all isolates of the Leishmania subgenus and absent in all isolates belonging to the Viannia subgenus . Of note , none of these 4 peaks were present in the isolate identified as L . martiniquensis . The discriminating power of other peaks was then interpreted in the context of the 2 subgenera . Within the Leishmania subgenus , peaks 5589 ( +/−3 ) and 11180 ( +/−6 ) were present only in L . mexicana/L . amazonensis isolates and absent in other isolates , identified by reference methods as L . killicki , L . tropica , L . major , L . infantum , L . donovani ( Table 1 & Fig . 1B ) . Similarly , within this Leishmania subgenus , peaks 5630 ( +/−2 ) & 5937 ( +/−2 ) , or 5753 ( +/−3 ) were present in isolates considered by reference methods as L . major , L . tropica respectively . Peak 7875 ( +/−5 ) was present in the 5 isolates allocated by MLST to the L . donovani complex ( L . donovani & L . infantum ) . Peak 5726 ( +/−6 ) was present in the 3 isolates identified as L . donovani by reference methods and absent in the 2 L . infantum isolates . All those species-defining peaks ( Leishmania subgenus ) were absent in the single L . killicki isolate ( Table 1 ) . Within the Viannia subgenus , the pair of peaks 5987 ( +/−3 ) & 6173 ( +/−3 ) was present in all L . braziliensis/L . peruviana isolates and absent in all L . guyanensis/L . panamensis isolates . All L . guyanensis/L . panamensis isolates expressed either the 6015 ( +/−5 ) or the 6234 ( +/2 ) peak that were both absent in all L . braziliensis/L . peruviana isolates . Slight variations for the value of the peaks were observed ( Table 1 ) but – in this relatively limited set of isolates - did not jeopardize the manual , computer independent identification process . Figure 2: shows mass spectrometry spectra from four different isolates of Leishmania included in the reference library ( 2 from the Viannia subgenus 2 from the Leishmania subgenus ) . The four peaks discriminating subgenera and several peaks discriminating species complexes are shown on these spectra , and are labeled with their respective molecular weights thus allowing an easy analysis based on the algorithm ( Fig . 1B ) . A cluster analysis based on a correlation matrix was performed for Old and New world Leishmania isolates , in order to assess the ability of the MALDI-TOF MS to generate a classification consistent with that obtained by reference methods . As depicted in ( Fig . 1A ) , the resulting dendrogram for all Leishmania isolates showed separate clusters corresponding to the species typed by reference methods , falling appropriately into the 2 subgenera ( Leishmania and Viannia ) . Isolates considered by MLST as L . major were located on one branch , clearly distinguished from isolates considered as L . donovani/L . infantum and L . tropica . The single L . killicki isolate analyzed to date was located in the Leishmania subgenus , close to isolates considered as L . donovani/infantum . The dendrogram built from PCA differentiated clearly L . guyanensis/L . panamensis from L . braziliensis/L . peruvianas species complexes but segregations between L . panamensis and L . guyanensis , L . braziliensis and L . peruviana were not possible at this stage . Isolates belonging to the L . mexicana complex fell into the Leishmania subgenus on a distinct branch . It appeared close to the single trypanosomatid isolate from the French West Indies ( recently named L . martiniquensis , Desbois et al . , personal data , Fig . 1 ) .
Applied on a set cultured isolates spanning most Leishmania species of medical importance , MALDI-TOF MS generated a classification consistent with results obtained by reference methods ( MLST or heat-shock protein 70 gene sequencing ) . This was achieved using a simple , database independent analysis of MALDI-TOF spectra based on the algorithm . The simplicity of the analytical procedure allowed a minute output of results as soon as fair parasite growth was obtained in monophasic medium . Taken together , these observations suggest that MALDI-TOF may be a useful tool to facilitate treatment decision in cutaneous leishmaniasis . Treatment of CL should indeed be based on species identification [1] , [3] , [4] , [9] , [10] , [30]–[32] . For example , systemic antimony is generally more effective in patients infected with L . braziliensis than in patients infected with L . guyanensis [33] , [34] or L . mexicana [9] . Conversely pentamidine is frequently used to treat L . guyanensis/panamensis CL [35] but is poorly effective in L . braziliensis CL [36] . Many patients get infected in places where both species circulate and may therefore receive initially first course of a suboptimal treatment . In the Old World , L . major can be treated effectively and easily with a 3rd generation aminoglycoside ointment [37] , [38] but the efficacy of this topical formulation in patients infected with L . tropica or L . infantum is as yet unknown [4] . In all these situations , rapid species identification should help adopt the most appropriate option in a majority of patients . Because more than 80% of cultures in our context are positive in the first 10 days after sampling , and because it takes a few hours to obtain the MALDI-TOF spectrum , in many instances time-to-identification is now short enough to with-hold treatment decision until species identification is available . Analysis of a higher number of isolates will be necessary to deliver a more solid dendrogram , particularly to determine whether MALDI-TOF MS can achieve a robust discrimination between L . braziliensis and L . peruviana , L . panamensis and L . guyanensis , L . donovani and L . infantum , L . mexicana and L . amazonensis . Fortunately , in current algorithms therapeutic decision in cutaneous leishmaniasis is not heavily impacted by these differentials [3] , [6] . The approach presented here has no taxonomic ambition but was evaluated for potential use in medical practice . We limit our conclusions to the ability of MALDI-TOF MS to generate clusters congruent with those raised by reference methods . Because Leishmania species determination is complex , extension of explorations will be performed in the context of multinational networks such as the LeishMan consortium ( http://www . leishman . eu ) that merges information from several European countries and benefits from the presence of experts with strong expertise in Leishmania species identification [14]–[16] . Interpretation of complex spectra in very rare cases of co-infection with two Leishmania species will be attempted in this context . Optimization of pre-analytical steps , including culture conditions , parasite concentration in pellets and protein extraction was important for a robust interpretation of spectra . We selected a 72–96 h incubation period , corresponding to the exponential phase or early stationary phase of growth to limit variations in protein content . Schneider , an axenic medium ( 20% fetal calf serum , penicillin and streptomycin ) , was chosen as the reference because its supports rapid growth of most isolates and is associated with reproducible spectra . The need for a culture step is a weakness of mass spectrometry shared with several other typing methods . In rare instances , parasite isolation was difficult or slow . We partially circumvented this bottleneck by culturing clinical samples simultaneously on Schneider and NNN medium . Once adapted , NNN-dependent isolates were transferred for one cycle in Schneider medium then processed for MALDI-TOF analysis . In the long-term , the approach may be further simplified by using dipsticks developed from discriminating peaks . Sensitive protein detection using immunochromatography directly from a lesion scraping or aspirate – as currently developed for diagnosis- may indeed by-pass the culture step and further accelerate species identification . Another relative limitation of mass spectrometry is that the method is currently handled by reference centers only . However , because mass spectrometers have a wide spectrum of medical applications in microbiology , prices are dropping and cheaper versions are emerging . In the short term , we found that MALDI-TOF ( MS ) was a promising approach to generate spectra from Leishmania promastigotes with high identification at the species level consistent with the reference method . A limitation of the technique is the need for cultivation parasites . Nevertheless , as compared with molecular biology [39] , this approach offers great advantages , in particular speed , simplicity , cost for isolate identification and was easy to integrate into the organization of a clinical laboratory . Not least , the intuitive interpretation of spectra was well-suited for allowing for close interactions between parasitologists and clinicians . These strengths should predictably facilitate rapid treatment decision in cutaneous leishmaniasis . | Cutaneous leishmaniasis is a disease due to a small parasite called Leishmania . This parasite causes disfiguring skin lesions that last for months or years . There are many different subtypes of Leishmania , each giving rise to lesions of different severity and responding to therapies in its own way . Treating physicians must know as soon as possible which subtype of Leishmania is involved to propose the best treatment . Because it is impossible to differentiate the Leishmania subtypes microscopically , the identification of the culprit subtype currently requires complex and expensive typing methods , the results of which are generally obtained several weeks after the diagnosis . Here , we have evaluated the ability of a new method using mass spectrometry to differentiate Leishmania subtypes . Our results were consistent with those provided by reference typing methods and were obtained rapidly after the parasite had been cultured in vitro . This new method may help physicians know very soon which Leishmania subtype is involved thereby facilitating treatment choice . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"dermatology",
"medicine",
"and",
"health",
"sciences",
"clinical",
"research",
"design",
"synthetic",
"biology",
"immunology",
"microbiology",
"pediatrics",
"research",
"design",
"developmental",
"biology",
"clinical",
"medicine",
"global",
"health",
"veterinary",
"science",
"veterinary",
"medicine",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"veterinary",
"diseases",
"veterinary",
"microbiology",
"medical",
"physics",
"proteomics",
"biophysics",
"physics",
"systems",
"biology",
"biochemistry",
"diagnostic",
"medicine",
"clinical",
"immunology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"evolutionary",
"biology"
] | 2014 | Easy Identification of Leishmania Species by Mass Spectrometry |
Peroxiredoxins ( Prxs or Prdxs ) are a large protein superfamily of antioxidant enzymes that rapidly detoxify damaging peroxides and/or affect signal transduction and , thus , have roles in proliferation , differentiation , and apoptosis . Prx superfamily members are widespread across phylogeny and multiple methods have been developed to classify them . Here we present an updated atlas of the Prx superfamily identified using a novel method called MISST ( Multi-level Iterative Sequence Searching Technique ) . MISST is an iterative search process developed to be both agglomerative , to add sequences containing similar functional site features , and divisive , to split groups when functional site features suggest distinct functionally-relevant clusters . Superfamily members need not be identified initially—MISST begins with a minimal representative set of known structures and searches GenBank iteratively . Further , the method’s novelty lies in the manner in which isofunctional groups are selected; rather than use a single or shifting threshold to identify clusters , the groups are deemed isofunctional when they pass a self-identification criterion , such that the group identifies itself and nothing else in a search of GenBank . The method was preliminarily validated on the Prxs , as the Prxs presented challenges of both agglomeration and division . For example , previous sequence analysis clustered the Prx functional families Prx1 and Prx6 into one group . Subsequent expert analysis clearly identified Prx6 as a distinct functionally relevant group . The MISST process distinguishes these two closely related , though functionally distinct , families . Through MISST search iterations , over 38 , 000 Prx sequences were identified , which the method divided into six isofunctional clusters , consistent with previous expert analysis . The results represent the most complete computational functional analysis of proteins comprising the Prx superfamily . The feasibility of this novel method is demonstrated by the Prx superfamily results , laying the foundation for potential functionally relevant clustering of the universe of protein sequences .
Peroxiredoxins ( Prxs ) are a large and ubiquitous superfamily of thiol dependent peroxidases , which have long been known to be involved in the reduction of aliphatic and aromatic hydroperoxides and peroxynitrite in biological systems [1–3] . Historically , these proteins have also been called TSA ( thiol-specific antioxidant ) , AhpC ( alkyl hydroperoxide reductase ) , and TPx ( thioredoxin peroxidase ) . Prxs are known to protect cellular components from oxidative damage [4 , 5] . Indeed , it has been argued that Prxs are one of the most important peroxide scavengers in biological systems [6–9] . In addition to a peroxide scavenger role , Prxs are involved in essential biological processes such as redox signaling , which , because of the Prx reaction efficiency , can occur by one of two mechanisms . In the first mechanism , oxidation of redox-regulated proteins is not caused by H2O2 directly , but is rather mediated by Prxs , such that Prx CP is first oxidized by H2O2 , which then reacts directly with the regulated kinase or phosphatase modifying its function . The regulated protein is subsequently regenerated by a cellular reductant . This signal transduction mechanism has been extensively reviewed [10–12] . In the second signaling mechanism , redox-regulated proteins may be directly oxidized by H2O2 [11 , 13–16] . However , thiol oxidation by H2O2 in redox regulated proteins is typically much slower in cellular proteins than the corresponding H2O2 detoxification by Prxs [17] . Thus , signal propagation occurs by Prx inactivation: Prxs are subject to H2O2 hyperoxidation at the active site cysteine , peroxidatic Cys ( CP ) , which inactivates them ( until they are repaired by the enzyme sulfiredoxin ) [18 , 19] . The “floodgate hypothesis” posits that localized Prx inactivation ( e . g . via hyperoxidation ) serves to promote H2O2-mediated oxidation of redox-regulated proteins [20] and examples of such signaling in cells are emerging [21 , 22] . Hyperoxidation is also reported to play a role in circadian rhythms [23] and chaperone function [24] . Fine control of the Prx reaction mechanism is clearly essential; thus , understanding molecular function of this large and complex superfamily would provide insight into broader biological mechanisms . As one would expect , peroxide detoxification and redox regulatory systems can be quite complex . For example , mammalian cells express six Prx isoforms: 2-Cys ( PrxI , PrxII , PrxIII , and PrxIV ) , atypical 2-Cys ( PrxV ) , and 1-Cys ( PrxVI ) [25] . Chloroplasts contain three Prx isoforms [26] . All Prxs contain CP preceded in the sequence by a conserved Pxxx ( T/S ) xxCP , a definitive motif for the Prx superfamily . An Arg is also absolutely conserved , but is contributed by a sequence fragment close in structure and distant in sequence . These residues activate the peroxide substrate , catalyze peroxide bond breakage , and catalyze attack of the CP thiolate on the substrate hydroxyl [27–29] . The extent and importance of the Prx proteins has led to several approaches to cluster the superfamily based on active site details . At its most simple , Prxs are classified into typical 2-Cys , atypical 2-Cys , and 1-Cys Prx families based on the presence or absence and position of a resolving Cys , CR [30 , 31]; however , proteins may have structural features of one of these classes , but mechanistic details of another [32] . Detailed sequence comparison and evolutionary analysis determined that Prxs diverged from an ancestor of the thioredoxin fold family and identified four classes of Prx , which these researchers called Prx1 , Prx2 , Prx3 , and Prx4 [33] . Subsequent work based on detailed sequence analysis divided the Prx superfamily into six isofunctional families: AhpC/Prx1 ( abbreviated Prx1 ) , Prx6 , Prx5 , Tpx ( thiol peroxidase ) , PrxQ/BCP ( bacterioferritin comigratory protein , abbreviated PrxQ ) and AhpE [28 , 32 , 34] . This level of detailed molecular functional annotation is typically lacking in the sequence databases , as we have previously shown [35] . More recently , we have used a bioinformatics approach based on active site profiling [36] to identify sequences in a given isofunctional family based on active site features [35 , 37] . Active site profiles ( ASPs , Fig 1 ) are used to identify and compare functional site features . The Deacon Active Site Profiler ( DASP ) , a tool that uses ASPs to search databases for sequences containing active site features similar to those in the ASP [37 , 38] identified many additional Prx members of each expertly identified isofunctional group [35] . Using this single search approach , we identified over 3500 proteins in the six Prx functional subgroups; these sequences are available in the Prx database , PREX [39] and in the Structure Function Linkage Database , SFLD [40 , 41] . SFLD curators subsequently added sequences to these groups using their hidden Markov model ( HMM ) approach . The significant question is: could one automatically identify such isofunctional families within a protein superfamily without expert analysis ? Databases such as CATH [42 , 43] , PFAM [44 , 45] and SCOP [46 , 47] have clustered large superfamilies of proteins based upon domain characteristics and/or structural and sequence classification . Such approaches capture broad levels of functional similarity . On the other hand , in SFLD , proteins are clustered based on functional similarity [41] . An SFLD superfamily contains proteins that share part ( but not all ) of their enzyme mechanism . At a more detailed level , SFLD families contain proteins which exhibit the same enzyme mechanism ( i . e . , are isofunctional ) . CATH PFAM , and SCOP families are more similar to what is defined as an SFLD superfamily [40]; such broad groups usually contain multiple isofunctional families . Our goal is to develop a method to more automatically identify isofunctional clusters . Several approaches aim to cluster sequences into isofunctional clusters , including FunFHMMer [48] ( an updated version of GeMMA [49] ) , SCI-PHY [50] , and ASMC [51] . These methods start with known superfamily sequences and subdivide that large set using clustering and pattern recognition of full sequences . SCI-PHY starts with a multiple sequence alignment , builds a hierarchical tree using agglomerative clustering , and identifies the point at which to prune the tree . SCI-PHY includes phylogenetic details in the clustering . ASMC starts with a PFAM family , uses modeling and analysis of specificity determining positions ( SDPs ) to cluster the PFAM family , and structural modelling to create active sites; ultimately structural comparisons are performed to identify functional groups . FunFHMMer starts with and clusters a CATH-Gene3d superfamily . Essentially , FunFHMMer builds weighted HMMs of the identified clusters , so new members of each group can be identified . Both ASMC and FunFHMMer identify SDPs or mechanistic determinants that are weighted heavily in creating profiles . Remaining challenges focus on determining when subdivision is complete and identifying the SDPs more automatically . The method described here , MISST ( Multi-level Iterative Sequence Searching Technique ) , presents a novel approach to identifying functionally relevant clusters . Previous methods start with the complete superfamily and divisively cluster that superfamily , while the current method begins with a few examples and agglomeratively builds the isofunctional clusters from those representatives . To define groups in MISST , we build on the observation that suggests if a group is isofunctional , a DASP search using that group as the input profile self-identifies its members and no other proteins , while groups that are not isofunctional do not self-identify in this way [52] . That is , a group of proteins is deemed a functionally relevant cluster if a database search ( using DASP ) returns all proteins in the group at significant scores and no ( or few ) other proteins at significant scores ( within a range of uncertainty ) . The iterative searches of MISST are built on this observation . The first step in this approach is to identify the starting set of isofunctional clusters , a process called TuLIP ( Two-Level Iterative clustering Procedure ) , during which proteins of known structure that share common active site features are clustered [52] . This process is also built on the same premise: an isofunctional cluster is one that self-identifies in a DASP search . Briefly , TuLIP starts with all structures from a protein superfamily and iteratively subdivides those into smaller and smaller groups based on active site features . At each iteration , each cluster is used in a DASP search of the sequences in the PDB . For each cluster , if the DASP search self-identifies–that is all proteins in the cluster are identified in the search and nothing else–that cluster is deemed a functionally relevant group . All clusters that do not pass this criterion are further subdivided and searched again . Results on the enolase superfamily demonstrate that TuLIP does identify the functionally relevant subgroups and families [52] . In this work , a comprehensive atlas of the Prx superfamily is identified through application of the TuLIP and MISST processes . Four functionally relevant clusters were identified by TuLIP from the known Prx structures . Through MISST iterations , sequences are added to the groups and the four clusters are subdivided into six clusters which correspond to the six expertly identified functionally relevant groups , even though this expert information of six groups was not input into the process . Because TuLIP and MISST involve iterative DASP searches , a modified process , DASP2 , was used in this work . DASP2 database search results are essentially identical to DASP search results , however DASP2 is significantly more efficient than DASP [53] . This agglomerative and divisive approach allowed us to assign molecular functional detail to over 38 , 000 sequences , many of which were previously uncharacterized or annotated as a general Prx ( or one of its synonyms ) . The current work suggests the feasibility of automation of MISST . Though more testing and validation is required , the MISST process should be generally adaptable for the analysis of other protein superfamilies to produce high-quality molecular function annotation and identification of isofunctional clusters within the protein universe .
Identification of functionally relevant clusters among proteins of known structure is the first step in our process and is accomplished using TuLIP , a two-stage approach to clustering structures based on active site features [52] ( see Methods for details ) . TuLIP identifies four functionally relevant clusters from 47 non-redundant peroxiredoxin ( Prx ) structures: three clusters ( Sct2 , Sct3 , and Sct4 ) during the first stage and one ( Rlx6 ) during the second stage ( Fig 2A ) . A good , though not perfect , correspondence is observed between expertly-identified subgroups , as deposited in SFLD , and TuLIP-identified groups ( Fig 3A ) . Prx5 maps one-to-one to TuLIP group Sct3 . TuLIP group Sct4 contains all proteins in three Prx subgroups: AhpE , Prx1 , and Prx6 , a result suggesting similar active site features , which is , indeed , observed ( S1 Fig ) . Prx1 and Prx6 had previously been identified as being closely evolutionarily related , as well [33] . All Tpx proteins are identified in TuLIP group Sct2; Sct2 also contains four PrxQ proteins ( Fig 3A ) . The two other PrxQ structures were grouped into their own cluster , Rlx6 . This subdivision of the PrxQs of known structure was previously observed in hierarchical clustering of active site signatures [35] . Hierarchical clustering based on the canonical Prx active site motif ( S1 Fig ) suggests that residue differences at the PrxQ active site of proteins of known structure are driving this subdivision . The TuLIP clustering results are not unexpected from the limited dataset of known structures and what is known about functional similarities . However , the results do present a challenge for the agglomerative and iterative process of searching sequence space: an ideal process would subdivide Sct4 into the expertly identified functionally relevant clusters and would recombine the PrxQ subgroup . MISST is an iterative search process developed to be both agglomerative and divisive . That is , the process was developed to add ( agglomerate ) sequences containing similar functional site features to each TuLIP group and to subdivide TuLIP groups when functional site features suggest distinct clusters . As an illustration , MISST should identify the two groups represented in the ASP in Fig 1B without curator intervention . The MISST process is outlined in Fig 4 and described in detail in Methods . Briefly , the process involves iterative DASP2 searches of GenBank , each followed by evaluation for cluster division , combination , and self-identification . DASP2 is a more efficient version of the DASP sequence-searching method that focuses not on the complete protein sequence , but rather only on a protein’s functional site features [37 , 38 , 55] . Groups defined by MISST should , thus , be identified and subdivided based on their mechanistic differences . Notably , no step in the MISST process requires human evaluation—the process should be automatable , although adjustment of two parameters may be needed once the process is automated . ASPs were created from sequences in each of the four TuLIP-identified groups: Sct2 , Sct3 , Sct4 , and Rlx6 ( profiles are provided in S1 File ) . Each ASP was used as input into an iterative process of DASP2 GenBank searches ( see MISST flow chart , Fig 4A ) . Following each iteration , each group was evaluated for self-identification ( Fig 4A ) and need for subdivision ( Fig 4C ) . If a group self-identifies , it is removed from the iterative process and set aside for final analysis . For all other groups , a new ASP is created from functional site pseudo-signatures ( see Methods ) of sequences identified at scores ≤1e-12 ( Search0 ) or ≤1e-14 ( subsequent search iterations; see Supplemental Methods in S3 File for justification , validation , and broader applicability of these score thresholds ) . Five search iterations ( Search0 through Search4 ) were performed ( Fig 2B ) . All groups satisfied self-identification criteria after Search3 except Rlx6_PrxQ which satisfied the criteria after Search4 . Through the iterations , sequences were added to each group and the four original TuLIP groups were divided into six . The process of adding sequences and splitting groups is represented in the dendrogram in Fig 2B; proteins found in the final groups are visually represented as networks in Fig 2C . Qualitatively , the six groups correspond almost perfectly with the six functionally relevant groups previously identified by experts [35] ( Fig 3B ) . These searches identified 38 , 739 sequences ( Table 1 ) in six groups ( DASP2 score threshold ≤1e-14 ) . Proteins identified in each cluster are provided in S2 File . 6 , 855 of these proteins are annotated in SFLD to the subgroup matching the MISST group [41] . 30 , 096 proteins were not previously identified by a single DASP search [35] or by SFLD HMM analysis ( Table 1 ) ; new sequences were identified due in part to their absence from the GenBank database during earlier analyses and to the more robust analysis method used here . To ascertain whether all 38 , 739 proteins are likely Prx superfamily members , we determined how many contained the canonical Prx active site motif Pxxx ( T/S ) xxCP [3 , 56 , 57] . Across all searches , this fragment is found in 99 . 3% of all MISST-identified sequences , indicating almost all sequences likely belong to this superfamily . We next explore how the MISST process agglomerates sequences and subdivides groups . We then quantitatively compare the MISST-identified groups to the previously identified sequences . Because MISST utilizes DASP2 with its focus on functional site features as the search mechanism , we can hypothesize mechanistic determinants important for each group’s function and compare the functional site features of these expanded groups to those described by experts [35 , 58] . MISST iterations initiated with seven Prx5 proteins in TuLIP group Sct3 ultimately identify 5434 proteins . This coherent group was not further split by PSSM analysis ( Fig 2B , purple dendrogram branch ) , likely because of the strong intragroup active site similarity . 1039 of the MISST-identified proteins are identified as Prx5 sequences in SFLD , representing 97 . 8% coverage ( recall ) . The Prx5 proteins deposited in SFLD were identified through one DASP iteration [35]; a few more were added through the SFLD curation processes . This group contains no proteins from any other Prx subgroup ( Fig 3B ) ; consequently , Sct3 is mapped to Prx5 for subsequent analysis and herein called Sct3_Prx5 . Sct3_Prx5 includes 252 proteins identified in SFLD as belonging to the Prx superfamily , but uncharacterized with respect to subgroup; thus , the functional subgroup of these proteins can now be defined more precisely . 4143 Sct3_Prx5 proteins were not previously identified as Prx5 ( Table 1 ) demonstrating that , if the new identifications are correct , search iterations of MISST add significantly to our knowledge of functionally related proteins . Consequently , the probability that these proteins are actual Prx5 proteins was evaluated by determining the presence or absence of the Prx5-specific active site motif P ( G/A ) A ( F/Y ) ( T/S ) ( P/G ) xCP [9] ( Fig 5A , part of red brace ) . 97 . 4% of all Sct3_Prx5 sequences contain this motif . The percentages do not differ between previously known and newly identified proteins: 98 . 2% of previously identified Prx5 proteins , 96 . 4% of Prx sequences in SFLD that are uncharacterized relative to subgroup , and 97 . 2% of new ( non-SFLD ) proteins contain the motif . Given that the percent of both knowns and new proteins containing this motif is similar , there is high probability that the MISST iterations are consistently identifying proteins that belong to the Prx5 functional family . To quantitatively evaluate sequence identification , F-measure , the harmonic mean of precision and recall [59] , was calculated for Sct3_Prx5 sequences . For this analysis ( and similar analyses of other groups ) , Prx proteins in SFLD are the known sequences; “positive” sequences are the proteins in the subgroup under consideration , while “negative” sequences are Prx sequences in all other subgroups . Thus , if a known Prx5 was identified by MISST , a true positive was counted . If a sequence from another Prx subgroup was identified as part of Sct_Prx5 , a false positive was counted . False negatives were Prx5 sequences identified in the previous work [35] , but not identified in this search . A true negative is counted if MISST did not identify Prx sequences known to be members of other Prx subgroups . Sequences identified by MISST , but not by previous methods , were not included in this analysis , as their assignment as true or false positives or negatives could not be evaluated . This is a difference between MISST and other methods: instead of subdividing a superfamily in which all proteins are thought to be known at the start [49–51] , MISST agglomeratively adds proteins from the database and subdivides the groups . F-measure analysis demonstrates the high quality of assignments to Sct3_Prx5 ( Fig 6A ) : the F-measure is 0 . 99 at the DASP2 search score threshold ( ≤1e-14 , dashed line Fig 6A ) . As the DASP2 score threshold becomes more significant , recall gradually decreases ( as some proteins are missed ) ; however , precision never drops below 1 for Sct3_Prx5 . Neither precision nor recall decrease in this group as the DASP2 score threshold becomes less significant ( yellow , orange , and red bars , Fig 6A ) , indicating no false positives are identified at ≤1e-8 , even prior to cross hit analysis ( Fig 6B ) . Detailed analysis of the Sct3_Prx5 functional site pseudo-signatures identify mechanistic determinants distinctive to this subgroup ( Fig 5A; structures in S2 Fig ) . These determinants were not identified a priori as input . The Prx active site motif includes elements distinctive to the Sct3_Prx5 subgroup: P ( G/A ) A ( F/Y ) ( T/S ) ( P/G ) xCP ( bold indicates residues almost invariant across the superfamily; [9 , 35] ) . Outside of this motif , two defining features are observed: His is almost invariant at signature position 15 ( Fig 5A red brace ) and a pair of Arg residues ( RSxR ( Y/F ) ) at positions 33–37 ( yellow brace ) . The second of these conserved Arg residues is the one recognized to play a major role in activating the peroxide substrate for–O-O–bond scission at the Prx active site [9] . In the structure 1TP9 , the side chain of the His residue conserved in Prx5 proteins is hydrogen bonded to the side chain of this invariant Arg ( signature position 36; S2A Fig ) . The location of these side chains in the active site near the CP suggests a role in mechanism , perhaps with the His playing a role in proton transfer . Reasonably well conserved motifs in the pseudo-signatures of this subgroup also include VMxxW at signature positions 20–24 and ( C/V ) ( V/L/I ) ( S/A ) VN at signature positions 39–43 ( Fig 5A , purple and fuchsia braces , respectively ) . The Cys in this second fragment is found in 76% of sequences; 19% of sequences have Val at this position . Further , phylogenetic evidence suggests conservation of this Cys , which sometimes serves as the CR , may be based on phylogeny ( S3 File; S3 Fig , red brace ) . Starting with just nine structures , MISST agglomerates sequences into a coherent Prx5 cluster . Even though PSSM analysis was performed at each iteration , the Sct3_Prx5 group did not split , suggesting that the PSSM approach does not split functionally relevant clusters . Sct2 was originally comprised of four PrxQ sequences and nine Tpx sequences ( Fig 3A ) . Known Tpx structures contain the resolving cysteine , CR , in the α3 helix . The CR is not found in a consistent location in the four TuLIP-identified PrxQ proteins . Using the Sct2 TuLIP group as MISST input illustrates sequence agglomeration and increasing coherence within a cluster , despite the group’s initial heterogeneity . At Search0 , known PrxQ proteins are identified at less significant DASP2 search scores than the Tpx proteins ( Fig 7A ) . By the second iteration ( Search1 ) known PrxQ proteins are not identified ( Fig 7B ) . Iterative DASP2 searches produce more robust profiles and each successive search produces a more coherent set of sequences that exhibit common active site features . With each iteration , additional Tpx proteins accumulate , with a plateau reached in Search2 and Search3 ( Fig 7C ) . At Search3 , ≥ 95% of sequences used as input to Search3 and ≤ 15% new sequences were identified at significant DASP2 scores ( ≤1e-14 ) in the GenBank Search3; thus , self-identification criteria were satisfied following Search3 ( Fig 2B , green dendrogram branch ) . At this point , the group was homogeneous for Tpx proteins and is thus called Sct2_Tpx . The final Sct2_Tpx cluster contains 4930 sequences—860 are in SFLD and annotated to the Tpx subgroup , 244 are marked as Prx-uncharacterized in SFLD , and 3826 are not in SFLD ( Table 1 ) . F-measure shows high precision and recall values for Tpx proteins ( Fig 6A ) . After this group satisfied the self-identification criteria , no false positives were identified even at less significant scores of ≤1e-8 . 860 SFLD Tpx sequences represent 90 . 1% coverage ( recall ) of known subgroup members; F-measure is 0 . 95 at the DASP2 score threshold of ≤1e-14 ( Fig 6A ) . Of the final sequences in this Sct2_Tpx group , 98 . 5% contained the Prx active site motif distinctive for this subgroup: PS ( I/L/V ) DTx ( V/T/I ) CP ( Fig 5B , red brace ) , which refines the motif determined from the previously identified smaller dataset [9 , 35] . The sequences are 99 . 94% bacterial ( S4A Fig ) , consistent with what was previously reported on the smaller dataset [58] . Additional mechanistic determinants can be hypothesized for the Sct2_Tpx subgroup . Signature positions 15 and 16 are distinctive in this group: a branched residue ( Val or Thr ) followed by Arg or Lys ( Fig 5B , red brace ) . A distinctive AxxR ( F/W ) C motif is observed at signature positions 20–25 ( Fig 5B , purple brace ) . This conserved Cys is the CR in helix α-3 . As in Sct3_Prx5 and Sct4_Prx1 , the nearly invariant ( 99 . 3% ) Arg at signature position 36 is the active site residue required for efficient catalysis [9 , 60] . In the structure 3P7X , the side chain of this Arg is hydrogen bonded to CP . It is preceded by a very well-conserved Leu at signature position 33 , the only subgroup with a well conserved hydrophobic residue at this position ( Fig 5B , yellow brace ) . Both Arg ( gray ) and Leu ( black ) extend towards CP ( S2B Fig ) . Finally , the Sct2_Tpx subgroup contains a Ser that is almost invariant at signature position 42 ( Fig 5B , fuchsia brace ) . These residues are proximal to the active site , suggesting a functional role ( S2B Fig ) . This example illustrates how the iterative MISST process creates more coherent groups , even when the original TuLIP group is composed of two subgroups . While the PrxQ structures were not present in the final Sct2_Tpx MISST group , this subgroup was not lost in the MISST process . As discussed subsequently , the PrxQs were identified as a subdivision of the Rlx6 group . The clustering process described herein starts with proteins of known structure; however , the structure database is a very limited representation of the sequence space universe . Because of this limitation , TuLIP sometimes combines multiple subgroups into one cluster [52] , as is the case with Sct4 , which contains both Prx1 and Prx6 proteins . Consequently , any agglomerative process aimed at identifying functionally relevant groups must recognize the need for cluster subdivision . PSSM Analysis was developed as an automatable process to do just this . PSSM Analysis is performed at each MISST iteration after the first ( Fig 4A ) using the outlined process ( Fig 4C; details in Methods ) . Essentially , the active site pseudo-signatures identified in the GenBank search are used to quantitatively determine if and how the group should be subdivided . If subdivision is required , two new ASPs are created from the appropriate pseudo-signatures . These ASPs are input to a DASP2 search of GenBank . Search outputs are compared to verify the groups are , in fact , unique . Notably , PSSM Analysis was performed at each search iteration for both the Sct2 and Sct3 MISST groups , but distinct , functionally relevant groups were not identified within either group . The TuLIP-identified Sct4 group includes all known structures from the Prx1 , Prx6 , and AhpE subgroups ( Figs 3A and 4A ) . At Search1 , PSSM Analysis identifies two groups; these groups evolve distinctly through subsequent search iterations ( Fig 2B , red and blue dendrogram branches ) . Notably , though the AhpE subgroup is not identified in Sct4 after Search1 , the AhpE subgroup is not lost . It is ultimately identified in Rlx6 using this same PSSM Analysis procedure ( discussed subsequently ) . Analysis of each search iteration provides insight into the PSSM Analysis of Sct4 ( Fig 8 ) . The Search1 DASP2 score distribution is bimodal—Prx1 sequences at more significant and Prx6 sequences at less significant DASP2 search scores ( Fig 8A , blue and red bars ) . PSSM Analysis correctly identifies these two groups ( Fig 8A , yellow and green boxes ) . One ASP is created each for sequences in the yellow and green boxes; each ASP is used in Search2 of GenBank . Prx1 and Prx6 sequences are identified distinctly in Search2 ( Fig 8B , Search2 distributions ) . After just one more GenBank search iteration ( Search3 ) , each group passes self-identification criteria . Ultimately , 9660 and 5212 sequences are identified at significant DASP2 scores in Sct4_Prx1 and Sct4_Prx6 , respectively ( Table 1 ) . Of the proteins annotated in SFLD , 96 . 6% of Prx6 proteins and 95 . 7% of Prx1 proteins are identified ( Table 1 ) . Both searches identify Prx sequences annotated as Prx-uncharacterized in SFLD: 127 and 289 are identified as part of Sct4_Prx6 and Sct4_Prx1 , respectively . Finally , 4143 and 7241 GenBank sequences not annotated in SFLD were identified as Sct4_Prx6 and Sct4_Prx1 members , respectively ( Table 1 ) . The Prx active site motifs for Sct4_Prx1 and Sct4_Prx6 are distinct: PxDF ( T/S ) FVCP and Px ( D/N ) ( F/Y ) TPVCP , respectively ( Fig 5C and 5D , red braces ) . 93 . 8% and 96 . 7% of all sequences in Prx1 and Prx6 , respectively , exhibit these motifs , demonstrating that MISST iterations and the PSSM Analysis distinguish these small active site differences . F-measure at the score threshold of ≤1e-14 is high for both groups: 0 . 98 at a DASP2 score threshold of ≤1e-14 for each ( Fig 6A ) . Thus , PSSM Analysis can effectively subdivide one group into two functionally relevant clusters . As with the other groups , we can identify mechanistic determinants that distinguish Sct4_Prx6 and Sct4_Prx1 . A key distinguishing feature is the TFVC versus TPVC for Prx1 and Prx6 , respectively: this one residue in the canonical Prx active site motif distinguishes these two subgroups ( Fig 5C and 5D , red brace; S2 Fig , cyan side chains ) . Another distinguishing feature is a Phe-Tyr ( Prx1 ) compared to Ser-His ( Prx6 ) at signature positions 2 and 3 ( Fig 5C and 5D , blue brace ) . In 2V2G , this His is in the active site , near the CP ( S2D Fig , yellow side chains ) . Again , Arg at position 36 in Prx1 is the active site residue required for efficient catalysis; the fragment containing this Arg is not part of the Prx6 signature . In both subgroups , the almost invariant Ser ( at signature position 42 ) and the almost invariant His ( at signature position 21 ) form a potential path for proton transfer in these subgroups ( S2C and S2D Fig , light pink side chains ) . CR is not observed within the Prx1 group profile because it is contributed from a different chain ( the partner subunit of the dimer ) . There is no CR in most Prx6 members [35] . Interesting phylogenetic observations at specific positions , including the well-known GG ( L/I/V ) G motif [31] , are discussed in S3 File . Previous sequence analysis methods identified Prx1 and Prx6 as only one group , which the authors named Prx4 [33] . Subsequent expert analysis clearly showed that Prx6 was a distinct functionally relevant group [35] . MISST , a method that focuses on differentiating active site features , has accomplished that which was previously accomplished only by expert curation—to divide these two closely related isofunctional clusters without human curation . This opens the exciting possibility of functionally relevant clustering of superfamilies for which functional groups are not known . PrxQ and AhpE were members of original TuLIP groups , but were lost from Sct2 and Sct4 searches , respectively , during MISST iterations . TuLIP group Rlx6 contained two of the six PrxQ structures known at the time this research was completed . The task is even more difficult because AhpE is a very small subgroup containing only 25 proteins in 2011 [35] and 112 in the current SFLD; previously , these proteins were found in only one class of bacteria ( actinobacteria ) [58] . Only one structure is available in the PDB database . Are these groups that are less well represented by structures identified through the iterative MISST process applied to TuLIP group Rlx6 ? The answer to this important question is yes . Analysis of the Rlx6 MISST search iterations illuminates how AhpE and PrxQ sequences are identified and subdivided ( Fig 2B , pink and yellow dendrogram branches ) . The Search0 ASP input contained only two PrxQ proteins ( Fig 2A ) ; Search0 output contained mostly PrxQ proteins , with a few AhpE proteins ( not shown ) . Per the MISST process ( Fig 4 ) , an ASP was created for sequences identified at a DASP2 score threshold of ≤1e-12 . This ASP was input to Search1 . Search1 output includes a small number of AhpE and PrxQ proteins at more significant scores; most PrxQ proteins are identified at less significant DASP2 scores ( Fig 9A ) . PSSM Analysis divides Search1 sequences into two groups: AhpE and PrxQ ( Fig 9A , blue and green boxes ) . An ASP is created for each group , and each ASP is input to DASP2 Search2 . Results are distinct: one search is populated with mostly AhpE and a few PrxQ proteins , the other populated almost solely with PrxQ proteins ( Fig 9B ) . The Rlx6_AhpE and Rlx6_PrxQ groups subsequently remain distinct ( as determined by the agreement criterion; Fig 4B ) and pass self-identification criteria at Search3 and Search4 , respectively ( Fig 2B , yellow and pink dendrogram branches ) . The two groups map easily to subgroups identified by experts . One , Rlx6_PrxQ , contains 12 , 014 sequences; 1786 of these sequences are found in SFLD , which represents 92 . 1% of known PrxQ proteins ( Table 1 ) . 739 sequences are annotated in SFLD to the Prx superfamily but not a specific subgroup . MISST identifies 9489 sequences in this cluster that were not previously assigned to the Prx superfamily ( Table 1 ) . Consistent with the other MISST-identified groups thus far discussed , F-measure ( and , thus , precision and recall ) is quite high , 0 . 96 , for Rlx6_PrxQ ( Fig 6A ) . Rlx6_AhpE is by far the smallest subgroup identified by MISST: only 1489 sequences are identified in this cluster . 98 of those proteins are currently annotated as AhpE in SFLD , which represents 87 . 5% of the 112 known AhpE proteins . 1254 Rlx6_AhpE proteins were not previously identified as Prxs ( Table 1 ) . Notably , F-measure for this cluster is not as strong as the other MISST-identified groups—only 0 . 74 at the DASP2 score threshold of ≤1e-14 . In addition , the F-measure is never above 0 . 78 , even at less significant score thresholds ( Fig 6A ) . Detailed analysis explains this result . There are 112 nonredundant AhpE sequences in SFLD . At thresholds of ≤1e-14 , ≤1e-12 , and ≤1e-10 , we identify 98 , 107 , and 108 of them , respectively; thus , recall is high , at 87 . 5% at ≤1e-14 and increases to 96 . 4% at ≤1e-10 . However , 54 Rlx6_AhpE proteins identified at the DASP2 search score threshold of ≤1e-14 were previously identified as PrxQ subgroup members [35] . These proteins decrease the precision of the result . The question of functional assignment of these 54 sequences is an important one; these sequences are listed in S2 File and discussed subsequently . The two clusters derived from Rlx6 exhibit common active site features , such as the Phe at signature position 2 ( Fig 5E and 5F , blue brace ) , which is highly conserved in both groups . However , the Prx active site motif is distinct between Rlx6_BCP and Rlx6_AhpE , including the canonical Prx active site motif: P ( K/A/R ) ( D/A ) xTxGC and PxAF ( T/S ) xxC for Rlx6_PrxQ and Rlx6_AhpE , respectively ( Fig 5E and 5F , red braces ) . 90 . 2% of proteins identified in Rlx6_PrxQ contained its motif , while 94 . 2% of the Rlx6_AhpE sequences , including 92 . 9% of those previously identified as AhpE and 88 . 9% of those previously annotated as PrxQs , contain its motif . Notably , Rlx6_PrxQ is the only subgroup with a Gly strongly conserved immediately preceding CP , which might suggest unique conformational or dynamical motion associated with PrxQ function . Other positions also distinguish these two Rlx6-derived groups . Glu is invariant at signature position 14 in Rlx6_AhpE , while the residue can be either Glu or Gln in Rlx6_PrxQ . The final two active site fragments are also distinct ( Fig 5E and 5F , fuchsia and purple braces , respectively ) . A G ( V/I ) SxD motif at positions 40–44 and a Leu at position 49 are strongly conserved in Rlx6_PrxQ . Notably , the invariant Gly , Ser , and Asp of the G ( V/I ) SxD motif are all in the 5ENU active site , along with the conserved Leu . These distinctive features suggest that , indeed , these two subgroups are functionally distinct . The question remains: what is the correct functional classification of the 54 sequences previously classified as BCP [35] and classified by MISST as AhpE ? A closer analysis of the active site signatures may explain the unexpected clustering . Signature logos were created for the 1786 Rlx6_PrxQ sequences that were previously annotated as PrxQ , the 98 Rlx6_AhpE proteins previously annotated as AhpE , and the 54 Rlx6_AhpE sequences previously annotated as PrxQ ( Fig 10 ) . Multiple positions in the active site signature illustrate why the 54 sequences previously annotated PrxQ are now identified in the AhpE MISST group , including a strongly conserved Ala-Phe dyad in the canonical PrxQ active site motif ( PxAF ( T/S ) xxC ) , a conserved Val or Ile immediately preceding CP , and four other positions ( Fig 10 , orange highlights ) . These results demonstrate the DASP2 method identified these 54 proteins in the Rlx6_AhpE subgroup because of common features at the active site . Further , specific residues can be identified that distinguish bacterial ( 81% ) and archaeal ( 19% ) proteins in the Rlx6_AhpE subgroup ( full discussion in S3 File ) . The biological relevance of these observations remains to be determined . In conclusion , the original Rlx6 TuLIP group contained just two of six PrxQ structures , and the lone AhpE structure was in Sct4 , not Rlx6 . Despite only one known AhpE structure , PSSM Analysis and iterative DASp2 searches extracted the AhpE functional group from the Rlx6 search results . These results demonstrate the MISST process can identify functional groups for which structural representation is limited . This is an important result , as many protein superfamilies do not contain comprehensive structural representation in all functional families . Over all six Prx functional groups , the iterative MISST process meets the challenges presented by the TuLIP results: all six Prx subgroups were identified in a robust and comprehensive fashion , even though not all groups were well-represented in the structure database . Results presented thus far demonstrate MISST can both add sequences to functionally relevant groups and subdivide groups into clusters exhibiting distinctive functional features . F-measure ( precision and recall , Fig 6A ) was described for each group individually . Further quantitative comparison between groups , including cross-hit counts and measures of performance , are essential to determine if these groups are distinct and functionally relevant . Cross-hits are defined as the same sequence identified in more than one MISST group at a given DASP2 search score threshold . This analysis demonstrates discreteness of MISST groups . In creating the final groups , a cross-hit analysis similar to that previously described [35] is performed ( see Methods ) . Only 20 proteins are removed in this final cross-hit analysis; the identities of the proteins which cross-hit are listed in S1 Table . To understand the discreteness of the MISST-identified groups , the correlation of cross-hits ( counted prior to this final cross-hit analysis ) with the DASP2 search score threshold was evaluated ( Fig 6C ) . At DASP2 search score thresholds of ≤1e-16 and more significant , all groups are distinct—the number of cross-hits is zero . At the significance threshold of ≤1e-14 , the threshold identified as a “trusted” threshold in the work described here ( see Supplemental Methods in S3 File ) , 20 cross-hits are identified corresponding to a cross hit rate of 0 . 052% , an extremely low false positive rate ( Fig 6C , table and red data point ) . Cross-hits increase drastically as the DASP2 search score threshold decreases in significance ( Fig 6C ) . We can observe the evolution of the cross-hits and , thus , better understand the relationship between group active sites by analyzing “fireworks plots , ” a form of network analysis ( S5 Fig ) . At a DASP2 search score threshold of ≤1e-8 , Prx5 and Tpx subgroups are most distinct and only exhibit a few cross-hits to other groups , which are mostly gone at a threshold of ≤1e-10 ( S5A and S5B Fig ) . At a score threshold of ≤1e-12 , the other four groups become more distinct ( S5C Fig ) . At a score threshold of ≤1e-14 , only twenty cross hits remain . Ten of these twenty cross-hits at ≤1e-14 are between AhpE and PrxQ ( S5D Fig ) , indicating the functional sites of these groups are more closely related to each other than they are to the other groups , as discussed above . The other ten cross-hits are distributed between Prx1 , Prx6 , and PrxQ ( S5D Fig ) . The remaining analysis , F-measure and Performance , assumes that the expert annotations deposited in SFLD [35] are correct . These sequences were identified by a single DASP search of GenBank using expertly-created ASPs . Subsequently , sequences were added using the SFLD HMM approach . The resulting sequences were curated by hand and deposited in SFLD; these annotations are the best known molecular functional annotations for the Prx superfamily . Only 412 sequences previously identified as Prx are not identified as part of the proper MISST group ( out of 7267 Prxs in SFLD with subgroup annotations ) . These sequences are evenly spread over the six groups and are counted as false negatives in the recall calculation of F-measure . 54 of these are the sequences previously annotated as PrxQ , but identified in this analysis as AhpE . 194 sequences were identified above the DASP2 score threshold of 1e-14 . About 25 of the sequences are no longer in GenBank . Since the 2011 analysis of the Prx superfamily , GenBank has grown from 11 . 9M proteins to over 54 . 8M proteins at the end of 2015 . With this growth comes many new sequences identified in our MISST searches that are not annotated in SFLD . To quantify the performance of MISST , all sequences not annotated in SFLD were not used for the F-measure and Performance analyses as the correct annotation is unknown . ( In the previous sections , we demonstrated the likelihood that these newly identified sequences were Prx by evaluating the presence and absence of the canonical Prx active site motif , Pxxx ( T/S ) xxCP , as well as the active site motif associated with each subgroup . ) To analyze the overall accuracy of the MISST process , a performance score was calculated [49 , 50] taking into account purity , edit distance , and VI distance [62] ( Fig 6D ) . These measures were calculated by defining the proteins in each group as TP , TN , FP , or FN; these definitions were based on the previous Prx annotation [35] ( see Supplemental Methods in S3 File ) . Purity provides a measure of the proportion of groups which contain only one subgroup . As Rlx6_AhpE is the only group containing false positives , purity remains at 83 . 3% ( 5 out of 6 groups are pure ) until highly significant DASP2 search score thresholds ( Fig 6D , blue ) . Edit and VI distances measure how many changes are required to transform one grouping method ( MISST ) to another ( SFLD ) . The high correlation between the six SFLD subgroups and the six MISST groups leads to low edit and VI distances , particularly at less significant score thresholds ( Fig 6D , black and red ) . The increase in edit and VI distance values at more significant scores is due to the presence of “singlets , ” which in this case are Prx proteins in the SFLD not identified as a member of any MISST group . Typically , edit and VI distances are used to compare two clustering methods which both start with the same set of proteins . However , MISST is an agglomerative method and does not begin with the full set of proteins; therefore , some proteins in the SFLD are not identified by MISST . Thus , as the DASP search score threshold becomes more significant , more proteins are classified as “singlets” because they are not identified in any MISST groups at the given threshold . Purity , edit , and VI distance were combined into an overall performance measure ( Fig 6D , purple ) . A maximal performance score of 90 . 3 is found at a score threshold of ≤1e-8; the performance at the threshold of ≤1e-14 is 88 . 9 ( Fig 6D , colored arrows ) . Performance does not reach 100 at any point because not all known Prx proteins are identified by the six MISST groups and some PrxQ-annotated proteins are identified in the Rlx6_AhpE group . Performance increases slightly at the less significant score thresholds , simply due to the behavior of edit and VI distance with “singlets . ” The value of 88 . 9 at a score threshold of ≤1e-14 compares well with the performance values reported for clustering of other gold-standard SFLD superfamilies ( amidohydrolase , crotonase , enolase , HAD , terpene cyclase , VOC ) by SCI-PHY ( performance ranged from 54 . 99 to 91 . 70 , with an average of 75 . 36 ) and GeMMA with a generalized cutoff ( performance ranged from 53 . 64 to 90 . 70 , with an average of 80 . 42 ) [49] . It is important to note that performance scores vary widely for both SCI-PHY and GeMMA , indicating more superfamilies must be tested using MISST to complete a full-scale comparison between methods . However , this initial test using the Prx superfamily demonstrates the feasibility of the current approach . Previous work has illustrated how different comparison measures ( sequence , structure , functional site ) can produce different clusters within a protein superfamily [54] . Here we explore that further , by evaluating full sequence similarity between the functionally relevant MISST clusters . A representative network ( RepNet ) was built from the 38 , 739 sequences identified in the six MISST groups . Each of the 1 , 369 nodes represents proteins sharing 55% sequence identity; each edge represents the pairwise BLAST score ( sequence comparison ) between the representatives of the two nodes . Nodes are colored based on the MISST group to which the sequences belong ( see Methods for more details ) . The network is filtered at a variety of BLAST score thresholds to visualize the full length sequence similarity among the MISST groups ( Fig 11 ) . As the threshold for edge becomes more stringent , groups begin to separate . Notably , and as expected , the sequences within MISST groups are more similar to each other than they are to proteins from other groups . This observation is the reason that full sequence comparison methods ( like BLAST ) do reasonably well at identifying the superfamily level of function , like peroxiredoxin . Notably , no single threshold can be identified to distinctly identify the six known subgroups , an illustration of why full sequence comparison methods are less successful at identifying detailed levels of molecular function , such as distinguishing between Prx1 and Prx6 . At less stringent edge ( BLAST score ) thresholds ( Fig 11B ) , some subgroups are indecipherable from one another ( such as Prx1 to Prx6 and AhpE to PrxQ ) , and at more stringent thresholds ( Fig 11D ) , some subgroups begin to split unnecessarily ( such as PrxQ , AhpE , and Prx5 ) . Unsurprisingly , the Prx1 and Prx6 subgroups are difficult to distinguish from one another until the most stringent threshold . Previous work has demonstrated that the similarity between these 2 subgroups makes it difficult to separate them based on sequence comparison alone [33] . MISST focuses on active site features to define isofunctional groups , thus eliminating reliance on full sequence comparison for detailed molecular function analysis . In this work , active site features are utilized to define functionally relevant clusters . A method , MISST , which uses self-identification of clusters to define functional relevance is introduced . The method is both agglomerative and divisive . As ASPs become more robust , DASP2 searches agglomerate more functionally related sequences . Likewise , at each stage , clusters are evaluated for the presence of groups that exhibit distinct functional site features . Functionally relevant clustering of the Prx superfamily presents several challenges for the method: How are sequences agglomerated ( Tpx and Prx5 ) ? How are clusters subdivided when they contain two distinct isofunctional groups ( Prx1 and Prx6 ) ? And , how are functionally relevant groups identified when structural representation is extremely limited ( AhpE ) ? A defining feature of functional annotation is the hierarchy under which groups of proteins are classified [40 , 41 , 54] , and it is important to understand how the MISST results fit into a functional hierarchy . Members of the Prx functional superfamily all perform a similar redox chemistry at CP; differences lie in substrate recognition and details of how CP is regenerated . The six expert-annotated groups of Prxs are classified as subgroups in the SFLD , which indicates that group members share more features among themselves than with members of other Prx subgroups . MISST distinguishes these six subgroups , and members thereof , identifying the differences between the mechanisms from which hypothesis-driven experiments can be developed . As expected , many new sequences were identified—the Prx data in SFLD is from 2010 and 2016 GenBank is significantly larger . Over 99% of newly identified sequences contain the canonical Prx active site motif . Additionally , with the exception of the AhpE subgroup , the phylogenetic distribution for each subgroup is reasonably consistent with the original Prx data , as recently reported by Poole and Nelson [58] . The current work demonstrates the feasibility of this novel , agglomerative approach of using self-identification to identify isofunctional clusters . Notably , the MISST process does not require human- or expert-based analysis and is automatable , with the exception of identification of the key functional residues which are input to TuLIP . Two MISST parameters may require further adjustment to demonstrate generalizability: score thresholds and self-identification criteria . However , our work on the enolase and Prx superfamilies suggests the score thresholds are generalizable ( S3 File ) . The feasibility of MISST is demonstrated here on the Prx superfamily . More extensive parameterization , validation , and generalizability will be demonstrated once the method is automated . Ultimately , we envision that MISST could be applied to cluster any protein superfamily automatically , thus laying the foundation for functionally relevant clustering of the universe of protein sequences .
The peroxiredoxin ( Prx ) superfamily contains six subgroups previously identified by expert analysis: Prx1 ( formerly AhpC/Prx1 ) , AhpE , PrxQ ( formerly BCP/PrxQ ) , Prx5 , Prx6 , and Tpx [28 , 32 , 34 , 35] . These expertly-identified subgroups are available in PREX [39] . Curators at the Structure-Function Linkage Database ( SFLD ) constructed hidden Markov models ( HMMs ) for each subgroup and have updated the proteins in each subgroup in SFLD ( Prx superfamily , EC 1 . 11 . 1 . 15 ) [40 , 41] . As of March 7 , 2016 , there were 7 , 267 annotated Prx proteins ( unique EFDIDs ) in SFLD , distributed among the subgroups as follows: 2 , 225 Prx1 , 112 AhpE , 1 , 939 PrxQ , 1 , 062 Prx5 , 975 Prx6 , and 954 Tpx . Additionally , there were 4 , 695 proteins assigned to the Prx superfamily but not assigned to a subgroup ( uncharacterized ) in the SFLD . Active site profiling is a method used to identify the residues in the structural vicinity of a protein’s functional site ( Fig 1 ) [36] . Briefly , key residues important for catalytic activity ( Fig 1 , black residues ) are identified using a combination of the Catalytic Site Atlas ( CSA ) [64] or literature research and structure alignment . All residues within 10 Å of each key residue ( Fig 1A , gray spheres ) are identified and extracted from the full protein sequence and aligned N- to C- terminus to create an active site signature ( Fig 1B ) . Fragments containing three residues or fewer are removed from the active site signature as they lack sufficient length for statistical significance . Multiple signatures are aligned to create an active site profile ( ASP ) , characterizing the active site features of all proteins in the group ( Fig 1B ) . An ASP score is calculated indicating the residue variation among the signatures in the profile [36] . ASP scores range from -0 . 5 to 1 . 0 , where 1 . 0 indicates perfect alignment and conservation across all signatures . The Deacon Active Site Profiler ( DASP ) is a tool that uses ASPs to search sequence databases for proteins with fragments similar to the active site motifs [37 , 38 , 55] . The ASP is separated into aligned motifs which contain contiguous fragments within the signatures ( Fig 1 , colored fragments ) . For each aligned motif , a position specific scoring matrix ( PSSM ) [65] is calculated , detailing the propensity for specific residues to appear in each position of the motif , normalized to the background frequency of each residue in the database [35 , 37 , 55] . Starting with the longest motif , a sliding window search is performed along each sequence in the database . A p-value defining the similarity between the ASP motif and the sequence fragment is calculated for every position; the most significant p-value indicates the best match between a fragment and the motif for a given protein . All motifs are searched in this manner to identify the best matching fragment with the caveat that fragment matches cannot overlap . For each protein sequence , the p-values for each “best matched” fragment are combined using QFAST [66] to calculate a DASP search score . This score represents the probability a given sequence contains the fragments matching the ASP motifs by chance . This process is completed across all protein sequences in the database , such that each protein is associated with a DASP search score indicating the statistical significance of the match between the protein fragments and the ASP fragments . To efficiently perform iterative DASP searches , a new version of DASP named DASP2 was developed to support variable input formats and decrease GenBank search times . DASP2 testing demonstrated DASP and DASP2 return essentially identical data , but DASP2 searches are significantly more efficient [53] . Additionally , expanding the supported input formats allows the identified fragments of one search to be used as the input of the next search , opening the door for iterative database searches used in the MISST process . While these changes do not alter the search results , the latest version supports efficient , iterative GenBank searches which are critical to this work . Previously , Leuthaeuser and coworkers demonstrated that clusters identified using pairwise active site similarity networks often share more functional details than those identified using full sequence or full structure similarity networks [54] . Building on this , the Two Level Iterative clustering Process ( TuLIP ) was developed to identify functionally relevant groups of protein structures based on active site similarity . Validation was previously performed on the enolase and GST superfamilies . Results demonstrated significant correspondence to known functional groups [52] . Initially , an all-by-all network was created using the 47 non-redundant Prx structures in which each node represents one protein structure and each node pair is connected by an edge representing a pairwise ASP score . The edge threshold was incremented and the MCL clustering algorithm [67] was applied until distinct subnetworks form , such that no edges connect subnetworks to each other . At this point , an ASP is created for each subnetwork and used to search the PDB with DASP2 . If the PDB search using the subnetwork’s ASP identifies only itself ( the proteins within the subnetwork ) at significant DASP search scores , it is defined as “functionally relevant” and removed for further analysis . For all subnetworks which are not identified as functionally relevant groups , the edge score is incremented and the process repeated . This iterative clustering process is continued until each protein is either part of a functionally relevant group or separated out as a singlet , which signifies the end of the strict clustering stage . The full iterative approach is then repeated for the relaxed stage: a fully connected network is formed from all singlets and the edge threshold is incremented to form subnetworks which are used to search the PDB for identifying functionally relevant groups . The relaxed stage uses more relaxed parameters for evaluation of the functional relevance of each subnetwork . Again , any subnetwork that meets the relaxed parameters is defined functionally relevant and is removed . The edge threshold is then incrementally increased . Once all proteins are either members of a functionally relevant group or singlets , TuLIP is complete . Utilizing two stages of iteration allows identification of functionally relevant groups whose relationship might be obscured by the more coherent groups identified with the strict clustering parameters . TuLIP is performed only on proteins of known structure . A single DASP2 search can expand the group into sequence space; however , the identified sequences are limited by the diversity of the search ASP , which , in turn , is limited to those sequences represented in the structure database . To expand functionally relevant clustering , so that the diversity of sequences and functionally relevant groups are fully comprehended , the Multi-level Iterative Sequence Searching Technique ( MISST ) was developed ( Fig 4 ) . This process utilizes iterative DASP2 GenBank searches to populate each TuLIP group with sequences sharing active site similarity , thus increasing robustness of the search ASP . Additionally , a novel PSSM Analysis method identifies when and how a MISST group should be subdivided into distinct functionally relevant groups . To initiate MISST , an ASP is created for each TuLIP group; each ASP is used in an initial DASP2 search , Search0 , of GenBank ( Fig 4A ) . Given the limited representation in the structure database , the active site diversity of these initial ASPs is limited; thus , the goal of Search0 is to create a more robust ASP better representing each group’s functional site diversity . A DASP2 score of ≤1e-12 was chosen as the threshold for inclusion of sequences in the more robust profile . Previous work had identified ≤1e-8 or ≤1e-10 as “generous” and “trusted” DASP score thresholds in a single search of Prx subgroups [35] . Subsequent work on the enolase superfamily demonstrated that cross-hits ( sequences identified as members of more than one functional group ) decreased to zero at ≤1e-13 in the 26 subgroups and families of the enolase superfamily [52] . Balancing performance , precision and recall on the enolase superfamily , a “trusted” score threshold of ≤1e-12 was identified and is therefore used here . A detailed analysis and discussion of these score thresholds is provided in Supplemental Methods in S3 File . An ASP is created from the pseudo-signatures of sequences identified with DASP2 search scores more significant than the score threshold . To create each pseudo-signature , fragments identified in each sequence as matching each ASP motif are concatenated ( in length order , longest to shortest ) . The pseudo-signatures are aligned to create a new ASP for each group; each ASP is then used as input into a second DASP2 search of GenBank , termed Search1 ( Fig 4A ) . At this point , an iterative process of sequence acquisition and data analysis begins for each TuLIP group . The DASP2 score threshold for Search1 and beyond is ≤1e-14 , rather than ≤1e-12 used at Search0 . 1e-14 was determined to be a more appropriate threshold because the ASPs become more robust and the DASP2 scores of known true positives shift to more significant scores with the addition of new sequences at each search ( S3 File , S6 Fig ) . Notably , there is no score shift after Search 1 as the average DASP search score for true positives does not improve between Search 1 and Search 2 or beyond ( S3 File , S6 Fig ) . Beginning with Search1 , each group is analyzed against two self-identification criteria to determine if the group is self-contained and stable ( Fig 4A ) . This approach to identifying functionally relevant groups is novel as groups are not identified based on a specific threshold , but instead all groups are required to pass a self-identification test to be considered functionally relevant . In this way , groups which are functionally distinct and easier to identify can be fully identified in few iterations , while groups sharing similar active site features with other groups may take more iterations to be distinctly identified . This approach prevents the simultaneous subdivision of some groups and combination of other groups that is prevalent in most clustering . A group is complete and removed from the iterative process when a GenBank search demonstrates self-identification; that is , all inputs are identified with significant DASP2 search scores and nothing else is identified with significant DASP2 search scores , within a small range of error . Quantitatively , two metrics define the self-identification criteria: percent new hits and percent inputs hit . The first metric tracks whether the search identified sequences not identified in the previous search: if ≤15% of the sequences identified at a score threshold of ≤1e-14 are “new” ( not identified ≤1e-14 in the previous search ) , the group meets this metric . The second metric evaluates whether the proteins used as input were identified in this search . To pass , ≥95% of input proteins must be identified at a DASP2 score threshold of ≤1e-14 ( see Supplemental Methods in S3 File for more detail ) . A MISST group is removed from the iterative process if it meets both metrics ( Fig 4A ) . The values of these two parameters were chosen based on data from the Prx superfamily , but will be evaluated on other superfamilies in the future . For completed groups in the current data set , percent new hits averaged 5 . 4% with a range from 2 . 2% to 11 . 3% and incomplete groups averaged 63 . 4% with a range from 29 . 9% to 98 . 2% . Similarly , percent inputs hit averaged 99 . 7% with a range from 99 . 5% to 100% for complete groups and averaged 66 . 8% with a range from 50 . 9% to 99 . 3% for incomplete groups . Preliminary analysis with other SFLD superfamilies ( enolase , crotonase , and radical SAM ) suggests these parameters are relatively generalizable , but comprehensive testing is required on more data sets . Once all groups meet the self-identification criteria , a final ASP is constructed from each MISST-identified group and used to search GenBank one additional time to obtain the final MISST search results for that superfamily . The ASPs of completed searches can additionally be used at any future time to identify new sequences recently added to GenBank . At each iteration , all groups that do not pass self-identification criteria are evaluated using the following protocol ( Fig 4A , gray box ) : These three steps are completed for each MISST group at each search iteration ( Fig 4A ) . After completion of these three steps , PSSM Analysis ( Fig 4C; see subsequent section ) is performed to determine potential group subdivision . Position Specific Scoring Matrix ( PSSM ) Analysis is a novel approach using PSSMs to determine whether a group of protein sequences contains more than one identifiable functionally distinct group based on residue similarity within the active site signatures . In this way , MISST groups that contain multiple functionally-distinguishable families can be appropriately subdivided . PSSM Analysis begins by placing every protein identified by one group’s search into order of magnitude “bins” based on the DASP2 search score at which they were identified . Each order of magnitude is considered a bin , such that proteins with DASP search scores >1e-9 and ≤1e-8 are placed into the bin labeled “8” ( Fig 4Ci ) . All proteins with DASP search scores ≤1e-25 are placed into the bin labeled “25 . ” Bin-specific ASPs are created from the proteins in each bin ( using the pseudo-signatures described previously ) and a PSSM [65] is calculated for the each ASP , resulting in 18 bin-specific PSSMs ( Fig 4Ci ) . The PSSM values are based on the count of each residue in each position of the profile , normalized to the overall count of that residue in the database . To identify the similarity between proteins in each pair of bins , a modified Pearson correlation coefficient is calculated pairwise between bin-specific PSSMs . A PSSM is a two-dimensional array , the first dimension representing each of the 20 amino acids; the second dimension representing a position in an ASP ( positions in an ASP are indicated by arrows in Fig 1B ) . The standard Pearson correlation coefficient is calculated between analogous columns of a pair of PSSMs . To get the overall comparison between two PSSMs , column correlations must be summarized , but averaging correlation coefficients can lead to bias [70] . Therefore , a Fisher transformation is executed prior to computing the average . Due to the nature of the transformation , all coefficients >0 . 9999 are set equal to 0 . 9999 , and the Fisher transform is performed to produce a z-score for each column . The z-scores are then averaged across all columns and back transformed to r , producing the modified Pearson correlation coefficient , which correlates the active site similarity between the proteins in two bins . To define when a group should be subdivided , a fully connected network is created , with each node representing proteins in a scoring bin ( from 8 to 25 ) and each edge representing the pairwise correlation coefficient between bin-specific PSSMs ( Fig 4Cii ) . Beginning at the lowest correlation value ( rounded to two decimal places ) , a filter threshold is applied to the network , removing all edges below the threshold . The filtered network is clustered using MCL clustering [67] to produce subnetworks ( Fig 4Ciii ) . If distinct subnetworks are formed , in which no edges connect the two ( or more ) subnetworks to each other , the subnetworks are evaluated based on the following criteria to determine if they might represent functionally distinct groups: 1 ) each subnetwork must contain at least three nodes; and 2 ) the nodes ( bins ) must represent contiguous DASP2 scores ( e . g . 8 , 9 , and 10 rather than 8 , 10 , and 12 ) . If the subnetworks meet both criteria , the subnetwork containing the nodes with the least significant DASP2 scores is removed as a potential functionally relevant group , while the remaining subnetwork is subdivided further . If a subnetwork does not meet both criteria , it is not identified as a potential functionally distinct cluster . The filter threshold is increased by 0 . 02 each iteration and the clustering process is repeated . At the edge threshold of 0 . 98 , PSSM Analysis is completed . If a group has subdivided , ASPs are built from the pseudo-signatures of proteins in each subnetwork and used in the subsequent MISST iteration and search of GenBank ( Fig 4A ) . If the network reaches the 0 . 98 edge threshold and no subnetworks have been identified , an ASP is created from the pseudo-signatures of the sequences with DASP2 search scores ≤1e-14 . MISST iterations continue , as outlined in Fig 4 . Once all groups pass self-identification criteria , a final DASP2 search of GenBank is completed for each MISST-identified group . In this work , these final searches were completed in March 2016 . Cross hit analysis then identifies the number of shared sequences between the six groups identified at the significance threshold of ≤1e-14 . Cross-hits are identified and removed using the same procedure utilized during the MISST process ( Fig 4A ) . The final list of all proteins identified in each MISST group along with their DASP2 search score , SFLD annotation , and pseudo-signature can be found in S2 File . The results of these searches were compared to the expert-identified subgroups using quantitative methods previously used to evaluate other similar processes [49 , 50 , 59 , 62] . To calculate these measures , the MISST groups were compared to the sequences in the SFLD as of March 6th , 2016 ( http://sfld . rbvi . ucsf . edu/django/ ) . Each of the 6 MISST groups contained the majority of one subgroup; consequently , the analysis was completed using a 1-to-1 correspondence of MISST group to known functional groups ( defined in Table 1 ) . To evaluate how well our clusters compared with known functional clusters , measurements of purity , edit distance , and VI distance were performed , as previously described [50] . Additionally , the combined performance metric suggested by Orengo and colleagues [49] was calculated as well as the F-measure , which is the harmonic mean of precision and recall [59] . Details of these metrics are provided in S3 File . The consensus Prx motifs for each group were determined based on the conservation of residues in each position of the motif according to the following rules: 1 ) if the three most conserved residues make up ≤97% of that position , an x is used in the consensus sequence for that position , and 2 ) for all other positions , all residues identified in ≥3% of the MISST group sequences are annotated in the consensus sequence . Conservation graphs were built using Weblogo [61] . A representative network ( RepNet ) was created for all 38 , 739 sequences identified by the six MISST groups in the final searches using Cytoscape [63] . Using CD-Hit [68 , 69] , 1 , 369 clusters were identified where all members share 55% sequence identity with the representative protein . Each representative is a node in the RepNet and the edges connecting the nodes are pairwise BLAST scores between each pair of representatives . The nodes are colored by the MISST group the proteins were identified by . | Peroxiredoxins ( Prxs ) are a large , ubiquitous superfamily of proteins that are arguably the most important reductants of peroxide in biological systems . These proteins are involved in a diverse array of essential cellular functions , including peroxide reduction , signal transduction , circadian rhythms , chaperone function and apoptosis . Previously , Prxs have been classified multiple ways , based on biological role and evolutionary analysis . A more detailed expertly curated analysis identified six functionally relevant Prx classes and identified over 3500 proteins in these six classes; this set provides a validation for molecular function annotation methods . It is well-known that automated molecular functional annotation for individual protein sequences is difficult without detailed manual curation . In this work , we address this deficiency in available technologies by presenting a novel iterative method , MISST , for agglomeratively identifying superfamily members and clustering them into functionally relevant groups . Using this potentially automatable approach , 38 , 739 Prx sequences were identified from GenBank . MISST identified six functionally relevant clusters from these sequences , matching those previously identified by experts . Key mechanistic determinants and organismal distribution are explored . This analysis provides a significantly more complete understanding of this biologically important protein superfamily; the method lays a foundation for automated functionally relevant clustering of the protein universe . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"taxonomy",
"database",
"searching",
"phylogenetics",
"data",
"management",
"protein",
"structure",
"sequence",
"motif",
"analysis",
"protein",
"structure",
"databases",
"sequence",
"similarity",
"searching",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"computer",
"and",
"information",
"sciences",
"bioinformatics",
"proteins",
"biological",
"databases",
"evolutionary",
"systematics",
"molecular",
"biology",
"biochemistry",
"sequence",
"databases",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"macromolecular",
"structure",
"analysis"
] | 2017 | An Atlas of Peroxiredoxins Created Using an Active Site Profile-Based Approach to Functionally Relevant Clustering of Proteins |
The plant hormone auxin regulates many aspects of plant growth and development . Recent progress in Arabidopsis provided a scheme that auxin receptors , TIR1/AFBs , target transcriptional co-repressors , AUX/IAAs , for degradation , allowing ARFs to regulate transcription of auxin responsive genes . The mechanism of auxin-mediated transcriptional regulation is considered to have evolved around the time plants adapted to land . However , little is known about the role of auxin-mediated transcription in basal land plant lineages . We focused on the liverwort Marchantia polymorpha , which belongs to the earliest diverging lineage of land plants . M . polymorpha has only a single TIR1/AFB ( MpTIR1 ) , a single AUX/IAA ( MpIAA ) , and three ARFs ( MpARF1 , MpARF2 , and MpARF3 ) in the genome . Expression of a dominant allele of MpIAA with mutations in its putative degron sequence conferred an auxin resistant phenotype and repressed auxin-dependent expression of the auxin response reporter proGH3:GUS . We next established a system for DEX-inducible auxin-response repression by expressing the putatively stabilized MpIAA protein fused with the glucocorticoid receptor domain ( MpIAAmDII-GR ) . Repression of auxin responses in proMpIAA:MpIAAmDII-GR plants caused severe defects in various developmental processes , including gemmaling development , dorsiventrality , organogenesis , and tropic responses . Transient transactivation assays showed that the three MpARFs had different transcriptional activities , each corresponding to their phylogenetic classifications . Moreover , MpIAA and MpARF proteins interacted with each other with different affinities . This study provides evidence that pleiotropic auxin responses can be achieved by a minimal set of auxin signaling factors and suggests that the transcriptional regulation mediated by TIR1/AFB , AUX/IAA , and three types of ARFs might have been a key invention to establish body plans of land plants . We propose that M . polymorpha is a good model to investigate the principles and the evolution of auxin-mediated transcriptional regulation and its roles in land plant morphogenesis .
In angiosperms , the plant hormone auxin regulates many aspects of growth and development such as axis formation during embryogenesis [1] , initiation of leaf primordia at the shoot apical meristem [2] , root development [3] , and tropic responses to light or gravity [4 , 5] . A major auxin-signaling pathway is transcriptional regulation mediated by a co-receptor consisting of TRANSPORT INHIBITOR RESPONSE1/AUXIN SIGNALING F-BOX ( TIR1/AFB ) and AUXIN/INDOLE-3-ACETIC ACID ( AUX/IAA ) . In addition , AUXIN BINDING PROTEIN 1 ( ABP1 ) , which is evolutionarily conserved from charophyte green algae to seed plants , is known as an extracellular auxin receptor involved in rapid re-orientation of microtubules [6–9] . Studies in Arabidopsis and other angiosperms have revealed that auxin perception by the TIR1/AFB-AUX/IAA co-receptor triggers transcriptional regulation mediated by the AUXIN RESPONSE FACTOR ( ARF ) transcription factors , which directly bind to cis-elements ( auxin responsive elements , or AuxREs ) of auxin responsive genes and positively or negatively regulates the expression [10] . In the absence of auxin , AUX/IAAs bind to ARFs through their C-terminal regions , called domains III/IV , of the respective proteins and the complex represses the expression of auxin responsive genes [11 , 12] . Auxin functions as “molecular glue” that stabilizes the interaction between the F-box protein TIR1/AFB and the transcriptional repressor AUX/IAA [13 , 14] . This interaction promotes ubiquitination of AUX/IAA by the ubiquitin ligase complex that contains TIR1/AFB and subsequent degradation of AUX/IAA by the 26S proteasome [15] , which liberates the ARFs and allows them to play their roles in transcriptional regulation . In Arabidopsis , an ensemble of 29 AUX/IAAs and 23 ARFs is believed to regulate various auxin responses [16 , 17] . However , the high level of genetic redundancy of these transcription factors and the complex body plan composed of various organs make it difficult to depict a comprehensive picture of auxin regulatory events consisting of interactions and feedback among multiple factors . Auxin responses are also observed in basal land plant lineages , such as the bryophytes ( liverworts , mosses , and hornworts ) , and green algal lineages related to land plants , the charophytes [18] . Whole genome sequencing approaches of the moss Physcomitrella patens and the lycophyte Selaginella moellendorffii , a member of the basal lineage of vascular plants , have revealed that these two species have orthologues of basic components for auxin-mediated transcriptional regulation with relatively lower redundancy than observed in flowering plants [19 , 20] . Additionally , it was reported that P . patens possesses the auxin perception mechanism mediated by TIR1/AFB and AUX/IAA , which regulates the chloronema-caulonema transition and rhizoid formation [21] . On the other hand , there is limited knowledge about auxin-mediated regulatory systems in green algae . It is generally accepted that the ancestor of land plants was closely related to charophytes [22] . A recent study on the draft genome sequence of Klebsormidium flaccidum , a filamentous charophyte lacking differentiation of specialized cells , reported the absence of TIR1/AFB , AUX/IAA , and ARF genes in its genome [23] , suggesting that auxin-mediated transcriptional regulation evolved subsequent to the divergence of Klebsormidium and the lineage leading to land plants . Marchantia polymorpha is a liverwort species belonging to the earliest diverging clade of extant land plants [22] and has long history as an experimental organism . M . polymorpha is a complex thalloid liverwort and spends most of its life cycle as a haploid flat thallus which grows apically and has distinct dorsiventrality . On the dorsal side , air chambers are regularly arranged [24] , and asexual reproductive organs , gemma cups and gemmae , are repeatedly formed [25] . On the ventral side of thallus , scales and rhizoids are produced [26 , 27] . M . polymorpha is dioecious and produces gametangiophores ( archegoniophores that produce egg cells and antheridiophores that produce sperms ) for sexual reproduction under certain environmental conditions [28 , 29] . Following fertilization , a diploid zygote develops into a multicellular sporophyte , or embryo , on archegoinophores , in which a set of cells undergo meiosis , resulting in a sporangium producing single-celled haploid spores [30] . The most common endogenous auxin , indole-3-acetic acid has been detected in M . polymorpha [31] . Application of exogenous auxin has revealed that auxin is involved in rhizoid initiation and elongation [32–34] , thallus growth [34 , 35] , regeneration from excised thalli [36] , and apical dominance [37] in M . polymorpha and in dormancy of gemma in Lunularia , a related liverwort genus [38] . Recently , M . polymorpha has received attention for its critical evolutionary position [39] . Molecular genetic tools including transformation techniques [40 , 41] , homologous recombination [42] , and CRISPR/Cas9-mediated genome editing [43] have been developed . By applying molecular techniques , we demonstrated that the auxin response reporter which expresses β-glucuronidase under the soybean-derived GH3 promoter ( proGH3:GUS ) specifically responds to auxin in a dose-dependent manner [34] , suggesting a possible conservation of regulatory machinery for auxin-mediated transcriptional activation in M . polymorpha . In this study , we performed an in silico search for auxin signaling factors in M . polymorpha , and demonstrate that it has a minimal but complete auxin-mediated transcriptional system . We also describe critical roles of AUX/IAA-mediated auxin signaling in M . polymorpha development throughout its life cycle . Moreover , we investigated protein interactions and functional diversities of ARFs in M . polymorpha . From these results , we discuss how various auxin responses are regulated using the minimal set of auxin signaling factors in M . polymorpha .
To investigate whether M . polymorpha has basic components of auxin signaling , genes for known auxin signaling factors were surveyed by BLAST searches against M . polymorpha transcriptome and genome databases . For AUX/IAA , only a single gene was found in the M . polymorpha genome ( MpIAA ) , that contains four conserved domains ( domains I to IV ) . In the N-terminus , the predicted MpIAA protein has a domain I motif ( LxLxL ) , which is predicted to bind directly to the transcriptional co-repressor TOPLESS ( TPL; Figs 1A and S1 ) [44] . In the C-terminus , MpIAA contains a domain II sequence , the AUX/IAA degron , and domains III and IV , which comprise a protein-protein interaction domain ( Figs 1A and S1 ) . In the domains III/IV of MpIAA , invariant lysine and acidic residues , which are shown to be important for AUX/IAA-ARF oligomerization in Arabidopsis , are conserved ( S1 Fig ) [45 , 46] . The amino acid sequence of MpIAA is considerably longer ( 825 amino acid residues ) as compared to AUX/IAA proteins of vascular plants , and contains an additional glutamine-rich stretch between domains I and II . To understand if this structure is evolutionarily conserved , we obtained partial sequence of AUX/IAA homologues from two other Marchantiales species , Conocephalum conicum and C . japonicum , via degenerate RT-PCR . The Conocephalum sequences also contain long glutamine-rich regions in their N-termini , implying that this region has a conserved function among at least the Marchantiales ( S1 Fig ) . In phylogenetic analyses , MpIAA resided in a clade with P . patens and S . moellendorffii distinct from the clade of Arabidopsis AUX/IAA sequences ( Fig 1B ) . The M . polymorpha genome encodes three ARFs ( MpARF1 , MpARF2 , and MpARF3 ) . The three MpARFs have a structure common to typical ARFs , containing a B3-like DNA binding domain ( DBD ) and a protein-protein interaction motif , called domains III and IV ( Figs 1A and S2 ) . Between the DBD and domain III , MpARF1 harbors a glutamine-rich region ( Fig 1A ) , which is known as a feature of activator ARF members [10] . Finet et al . ( 2013 ) classified 224 ARF proteins across diverse land plants into three clades: clade A , clade B and clade C [47] . Phylogenetic analysis showed that MpARF1 belonged to the clade A , which represents activator ARF proteins such as ARF5/MONOPTEROS ( MP ) of Arabidopsis ( Fig 1C ) . MpARF2 resided in clade B ( Fig 1C ) . In Arabidopsis , some ARF members in this clade have been shown to function as transcriptional repressor [10 , 48] . MpARF3 has a relatively longer DBD , and its domains III and IV show lower similarity to ARFs in the clades A and B ( Figs 1A , S2A , and S2B ) . MpARF3 mRNA contains the possible target sequence of microRNA160 ( S2C Fig ) . These features are common with Arabidopsis ARF10 and ARF16 , which possibly function as repressors [49] . Phylogenetic analysis supported the placement of MpARF3 in the clade C including Arabidopsis ARF10 , ARF16 , and ARF17 ( Fig 1C ) . In summary , MpARF1 , MpARF2 , and MpARF3 were phylogenetically classified into the clades A , B , and C , respectively , suggesting that three functionally diverged types of ARFs existed in the common ancestor of extant land plants . BLAST searches revealed that M . polymorpha genome harbors two genes that , respectively , exhibit high similarity to TIR1/AFB and a jasmonic acid receptor , CORONATINE INSENSITIVE1 ( COI1 ) , of Arabidopsis . One encodes a protein with 54% identity to Arabidopsis TIR1 and is phylogenetically classified in the TIR1/AFB clade ( named MpTIR1 , Figs 1D and S3 ) . The other sequence is 44% identical to COI1 , and phylogenetic analysis supported that this sequence belonged to the COI1 clade ( Fig 1D ) . These results suggest that M . polymorpha has only one TIR1/AFB auxin receptor . Taken together , M . polymorpha has all required components for auxin-mediated transcriptional regulation with minimal genetic redundancy . We also performed BLAST and HMMER searches using another auxin receptor , ABP1 , as query . ABP1 is broadly conserved in the green lineage including charophytes green algae [7] . However , to our surprise , no homologue of ABP1 was found in the M . polymorpha genome . Studies in angiosperms , including Arabidopsis , have shown AUX/IAA as the key component in auxin perception and signaling . Dominant mutations in domain II of AUX/IAA inhibit its auxin-dependent degradation and results in auxin resistance [50] . Because M . polymorpha has only the single AUX/IAA , we focused on MpIAA to investigate the mechanism and function of auxin signaling in M . polymorpha . To examine if MpIAA is involved in auxin signaling in M . polymorpha , we generated transgenic plants which expressed MpIAA with or without substitutions of two conserved proline residues in the domain II degron sequence into serines under the control of the MpELONGATION FACTOR 1α constitutively active promoter ( proMpEF1α:MpIAAmDII and proMpEF1α:MpIAA; Fig 2A ) [51] and analyzed their responses to exogenous auxin . In the absence of auxin , proMpEF1α:MpIAA plants were indistinguishable from wild type ( WT ) . In contrast , in the absence of exogneous auxin proMpEF1α:MpIAAmDII plants displayed morphological defects such as dwarfism and curling of the thallus . In the presence of high concentrations of exogenous naphthaleneacetic acid ( NAA ) , a synthetic auxin , proMpEF1α:MpIAA plants exhibited growth arrest and produced many rhizoids as reported for WT previously [34 , 52–55] , while proMpEF1α:MpIAAmDII plants were insensitive to NAA treatment ( Fig 2B ) . Thus , we conclude that MpIAA conveys auxin signaling through domain II function in a similar manner as that demonstrated in angiosperms . We observed that the expression level of transgenes in the lines obtained was significantly lower in proMpEF1α:MpIAAmDII than that of proMpEF1α:MpIAA ( S4 Fig ) , suggesting that high levels of expression of MpIAAmDII driven by the MpEF1α promoter might be deleterious . Therefore , we utilized an inducible system in which protein nuclear localization is modulated by a glucocorticoid receptor ( GR ) domain [56 , 57] , to investigate in more detail the function of MpIAA and to determine developmental processes where auxin-mediated transcriptional regulation is involved in the life cycle of M . polymorpha . We generated transgenic plants expressing a chimeric protein of MpIAAmDII C-terminally fused with GR under the control of the MpIAA promoter ( proMpIAA:MpIAAmDII-GR ) . In these transgenic plants auxin signaling should be repressed upon treatment with dexamethasone ( DEX ) . To confirm that this experimental strategy works in plants , we first introduced proMpIAA:MpIAAmDII-GR into the M . polymorpha lines harboring the auxin response reporter proGH3:GUS [34] . Twelve hours of exogenous auxin treatment increased GUS activity in proGH3:GUS control plants . DEX treatment applied to proMpIAA:MpIAAmDII-GR/proGH3:GUS plants completely repressed auxin-dependent expression of the GUS reporter gene , whereas DEX treatment did not affect auxin-induced GUS activity in proGH3:GUS plants ( Fig 3 ) . These results suggest that the presumable accumulation of MpIAAmDII protein in the nucleus represses an auxin-dependent transcriptional response as has been shown in Arabidopsis [58] . Taken together , the mechanism of AUX/IAA-mediated auxin response is conserved in M . polymorpha . We next examined the significance of MpIAA-mediated auxin signaling in the regulation of cellular morphology in M . polymorpha . We first analyzed the morphological and cellular responses of M . polymorpha thalli to exogenously supplied auxin . In the WT thallus , NAA treatment caused epinasty of thalli ( Fig 4A and 4B ) , protrusion of air chambers ( Fig 4D and 4E ) , and elongation of gemma cups ( Fig 4G and 4H ) . Quantification of cell parameters revealed directional expansion of dorsal epidermal cells ( S5 Fig ) , which therefore may be responsible for the above phenotypes . In the absence of DEX , proMpIAA:MpIAAmDII-GR plants responded to NAA as did WT ( Fig 4J , 4K , 4M , 4N , 4P , and 4Q ) . Treatment of proMpIAA:MpIAAmDII-GR plants with DEX alleviated the NAA-induced phenotypes ( Figs 4L , 4O , 4R , and S5 ) , whereas the same treatment to WT thallus did not ( Figs 4C , 4F , 4I , and S5 ) . These results suggest that cell expansion in dorsal epidermal tissue is promoted by auxin through MpIAA-mediated transcriptional regulation . The results described thus far suggest the central role of the sole AUX/IAA , MpIAA , in the auxin-mediated transcriptional regulation in M . polymorpha . To investigate spatiotemporal expression pattern of MpIAA , we generated transgenic plants expressing GUS reporter gene under the regulation of MpIAA promoter ( proMpIAA:GUS ) . High levels of GUS activity was observed throughout proMpIAA:GUS thalli including the gemma cups ( Fig 5A ) . Cross-sections of the proMpIAA:GUS thallus revealed that GUS staining was observed in all layers of thallus tissue , including dorsal air chambers , gemma cups and developing gemmae , internal parenchymatous tissue , and ventral scales and rhizoids ( Fig 5B and 5C ) . We also observed expression during the reproductive phase . Antheridiophores showed strong GUS staining ( Fig 5D ) , including GUS signal in antheridia and surrounding tissues ( Fig 5E ) . Compared with antheridiophores , archegoniophores showed relatively weak GUS staining , which was observed in the tips of digitate rays ( Fig 5F ) . Cross-sectional analysis revealed that intensive GUS activity was observed in archegonia including the egg cells ( Fig 5G ) . We also observed expression in the diploid sporophyte generation following fertilization of the haploid gametes . Strong GUS staining was observed in young sporophytes ( Fig 5H and 5I ) , and during sporophyte development , a gradient of GUS activity was evident along apical-basal axis . GUS staining of apical sporogenous tissue was relatively weaker , and diminished as sporophyte matured , whereas that of the basal region consisting of the foot and seta remained ( Fig 5I–5K ) . These GUS staining patterns in the sporophyte were observed in reciprocal crosses between proMpIAA:GUS and WT . These results suggest that MpIAA is widely expressed in both gametophyte and sporophyte generations with some tissue specificity , and that MpIAA-mediated auxin signaling would function in various tissues and organs in both generations of the life cycle . In proMpIAA:MpIAAmDII-GR plants , auxin responses could be repressed in a DEX-dependent manner ( see above ) . In order to investigate tissue- or stage-specific functions of MpIAA-mediated auxin signaling , we analyzed morphological phenotypes of proMpIAA:MpIAAmDII-GR plants caused by repression of auxin responses during various developmental stages . Without DEX treatment , proMpIAA:MpIAAmDII-GR plants developed normal thalli with gemma cups and regularly-arranged air pores on their dorsal sides ( Fig 6A ) . Gemma cups formed serrated structures on their rims and produced many gemmae from their bases ( Fig 6B and 6C ) . On the ventral side of thallus , ventral scales and numerous rhizoids were observed ( Fig 6D ) . Compared to the control condition , gemmalings grown in the presence of 10 μM DEX for 14 days showed severe growth inhibition ( Fig 6A and 6E ) . The eight gemmalings observed exhibited some or all of the following morphological abnormalities: five produced a cellular mass lacking dorsiventrality ( Fig 6F ) , four formed air pores ectopically , and five produced serrated structures , which were reminiscent of the gemma cup rim ( Fig 6G ) . Adventitious gemma-like multicellular bodies were frequently ( six of the eight ) formed as clusters on the surface of gemmalings ( Fig 6H ) . We could not find any ventral scales in the apical regions of gemmalings treated with DEX . These results suggest critical roles of MpIAA-mediated auxin signaling in gemmaling growth and differentiation , especially with respect to ventral structures . We then applied DEX treatment to 7-day-old thallus precultured in the absence of DEX . At this stage , gemmalings had developed into mature thalli with organogenesis in a proper dorsiventral topology . DEX treatment for 7 subsequent days conferred hyponasty resulting in V-shaped thalli ( Fig 6I ) . Although gemma cups were observed on dorsal side of the DEX-treated thalli , the cups were shallow and elongated along with the apical-basal axis , generating many serrated structures ( Fig 6J ) . At the bottom of gemma cups , in spite of the normal development of gemma primordia , mature gemmae did not develop ( Fig 6K ) . On the ventral side , the number of rhizoids was decreased , especially smooth rhizoids ( Fig 6L ) . These results suggest involvement of endogenous auxin and MpIAA-mediated transcriptional regulation in the harmonized growth of dorsal and ventral thallus tissues . MpIAA-mediated auxin response also functions in the development of gemma cups , gemmae and rhizoids . To investigate the role of MpIAA-mediated auxin signaling in gametangiophore growth , we started periodical DEX treatment to gametangiophores after they became visible ( smaller than 5 mm in height ) . DEX treatment resulted in short stalks , while , in the control condition , male and female gametangiophores had vertically elongated stalks ( Fig 6M–6P ) . Additionally , DEX treatment to proMpIAA:MpIAAmDII-GR plants compromised the tropic growth of gametangiophore stalks ( Figs 6M–6P and S6 ) . These results suggest involvement of MpIAA-mediated auxin signaling in both tropic and differential growth , which was reported as responses in gametangiophores to exogenous auxin application [59] . Finally , we investigated MpIAA-mediated auxin signaling in sporophyte development . Without DEX treatment , sporophytes developed on archegoniophores , producing yellow sporangia in approximately 4 weeks after crossing ( Fig 6Q and 6S ) . Periodical DEX treatment that was initiated on the day following crossing conferred developmental arrest of the sporophyte ( Fig 6T ) . We did not observe any mature sporangia 4 weeks after crossing ( Fig 6R ) . These results suggest that proper MpIAA-mediated auxin signaling is critical for sporophyte development . Phenotypic analysis of proMpIAA:MpIAAmDII-GR plants demonstrated that MpIAA-mediated auxin signaling regulates many aspects of growth and development of M . polymorpha . Therefore , our next question was how M . polymorpha generates various auxin responses using the simplified components for auxin-mediated transcriptional regulation . To tackle this question , we focused on the three ARF genes of M . polymorpha . We speculated that each MpARF might have a different specificity in protein-protein interactions and/or transcription activities . In Arabidopsis , interactions between AUX/IAA and ARF proteins have been examined by yeast two-hybrid ( Y2H ) assays and bimolecular fluorescence complementation ( BiFC ) [5 , 16] . To investigate if MpIAA and MpARFs have ability to form homo- or hetero-dimers , we first performed Y2H assays using C-terminal regions of MpIAA and MpARFs . Our Y2H assay showed that MpIAA could interact with all three MpARFs ( Fig 7 ) . Interactions between MpARFs were observed in all combinations except for MpARF3 homotypic interaction ( Fig 7 ) . We also examined the strengths of the observed Y2H interactions by quantitative measurements of 0β-galactosidase reporter activities ( S1 Table ) . MpIAA showed high β-galactosidase activities with all MpARFs , but there were significant differences among the combinations . MpARF1 showed a notably higher activity in combination with MpARF1 than with the other MpARFs , while MpARF2 did with MpARF3 , suggesting different affinities in the interactions among the three MpARFs . The interactions between MpIAA and MpARFs in planta were also examined by BiFC assay in Nicotiana benthamiana leaves . We analyzed all combinations of interactions between proteins fused to N-terminal and C-terminal halves of YFP under the condition where negative control experiments with empty vectors yielded no fluorescent signal . BiFC assays revealed that MpIAA interacted with MpIAA itself and all MpARFs . All combinations of MpARF-MpARF except for MpARF3-MpARF3 produced signal ( Fig 8 ) . These results confirmed protein interactions between MpIAA and MpARFs , as well as between MpARFs . To characterize the transcription activity of each MpARF , we performed transient transactivation assays using cultured tobacco BY-2 cells . The effector constructs carried full-length or the middle region sequences of MpARFs fused with the Gal4 DNA binding domain . The reporter vector expressed firefly luciferase ( F-Luc ) under the control of a promoter containing six repeats of the Gal4 binding site . As a transformation control , we prepared the plasmid carrying the Renilla luciferase ( R-Luc ) gene driven by the cauliflower mosaic virus 35S promoter ( Fig 9A ) . These constructs were simultaneously introduced into BY-2 cells by particle bombardment . Transcriptional activity was evaluated by the relative activity of F-Luc to R-Luc . Both the middle region and full-length sequences of MpARF1 showed approximately two-fold higher activity than the effector expressing only Gal4 DBD . In the case of MpARF2 , luciferase activity was lower than the control ( Fig 9B ) . These results suggest that MpARF1 and MpARF2 can function as a transcriptional activator and repressor , respectively . We could characterize MpARF3 as neither an activator nor a repressor from this experiment , as the middle region and full-length sequences of MpARF3 showed just slightly lower and higher luciferase activities , respectively , than the control ( Fig 9B ) . Taken together , our results suggest that M . polymorpha has three types of ARFs with different characteristics in their transcriptional activities .
Our results revealed that the liverwort M . polymorpha has a single TIR1/AFB , a single AUX/IAA , and three phylogenetically and functionally diverged ARF homologues . In Arabidopsis , it has been shown that AUX/IAA functions as a repressor through the interaction via domain I with the co-repressor TPL [44] . MpIAA has a conserved LxLxL motif in domain I ( S1 Fig ) . The M . polymorpha genome encodes a homologue of TPL , and it is suggested that MpTPL is involved in auxin-mediated transcription ( in an accompanying paper ) . In the present study , we showed that expression of domain II-modified MpIAA conferred an auxin-resistant phenotype and suppressed the transcriptional response to exogenously supplied auxin as monitored by proGH3:GUS ( Figs 3 and 4 ) . Additionally , it was reported that knock-down of MpIAA resulted in auxin hypersensitivity ( in an accompanying paper ) . These results suggest that auxin-mediated degradation of MpIAA , presumably promoted by MpTIR1 , is critical for transcriptional regulation . Our results also showed interaction between MpIAA and MpARFs through domains III/IV ( Figs 7 and 8 , S1 Table ) . Loss- and gain-of-function mutants of MpARF1 show auxin-resistance and hypersensitivity , respectively ( in an accompanying paper ) [43] . Taken together , these results suggested that M . polymorpha possesses an auxin-mediated transcriptional regulation system that involves AUX/IAA and ARFs . Previous genomic analyses revealed that the lycophyte S . moellendorffii and the moss P . patens have all the basic components for auxin-mediated transcription [19 , 20] , and that the filamentous charophyte alga K . flaccidum has none of the components [23] . Expressed sequences showing high similarity to the DBD and domains III/IV of ARFs were found in two other charophyte species , Coleochaete orbicularis and Spirogyra pratensis , respectively , that in some analyses represent the sister lineage to extant land plants [60 , 61] . Although it is still controversial whether aquatic ancestors of land plants had acquired the auxin-mediated transcriptional regulation , these data suggest that the origin of auxin responses using the three types of ARFs dates back to at least the last common ancestor of extant land plants . In comparison with AUX/IAAs of vascular plants , the predicted amino-acid sequence of MpIAA , is much longer , and contains a long glutamine-rich region between domains I and II , which is conserved at least in the Marchantiales ( Figs 1A and S1 ) . Glutamine-rich domains are known to activate transcription in eukaryotes [62] . Activator ARFs , including MpARF1 , also contain a glutamine-rich domain ( Fig 1A ) [10] . Since ARFs and AUX/IAAs also exhibit similarity in their C-terminal interaction domains , these genes likely evolved from a common ancestral gene . It is possible that the glutamine-rich domain of Marchantiales AUX/IAAs might be a remnant of the ancestral gene and may retain an unknown function in this lineage . We previously demonstrated that proGH3:GUS activity could reflect the sites of endogenous auxin responses [34] . Expression of MpIAA was observed in various tissues including those showing high proGH3:GUS activities , such as the base of gemma cups , stalks and lobes of archegoniophores , antheridia and developing sporophytes ( Fig 5 ) . The significance of MpIAA expression in these tissues was supported by observation of phenotypes of proMpIAA:MpIAAmDII-GR plants ( Fig 6J , 6K , and 6M–6T ) . These results suggest that auxin responses monitored by proGH3:GUS can be accounted for by MpIAA function . Interestingly , developmental defects of DEX-treated proMpIAA:MpIAAmDII-GR plants were also observed where no proGH3:GUS activity was detected , such as gemmalings and rhizoids ( Fig 6E–6I and 6L ) . This could be possibly due to a high threshold of proGH3:GUS expression in response to auxin , or limitation of this reporter only representing the transcriptional activity mediated by activator-type ARF ( MpARF1 ) . If the latter is the case , non-activator-type ARFs ( MpARF2 and MpARF3 ) could have important developmental roles in M . polymorpha . Auxin regulates many aspects of plant growth and development via modulating cell differentiation and expansion . In Arabidopsis , ARF10 belonging to the same clade as MpARF3 regulates cell totipotency in cultured cells [63] . In the moss P . patens , gain-of-function AUX/IAA mutants showed delayed caulonema differentiation from chloronema [21] . The present study showed that gemmaling produced undifferentiated cell mass by repression of MpIAA-mediated auxin signaling ( Fig 6 ) , suggesting that auxin-mediated transcription regulates cell differentiation in M . polymorpha . Consistently , transgenic plants expressing a bacterial auxin inactivating enzyme produced undifferentiated cell mass ( in an accompanying paper ) . With respect to cell elongation , various mutants of AUX/IAA , ARF , and TIR1/AFB genes in Arabidopsis show defects in cell expansion and tropic responses [4 , 5 , 64–67] . Our study revealed that MpIAA-dependent auxin signaling regulated directional elongation of epidermal cells and tropic responses ( Figs 4 , 6M–6P , S5 , and S6 ) . These results suggest that regulation of cell differentiation and expansion by auxin-mediated transcription was already present in the common ancestor of land plants . In Arabidopsis , it is reported that early-phase auxin-induced hypocotyl elongation occurs independently of TIR1/AFB-mediated transcription [68] . ABP1 has been proposed to be another auxin receptor that rapidly activates cell expansion in transcription-independent manner in angiosperms [6 , 69–72] . ABP1 is also found in green algae , although its function in these taxa remains to be elucidated [7 , 23 , 60] . To our surprise , no homologue of ABP1 was found in the M . polymorpha genome , suggesting that M . polymorpha lost ABP1 during evolution . This brings up new open questions , such as when was ABP1 function in auxin-mediated cell elongation established and what is the ancestral role of ABP1 in the plant and land plant lineages ? One of the important roles of auxin in plant development is axis formation . The present study revealed that the expression pattern of MpIAA exhibited a gradient along the apical-basal axis in the sporophyte , similar to the auxin-response reporter proGH3:GUS ( Fig 5I–5K ) . Repression of MpIAA-mediated auxin signaling caused arrest of sporophyte development ( Fig 6Q–6T ) . In the moss P . patens , it has been reported that the expression pattern of proGH3:GUS changes dynamically along apical-basal axis during sporophyte development , and that defects in auxin transport causes abnormal morphology of sporophyte [73 , 74] . In Arabidopsis , IAA12/BODENLOS and ARF5/MP are involved in formation of apical-basal axis in embryogenesis [75 , 76] . These results suggest that land plants would have common auxin-mediated mechanism for apical-basal axis formation during embryogenesis after fertilization . Past studies showed that in M . polymorpha excessive exogenous auxin treatment promoted of rhizoid formation both dorsal and ventral sides of gemmalings , and thus proposed involvement of auxin into dorsiventral patterning of thallus [32 , 34 , 35 , 55] . The present study revealed that repression of MpIAA-mediated auxin signaling inhibited development of rhizoids and ventral scales ( Fig 6 ) . These results at least suggest MpIAA-mediated auxin signaling promotes the development of ventral tissues . It would be intriguing to clarify whether auxin mediates it directly or through dorsiventral axis formation . In Arabidopsis , it is thought that the complex transcriptional regulation using 29 AUX/IAAs and 23 ARFs underlies robust auxin responses [16] . In addition , various combinations of TIR1 and AUX/IAA proteins form co-receptor complex with a wide range of auxin-binding affinities and show various auxin sensitivity of AUX/IAA degradation , which also contributes to complex auxin responses in angiosperms [77 , 78] . In this study , we demonstrated that the liverwort M . polymorpha regulates various developmental processes with a minimized auxin-mediated transcription system . Because M . polymorpha has only one AUX/IAA and one TIR1/AFB orthologue ( Fig 1 ) , it was expected that the variation in auxin responses could be attributed to the functional diversification of the three MpARFs . Our results showed that the three ARFs in M . polymorpha are phylogenetically diverged and have different transcriptional activities ( Figs 1C and 9B ) . Furthermore , different binding affinities of MpIAA observed with the three MpARFs ( S1 Table ) suggest different auxin responsiveness , and MpARF-MpARF interactions ( Figs 7 and 8 ) could add a higher level of regulation . This is also supported by the analyses of chimeric protein of TPL fused with domains III/IV of MpIAA and MpARFs ( in an accompanying paper ) . Recently , crystal structure analyses revealed that AUX/IAA and ARF proteins multimerize through domain III/IV , and that the DBDs of ARFs dimerize and function as a “molecular caliper” when they bind to palindromically orientated AuxREs [45 , 46 , 79] . Although the physiological significance of interactions between MpARFs thorough domains III/IV and potentially between DBDs is unclear , it is possible that M . polymorpha can regulate diverse auxin responses via a combination of protein interactions of functionally diverged ARFs . Still , would it be possible to explain such various auxin-response outputs only by the protein interaction variations ? Recent work shows that functionally diverged ARF proteins from Arabidopsis have little differences in their DNA-binding specificity [79] , raising the idea that the variation of ARFs may not contribute much to target specificities . Given the single-copy existence of the activator ARF in M . polymorpha , it might be more plausible that outputs are pre-determined depending on the respective cell types and that auxin just modulates switches via MpARF1 . This idea is consistent with the recently proposed model , where auxin is viewed as a signal that provides “impetus” to processes [80] . It is expected that investigation on how three MpARFs regulate pleiotropic auxin responses will provide insights into the mechanisms of eliciting the variety of auxin responses observed in land plants . The present study demonstrated that M . polymorpha has minimal but complete , relative to that known in flowering plants , auxin-mediated transcription system , regulating diverse morphological events including both cell expansion and differentiation . In addition to transcriptional regulation , various auxin responses can be generated by regulating auxin biosynthesis , metabolism , and transport . It is also necessary to investigate the regulation of these factors in M . polymorpha . We propose that M . polymorpha with a low genetic redundancy is a good model for investigating the evolution and mechanisms of morphogenesis controlled by auxin .
Male and female accessions of M . polymorpha , Takaragaike-1 ( Tak-1 ) and Tak-2 , respectively [28] , were maintained asexually . F1 spores generated by crossing Tak-1 to Tak-2 or proGH3:GUS #21 [34] , were used for transformation . Gametangiophore formation was induced by far-red irradiation as described previously [28] . M . polymorpha was cultured on half strength Gamborg’s B5 medium [81] containing 1% agar under 50–60 μmol photons m-2 s-1 continuous white fluorescent light at 22°C unless otherwise defined . A similarity search for M . polymorpha genes was performed using BLAST against transcriptome and genome databases from on-going project by US Department of Energy Joint Genome Institute ( http://www . jgi . doe . gov/ ) . The transcriptome data contained 3 . 0 × 106 reads by Roche 454 GS FLX and >1010 reads by Illumina Hi-Seq from >18 conditions/tissues in different growth stages . The genomic DNA was sequenced under the coverage of 26 . 7× and 54 . 0× by Roche 454 GS FLX and Illumina Hi-Seq , respectively . The protein sequences of MpIAA , MpARFs and MpTIR1 were aligned with sequences listed in S2 Table . Partial cDNA sequences of AUX/IAAs in C . conicum and C . japonicum were amplified by degenerate RT-PCR using the primer set , degenerate-IAA_L2 and degenerate-IAA_R1 . Primers used in this study are listed in S3 Table . PCR fragments were subcloned into pBC-SK+ and sequenced . These sequences were aligned using the MUSCLE program [82] implemented in Geneious software version 6 . 1 . 6 ( Biomatters; http://www . geneious . com/ ) with default parameters . For phylogenetic analysis we used the C-terminal region ( domain II-stop ) of AUX/IAA , the DNA-binding domain of ARFs , and full length sequences of TIR1/AFBs . Phylogenic trees were generated by PhyML program version 2 . 1 . 0 [83] implemented in the Geneious software using the LG model and four categories of rate substitution . Tree topology , branch length , and substitution rates were optimized , and the tree topology was searched using the nearest neighbor interchange method . Bootstrap values were computed from 1000 trials . HMMER search for ABP1 homologues was performed against a database of six-frame translation products derived from the M . polymorpha transcript database , using the hmmscan program in HMMER3 . 1 ( http://hmmer . org ) with the raw HMM file for ABP1 downloaded from the Pfam database ( http://pfam . xfam . org; ID: PF02041 ) . The coding sequence of MpIAA was amplified by RT-PCR using the primer set MpIAA_entry and MpIAA_stop , and cloned into pENTR/D-TOPO vector using the Gateway TOPO cloning kit ( Life Technologies ) . Mutations in domain II were introduced by PCR using the primer set mDII_L3 and mDII_R3 . Then the MpIAA and MpIAAmDII cassettes were transferred into pKIGWB2 using LR Clonase II ( Life Technologies ) according to the manufacture’s protocol , which generated proMpEF1α:MpIAA and proMpEF1α:MpIAAmDII constructs , respectively . To generate a construct for proMpIAA:GUS , the genomic fragment covering from 5 . 2 kb upstream of putative start codon to the 52nd codon was amplified using the primer set MpIAA_usEntry and MpIAA_R9 , and cloned into pENTR/D-TOPO vector . The resultant genomic fragment was transferred into pGWB3 [84] by LR Clonase II and translationally fused with GUS reporter gene . To generate a construct for proMpIAA:MpIAAmDII-GR , the genomic fragment covering 5 . 2 kb upstream region and the coding sequence of MpIAA was amplified using the primer set MpIAA_usEntry and MpIAA_nonstop , and cloned into pENTR/D-TOPO vector . Mutations in domain II were introduced as described above . The glucocorticoid receptor hormone binding domain ( GR ) was amplified from pOpOn2 . 1 [85] , and cloned into the AscI site of pENTR/D-TOPO vector . The resultant cassette containing genomic fragment of MpIAA fused with GR was transferred into pGWB1 [84] or pMpGWB201 . Transformation of M . polymorpha was performed as described previously [40] . Independent T1 lines were isolated , and single G1 lines from independent T1 lines were established by subcultivating single gemmae which arose asexually from single initial cells [25 , 86] . Plants grown from gemmae of G1 lines ( termed the G2 generation ) were used for experiments . Histochemical assays for GUS activity were performed as described previously [34] . GUS stained gemma cups and gametangiophores were embedded in 6% agar block and sectioned into ~100-μm-thick slices with LinearSlicer PRO 7 ( DOSAKA EM , Kyoto , Japan ) . NAA and DEX treatments were performed by submerging plants into half-strength Gamborg’s B5 liquid medium [81] containing 10 μM NAA and/or 10 μM DEX for 12 h . GUS activity was then measured by monitoring cleavage of the β-glucuronidase substrate 4-methylumbelliferyl β-D-glucuronide ( MUG ) as described previously [29] with some modifications . After adjusting the concentration of extracted protein to 10 μg/70 μl extraction buffer , 20 μl of methanol and 10 μl of 10 mM MUG were added . After incubation at 37°C for 30 min , 900 μl of 200 mM sodium carbonate was added to stop the reaction . Fluorescence ( 460 nm emission/360 nm excitation ) of liberated 4-methylumbelliferone ( MU ) was measured on Powerscan4 ( DS Pharma Biomedical , Osaka , Japan ) . In the vegetative phase , NAA and DEX treatments of proMpIAA:MpIAAmDII-GR plants were performed by growing on half-strength Gamborg’s B5 medium [81] containing 10 μM NAA and/or 10 μM DEX . In the reproductive phase , DEX treatment was performed by spraying 10 μM DEX solution every 1 or 2 days . For scanning electron microscopy , plant samples were frozen in liquid nitrogen and directly observed on a Miniscope TM3000 ( HITACHI , Japan ) . Measurement of size and aspect ratio of epidermal cells , and curvature of gametangiophores were performed using ImageJ ( http://imagej . nih . gov/ij/ ) from SEM and photographic images , respectively . For Y2H analyses the C-terminal regions of MpIAA ( aa 627–825 ) , MpARF1 ( aa 783–928 ) , MpARF2 ( aa 751–879 ) and MpARF3 ( aa 730–821 ) were cloned into pBTM116SBEN and pVP16PS , which were modified from pBTM116 and pVP16PS [87] , using EcoRI/BamHI and BamHI/NotI site , respectively . Resultant constructs were transformed into the L40 Saccharomyces cerevisiae reporter strain [87] , and transformants were selected on SD medium lacking tryptophan and leucine . Protein interactions were checked by histidine requirement or ONPG assays in the conventional method [87] . For BiFC analyses , C-terminal regions of MpIAA and MpARFs described above were cloned into pENTR/D-TOPO . Each insert fragment was introduced into pB4CY2 and pB4NY2 by the LR reaction . pB4CY2 and pB4NY2 were obtained from S . Mano , National Institute of Basic Biology , Japan . Preparation and infiltration of Agrobacterium cultures were performed as described previously [88] . Fluorescence from YFP ( observation , 520 to 560 nm; excitation , 515 nm ) was observed 20 to 21 h after infiltration . Fluorescent signals , chloroplast autofluorescence and bright-field images were captured using a confocal laser scanning microscope , FluoView 1000 ( Olympus ) . cDNAs encoding full-length or middle region ( MpARF1: aa 347–822 , MpARF2: aa 388–764 , MpARF3: aa 448–628 ) sequences of MpARF proteins were amplified with specific primers listed in S3 Table , and cloned into the vector to express them as a fusion protein with Gal4-DBD driven by the cauliflower mosaic virus 35S promoter [89] using the In-Fusion HD Cloning Kit ( Clontech ) . A reporter plasmid containing six repeats of the Gal4 binding site and F-Luc , and the transformation control plasmid carrying R-Luc driven by the 35S promoter were described previously [89] . These constructs were introduced simultaneously into cultured tobacco BY-2 cells by bombardment using Biolistic PDS-1000/He Particle Delivery System ( BIO-RAD ) . After 48 h incubation , F-luc and R-luc activities were assayed using the Dual-Luciferase Reporter Assay System ( Promega ) in accordance with the manufacture’s protocol . Luminescence was detected by Centro XS3 LB 960 Microplate Luminometer ( Berthold ) . Luciferase activity was normalized by protein concentration . Total RNA was extracted from 2-week-old thalli using TRIZOL Reagent ( Life Technologies ) . First-strand cDNA was synthesized from 0 . 5 μg of total RNA with ReverTra Ace reverse transcriptase ( Toyobo ) and oligo ( dT ) primer . PCR amplification of the transgene-specific sequence was performed using the primer set MpIAA_dN2 and attB2_R . PCR amplification of the cDNA encoding the EF1α was performed as described before [90] , and served as a control . These reactions were performed using the C1000 Thermal Cycler ( Bio-Rad ) . The genomic sequences from this article are available in DDBJ under the following accession numbers: AB981316 ( MpIAA ) , AB981317 ( MpARF1 ) , AB981318 ( MpARF2 ) , AB981319 ( MpARF3 ) , AB981320 ( MpTIR1 ) , and AB981321 ( MpCOI1 ) . | Auxin is an important plant hormone which regulates many aspects of plant growth and development , such as embryogenesis and directional growth in response to light or gravity . Recent molecular genetics advances in angiosperms revealed that transcriptional regulation is critical for auxin response . Although auxin response was also observed in charophytes , a class of green algae related to land plants , and bryophytes , little is known how plants acquired auxin signaling components during evolution . Recently , it was reported that a filamentous charophycean alga , Klebsormidium flaccidum , does not have auxin-mediated transcriptional regulation , suggesting its evolution around land adaptation of plants . Here , we revealed that a liverwort species , Marchantia polymorpha , has a simplified but complete auxin-mediated transcription mechanism compared with other land plant species . Despite the minimal auxin system , M . polymorpha exhibited pleiotropic auxin responses in morphogenesis including roles in organ differentiation and cell elongation . Phylogenetic and experimental analyses revealed that the three ARF transcription factors in M . polymorpha had different transcriptional activities and interacted in various combinations with different affinities . This suggests that the auxin-mediated transcriptional regulation was established in the common ancestor of land plants , and might be correlated with the change in body plans at this time in the evolution of land plants . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Auxin-Mediated Transcriptional System with a Minimal Set of Components Is Critical for Morphogenesis through the Life Cycle in Marchantia polymorpha |
Sex-specific traits that lead to the production of dimorphic gametes , sperm in males and eggs in females , are fundamental for sexual reproduction and accordingly widespread among animals . Yet the sex-biased genes that underlie these sex-specific traits are under strong selective pressure , and as a result of adaptive evolution they often become divergent . Indeed out of hundreds of male or female fertility genes identified in diverse organisms , only a very small number of them are implicated specifically in reproduction in more than one lineage . Few genes have exhibited a sex-biased , reproductive-specific requirement beyond a given phylum , raising the question of whether any sex-specific gametogenesis factors could be conserved and whether gametogenesis might have evolved multiple times . Here we describe a metazoan origin of a conserved human reproductive protein , BOULE , and its prevalence from primitive basal metazoans to chordates . We found that BOULE homologs are present in the genomes of representative species of each of the major lineages of metazoans and exhibit reproductive-specific expression in all species examined , with a preponderance of male-biased expression . Examination of Boule evolution within insect and mammalian lineages revealed little evidence for accelerated evolution , unlike most reproductive genes . Instead , purifying selection was the major force behind Boule evolution . Furthermore , loss of function of mammalian Boule resulted in male-specific infertility and a global arrest of sperm development remarkably similar to the phenotype in an insect boule mutation . This work demonstrates the conservation of a reproductive protein throughout eumetazoa , its predominant testis-biased expression in diverse bilaterian species , and conservation of a male gametogenic requirement in mice . This shows an ancient gametogenesis requirement for Boule among Bilateria and supports a model of a common origin of spermatogenesis .
Evolution of sexual reproduction , consisting of the origin and maintenance of sex , has been a central focus of evolutionary biology since the time of Darwin . The origin of sex has generally been simplified to the question of the origin of meiosis , which is known to have a single origin among all eukaryotes [1] , [2] . However , sexual reproduction in higher eukaryotes is more complex than meiosis alone , and has evolved independently in plants and animals . The fundamental component of animal sexual reproduction is gametogenesis , the differentiation of sexually dimorphic male sperm and female eggs . Unlike meiosis , which is required in both males and females , most other components of gametogenesis are sex-specific or sex-biased , such as sperm tail formation . These traits are subject to strong selective pressures from natural selection , sexual selection , and/or sexual antagonism [3] , [4] . Because of these selective forces , sex-biased reproductive-specific traits are known to diverge rapidly . Such patterns of rapid divergence are not only prevalent among morphological traits like male genitalia , but also extend to the molecular level , including DNA sequences , the expression profiles of sex-biased reproductive genes and regulatory pathways underlying sex determination . [5]–[17] . What remains unknown is to what extent features of sexual reproduction can be conserved . Most animals produce sex-specific gametes distinct in size , morphology , motility and development . Animal sperm are predominantly small and motile , with compact nuclei and often a beating flagellum , which are produced through a series of male-specific developmental steps called spermatogenesis . Eggs are usually large in size and immotile , and are produced through a distinct developmental process called oogenesis . The evolutionary origin of such dimorphic features of animal sexual reproduction is intriguing yet difficult to address experimentally since they left no trace in the fossil record . However , the identification of conserved male or female-specific gametogenic proteins across large evolutionary distances could uncover molecular traces of any ancient gametogenic machinery , providing evidence for a common origin of sexually dimorphic traits among animals . While reproductive proteins with conserved homologs in different phyla are not uncommon , most of them are involved in general cellular functions , and are hence also required for other non-reproductive processes [18] , [19] . Their sequence conservation likely results from their pleiotropic functional constraints ( i . e . additional functions in non-reproductive tissues ) rather than their reproductive functions . This is consistent with a model in which the pre-existing somatic cell machinery underwent reproductive specialization during the evolution of gametogenesis in multi-cellular ancestors . A few reproductive-specific proteins have more restricted roles in sexual reproduction in distant species . However they are either required in different sexes or different stages of sexual reproduction in distant lineages of metazoans , or functional information is only available for one metazoan lineage [20]–[25] . Furthermore , most of these proteins appear to be associated with common features of germ cells in both sexes , suggesting that the unique functions that differentiate germ cells from somatic cells are likely to be conserved in animals [26]–[29] . Like meiosis , germ cells are common to both sexes and are therefore not subjected to as strong selective pressures as sex-specific or sex-biased processes are . Besides meiosis and the specification of germ cells , most of the other components of gametogenesis appear to be sex-biased or sex-specific . Despite a large number of sex-specific gametogenesis proteins uncovered in the model organisms Drosophila , C . elegans and mice , no conserved male- or female-specific gametogenic factors common to all lineages of animals have been clearly demonstrated [30]–[33] . However , the major steps of dimorphic gametogenesis among animals are very similar . For example , the major steps in spermatogenesis consist of the development of germline stem cells , mitotic proliferation of spermatogonial cells , preparation for and entry into meiosis , meiotic divisions , and finally differentiation of haploid spermatids into highly specialized motile sperm . Most , if not all , of the developmental steps and the developmental sequence of those steps are similar among animals from different phyla [32] , [34] . These similarities raise the question of whether they evolved independently in different lineages by convergent evolution , or evolved from a single ancestral prototype . While the absence of a universal male- or female-specific reproductive factor is predictable due to the fast divergence of reproductive proteins and hence is compatible with the hypothesis of multiple origins of spermatogenesis and oogenesis , it does not exclude the alternative single-origin hypothesis . It remains possible that spermatogenesis and oogenesis each evolved from a single prototype , followed by the rapid divergence of most components of the reproductive machinery . However , a few core components of the ancient prototype critical for sperm or egg production could remain conserved . Identification of such conserved core components is the key to distinguishing between these two possibilities . Such ancient core reproductive components should fulfill the following criteria: present in most , if not all , of the major lineages of animals undergoing sexual reproduction; originated around the time when gametogenesis likely evolved; demonstrated conservation at the sequence , expression and functional levels in species from diverse animal phyla . The most stringent criteria for a conserved male or female gametogenesis factor requires a clear demonstration that these conserved components are only required for gametogenesis in one sex among animals , and not for any other processes , thus excluding any possibility that such factors are conserved due to essential functions outside of gametogenesis . A strong candidate for a conserved male gametogenic factor in animals appears to be BOULE , the ancestral member of the human DAZ gene family . The human DAZ gene family consists of a Y-linked DAZ gene and the autosomal DAZ-Like ( DAZL ) and BOULE genes , all of which share a conserved RNA recognition motif ( RRM ) and a more divergent DAZ repeat consisting of 24 amino acids rich in N , Y , and Q residues [35] . All DAZ family proteins studied so far appear to be restricted to reproduction [35] , [36] , and the DAZ gene is commonly deleted in men with few or no sperm [37] , [38] . Although no mutations in DAZL or BOULE have been shown to be responsible for infertility in men or women , homologs of DAZL and BOULE are required for fertility in other species , and over-expression of DAZ family proteins promote the differentiation of human embryonic stem cells towards the germ cell lineage [35] , [39]–[43] . Furthermore , a human BOULE transgene rescued partial testicular defects of fly boule mutants , suggesting functional similarity between these two distant homologs [44] . However , the only two metazoan Boule homologs whose function has been characterized in depth revealed opposite gametogenic requirements [39] , [40] . Loss-of-function phenotypes of Boule homologs in Drosophila and Caenorhabditis elegans reveal their divergent roles in reproduction , with boule required for male reproduction in flies and for oogenesis in the worm [39] , [40] . The prevalence of Boule homologs among other metazoan phyla remains unexplored , raising the possibility that Boule may have undergone adaptive evolution like many other reproductive genes and subsequently diverged at the functional level among different metazoan branches . While partial rescue of the fly boule mutant defect by a human BOULE transgene suggests functional conservation , this may only reflect similar biochemical properties of all DAZ family members [45] . Indeed , frog Dazl is able to partially rescue the Drosophila boule mutant , despite the fact that frog Dazl performs a reproductive function distinct from fly boule [46] , [47] . We sought to gain further insight into the metazoan evolution of Boule and to determine if it has a general conserved reproductive function , or also a conserved sex-specific function . We systematically examined the prevalence of Boule homologs in major animal phyla and also the molecular evolution of Boule in two distant bilaterian classes . To understand the functional evolution of bilaterian Boule , we surveyed the expression of Boule homologs in representative bilaterian species and determined the functional conservation of deuterostomian Boule through expression and genetic analyses of the mouse Boule homolog .
We first asked what other animal lineages might have Boule homologs besides insects , mammals , and nematodes . In order to distinguish Boule from homologs of other DAZ family members as well as other general RNA binding proteins , we established the signature features of Boule that would allow us to identify Boule homologs in distant lineages with confidence . We separately aligned the protein sequences of known Boule homologs among two distant metazoan groups , mammals and insects , and established a consensus sequence for the RNA recognition motif ( RRM ) in each group ( Figure S1 ) . To determine general features of the Boule RRM we aligned the mammalian and insect consensus sequences to each other and found a 92-amino acid consensus sequence . The most conserved residues were in the two RNP motifs ( PNRI ( V ) FVGG for RNP2 and DRAGV ( I ) SKGYGFV ( I ) for RNP1 ) that are known to be important for RNA binding in RRM proteins [48] . We also established a consensus sequence for the closely related Dazl RRM , and found it to be distinct from the Boule consensus sequence ( Figure S2 ) . Dazl homologs contain slightly different consensus sequences for both RNP2 ( VFVGGI ) and RNP1 ( KGYGFVSF ) , have distinct sequences surrounding the RNPs , and have a conserved deletion of two amino acids ( Figure S2 ) [35] . Interestingly , the mammalian Boule proteins appeared to share higher sequence similarity than insect homologs despite the fact that mammals have an additional Boule-like protein , Dazl , suggesting that the presence of Dazl did not relieve the selective pressure on Boule in any significant way . Not only is the sequence of the RRM highly conserved , but the proteins are similar in size , usually around 30 kDa , and contain a single RRM domain near the N-terminus [48] . While it is impossible to align all the exon-intron boundaries due to the extensive genomic divergence between distant species , we found that exon-intron structures spanning the region of the highly conserved RRM ( exons 2 , 3 , 4 , and 5 ) are conserved , except that Drosophila exons 3 and 4 are fused into a single exon ( Figure S1C ) . Thus , comparison of the mammalian and insect Boule genes reveals conservation not only in specific protein sequences , but also in aspects of the genomic structure underlying these sequences . Since sex-biased genes often undergo lineage-specific loss during evolution [13] , we assessed the prevalence of Boule homologs in each branch of metazoan evolution ( Figure 1 ) . Starting with the Boule RRM consensus sequence , we used Tblastn to search the genomes of species from major phyla representing the two clades of Bilaterians , deuterostomes and protostomes , for Boule homologs . Among deuterostomes , Boule homologs were identified in at least one species of every phylum ( Figure 1 ) : in Chordata ( human , Homo sapiens; mouse , Mus musculus; chicken , Gallus gallus; rainbow trout , Oncorhynchus mykiss; elephant shark , Callorhinchus milii; lamprey , Petromyzon marinus; ) , Tunicata ( sea squirt , Ciona intestinalis ) , Cephalochordata ( lancelet or amphioxus , Branchiostoma floridae ) , Echinodermata ( sea urchin , Strongylocentrotus purpuratus ) , and Hemichordata ( Acorn worm , Saccoglossus kowalevskii ) . Boule homologs were present in many protostomian species of the Ecdysozoa and Lophotrochozoa superphyla ( Figure 1 , ESZ and LTZ , Figure 2A ) . Boule was found in fruit flies ( D . melanogaster ) , mosquitoes ( Anopheles gambiae ) , lobster ( Homarus americanus ) , green shore crab ( Carcinus Maenas ) , wasp ( Nasonia vitripennis ) and nematodes ( C . elegans ) , representing the Arthropoda and Nematoda phyla ( Figure 1 , ESZ ) , and also in each phylum of the Lophotrochozoans such as Platyhelmintha ( flatworm , Schistosoma japonicum ) , Annelida ( leech , Helobdella robusta ) , and Mollusca ( snail , Biomphalaria glabrata ) ( Figure 1 , LTZ ) . Therefore , homologs of a known gametogenic protein—Boule—are present throughout both deuterostomes and protostomes . Next , we asked when Boule arose during evolution by determining whether Boule homologs are present in basal , non-bilaterian metazoans or beyond the animal kingdom in plants or fungi . Based on the consensus Boule features , we determined that Boule homologs are absent in fungi and plants , suggesting that Boule is restricted to the animal lineage ( Figure S3A , Figure 1 ) . We then explored the genomes of basal metazoan animal species and found that there is no Boule homolog in the most primitive animal , Trichoplax ( Figure S3B , Figure 1 ) . However , we identified a Boule homolog in the sea anemone , a species from the primitive Cnidaria phylum . Comparison of the consensus Boule sequence against the sea anemone genome ( Nematostella vectensis ) reveals two proteins with high similarity [49] . Surprisingly , the RRMs of both proteins contain characteristics of the Boule consensus sequence , while one of the sea anemone proteins has identical signature RNP1 and RNP2 motifs ( PNRIFVGG and GVSKGYGSVT ) to those of the Boule consensus domain ( Figure 3C ) . Furthermore , unlike the fused exons 3 and 4 in the Drosophila boule genomic structure , the sea anemone boule gene has separate exons 3 and 4 as in humans , suggesting that the ancestral Boule gene contained separate exons 2 , 3 , 4 and 5 that encoded the RRM . This gene ( XM_001637198 ) is predicted to encode a protein around 22 kDa , close to the typical size of Boule proteins . Hence , a Boule homolog is present in the sea anemone , a representative of Cnidaria ( Figure S3C , Figure 1 ) . A second sea anemone protein also has some similarity to the characteristic Boule RNP1 and RNP2 , but there are multiple differences in critical positions and a greater divergence from the Boule consensus sequence ( Figure S3C ) . Furthermore , the gene itself does not possess two conserved exon/intron junctions in the second half of the RRM domain that are present in all other species examined , including Drosophila . Therefore , the second protein is likely a more divergent duplicate of the ancient Boule gene , specific to the Cnidarian lineage . The sea anemone is one of the most primitive metazoan species that undergoes sexual reproduction . It has separate sexes , inducible spawning and external fertilization [49] . Our finding places the origin of the Boule gene prior to the divergence of Bilateria from Cnidaria , but likely after Trichoplax branched from the common ancestor of eumetazoans , making Boule one of the few ancient animal gametogenic proteins known so far . Further analysis of Boule homologs in other basal metazoan lineages could better pinpoint the origin of metazoan Boule . Dazl arose through a duplication of Boule , likely after protostomian and deuterostomian splitting , but the exact point of Dazl origin within deuterostome evolution has not been defined [35] . Homologs of Dazl have been identified in mammals , birds , reptiles , amphibians and fish [35] , [50] , [51] , but whether Dazl is present in other non-vertebrate deuterostomes is unknown . Using the Dazl RRM domain , we searched for Dazl homologs in the genomes of acorn worm from Hemichordata ( Saccoglossus kowalevskii ) , sea urchin from Echinodermata ( Strongylocentrotus purpuratus ) , lancelet from Cephalochordata ( Branchiostoma floridae ) , and sea squirt from Tunicata ( Ciona intestinalis ) . We could not detect any canonical Dazl homologs ( Figure 1 ) . The highest BLAST hit from those genomes were Boule homologs , suggesting that Dazl is not present in either non-chordate deuterostomes or primitive chordates , and is likely restricted to the vertebrate lineage . To further determine the origin of Dazl in vertebrate evolution , we searched the genomes of the jawless fish , lamprey ( Petromyzon marinus ) and could not identify a Dazl homolog ( http://genome . wustl . edu/genomes/view/petromyzon_marinus/ ) ( Figure 1 ) . Given that Dazl is present in bony fish such as zebrafish and medaka [52] , [53] , we then asked if Dazl is present in the cartilaginous fish , phylogenetically the oldest group of living jawed vertebrates . We searched the genome of the elephant shark ( Callorhinchus milii ) and found no evidence of a Dazl homolog , though a shark Boule homolog is present [54] . This analysis suggests that Dazl originated around the time of vertebrate radiation , likely in the ancestral lineage of bony fish ( Figure 1 ) . To further determine the evolutionary relationship of metazoan Boule homologs , we performed phylogenetic analysis of Boule homologs from the major animal branches , together with homologs of the other members of the DAZ family , Dazl and DAZ ( Figure 2B and 2C ) . Boule clearly represents the most ancient and widespread clade among the DAZ family members , present from sea anemone to human , whereas all Dazl and DAZ homologs can be clustered together in one branch . This is consistent with the distinct reproductive functions of DAZ and Dazl homologs , and the late arrival of Dazl in vertebrate evolution and DAZ in primate evolution ( Figure 2B and 2C ) [35] , [38] , [41] , [47] , [55] . Conservation of Boule homologs in major lineages of animals and their evolutionary relationship throughout animal evolution suggests that Boule is a fundamental component of eumetazoan reproductive machinery essential for the survival of most animal species . The ancient origin and widespread presence of such a reproductive gene is in stark contrast with the pervasive rapid evolution usually associated with reproductive genes , especially male reproductive genes [5] , [12] . The presence of Boule in various animals provided the rare opportunity to examine how selective forces shaped the molecular evolution of a reproduction-specific gene in distant lineages . We therefore examined Boule homologs that recently diverged from each other for any signs of adaptive evolution . We analyzed two separate groups of homologs to determine if Boule is under different selective pressure when Dazl homologs are present . We compared the entire boule coding sequences among seven Drosophila species ( D . melanogaster , D . sechellia , D . yakuba , D . virilis , D . erecta , D . willistoni and D . ananassae ) as well as eight representative mammalian species [56] . To determine if positive selection has played a role in Boule evolution , we compared the ratio of the rate of nucleotide changes that result in a non-synonymous amino acid substitution ( Ka ) to the rate of nucleotide changes that cause a synonymous amino acid substitution ( Ks ) . Positive selection is a process that favors the retention of mutations that are beneficial to the reproductive success of an individual . Neutral theory predicts that the rate of non-synonymous substitutions ( that by definition affect protein sequence ) is equal to the rate of synonymous substitutions . If a protein has evolved under positive selection , there are more non-synonymous substitutions ( Ka ) than synonymous substitutions ( Ks ) , and an accordingly high Ka/Ks ratio . If the protein evolved under purifying selection or negative selection , a process that removes deleterious alleles , there is a decrease or absence of non-synonymous substitutions , and therefore Ka/Ks is much smaller than that expected under neutral theory . A Ka/Ks greater than 1 is a strong indication of positive selection whereas only a Ka/Ks smaller than 0 . 1 usually suggests a role of purifying selection . Among all pairwise comparisons among Drosophila species , we found that non-synonymous substitutions ( Ka ) were not in excess of synonymous substitution ( Ks ) . Instead , Ka/Ks ratios for all pairwise comparisons were below 0 . 1 ( Table S1 ) , significantly lower than the ratio reported for rapidly diverging proteins [10] , [12] . Similarly , all pairwise comparisons among mammalian species revealed Ka/Ks ratios below 0 . 1 ( Table S2 ) , indicating that the presence of Dazl homologs in mammals had little impact on the selective pressure on Boule homologs . Furthermore , this suggests that positive selection was not the major force driving the evolution of Boule either in Drosophila or mammals . Instead , the low Ka/Ks ratio suggests that purifying selection was responsible for the strong functional constraint on the entire protein , making Boule an exception to the rapid evolution commonly seen in reproductive genes [5] , [9] . The prevalence and strong functional constraint of Boule throughout protostomes and deuterostomes suggests that Boule is likely a common reproductive factor with a critical function essential for the survival of bilaterian species . However the only Boule homologs functionally characterized exhibit divergent roles in reproduction , with Drosophila boule necessary for male reproduction and the C . elegans boule homolog , daz-1 , required for egg production [39] , [40] . Recently , the Boule homolog in the fish Medaka was reported to be expressed in both testes and ovaries [53] . Such divergent roles and expression during gametogenesis raised the question of what the ancestral function of Boule was , and whether the expression and function of Boule homologs might have diverged despite the high conservation of the functional motif . Since Boule function has only been examined in protostomes ( C . elegans and Drosophila ) , we reasoned that by determining Boule expression patterns in deuterostomes we could ascertain whether or not the expression or function of Boule is conserved among bilaterians . We chose two deuterostome species ( chicken and sea urchin ) from separate phyla and asked if Boule homologs are preferentially expressed in the testis or ovary . Like Drosophila and C . elegans , the sea urchin is also an invertebrate and has only the Boule gene , whereas chicken is a vertebrate with both Boule and Dazl . We identified homologs of Boule in chicken ( G . gallus ) and purple sea urchin ( S . purpuratus ) ( Figure 1 , Figure 2A ) [51] , [57] , and found that chicken Boule is expressed specifically in the testis , and is not present in ovaries or any other organs we examined ( Figure 3A , Figure S4 ) . However , chicken Dazl is expressed in both testes and ovaries , similar to mammalian Dazl [41] , [58] . The expression of the Boule homolog in sea urchin , a primitive deuterostomian species from the Echinodermata phylum , is also testis-biased and not expressed in any non-gonadal tissue ( Figure 3B ) . A transcript that lacks a complete RRM domain was detectable at low levels in ovary ( not shown ) . However , this ovarian transcript may not be functional and is likely the isoform previously reported in sea urchin ovary and eggs by in situ hybridization [57] . Together these results show that deuterostome homologs of Boule are also reproduction specific , like their protostome counterparts , but with a tendency toward testis-biased expression . Since the nematode Boule homolog is only required for ovarian function but not male gametogenesis , and Boule transcripts have been detected in the ovaries as well as in the testes of some other species [53] , [57] , [59] , we wondered if such ovarian expression in sporadic species is a lineage-specific phenomenon or if it is a common feature . Thus , we turned to the laboratory animal model , the mouse , for an in-depth gene expression and functional analysis . Although mammalian Boule is highly expressed in the adult mouse testis but not ovary , it is not known if Boule is expressed in the ovary during development [35] . In view of the different timing of meiotic initiation in female and male mammals , we determined the developmental expression profile of mouse Boule during both male and female embryonic gonadal development ( embryonic day 10 . 5 , E12 . 5 , E16 . 5 and E17 . 5 ) in comparison with adult gonads . We first characterized the entire mouse Boule genomic region and identified alternatively spliced isoforms ( Text S1 and Figure 3C ) . Using primers spanning all 12 exons , we found two major Boule transcripts , both of which were most highly expressed in the adult testis . The primary Boule transcript contained all 12 exons ( Bol1 ) and was expressed only in the adult testis , whereas a second transcript lacking exon 11 ( Bol2 ) was highly expressed in adult testes but also detectable at low levels in early embryonic gonads of both sexes and the adult ovary ( Figure 3C ) . We thus confirmed that the predominant expression of Boule during reproductive development is in the adult testis , including a testis-specific isoform , and also identified previously unreported low levels of Boule RNA in mouse ovaries . Together with previous findings , a total of seven out of eight bilaterian species examined ( human , mouse , cattle , chicken , fish medaka , sea urchin and fruit fly ) representing three different phyla express Boule in the adult testis [35] , [53] , [60] . Expression is in the same cell types ( spermatocytes and spermatids ) in the testes of the human , mouse and fish medaka , suggesting conservation of developmentally-regulated testicular expression of Boule in vertebrate animals [35] , [53] . The observation that Boule homologs show predominantly testis-biased expression in diverse species is consistent with a conserved male gametogenic function in bilateral animals . However , the oogenic requirement seen in C . elegans taken together with detectable levels of ovarian expression in several species suggests the possibility that an additional oogenic function is also conserved . Alternative Boule transcripts detected in mouse ovaries or embryonic gonads , albeit at much lower levels , could still play an important physiological function and therefore contribute to its sequence conservation . To ascertain if Boule is functionally conserved in deuterostomes and if ovarian expression of Boule is physiologically significant , it is necessary to examine the physiological function of Boule in deuterostomes . To determine if the male-specific requirement of Drosophila boule is functionally conserved among Bilateria , we set out to generate a mutation of a deuterostomian Boule homolog to investigate its physiological requirement . We used mice as a representative species of deuterostomes , and used gene targeting to delete the RNA binding domain , and thus disrupt the critical function of mouse Boule ( Figure 4 ) . We replaced exon 3 , which encodes a part of the RNA binding domain and is present in all Boule isoforms ( Figure 3C ) , with a lacZ-neo vector through homologous recombination in embryonic stem cells . The removal of exon 3 resulted in a deletion of the RNA binding domain and a frame-shift in the remaining transcript . This transcript is expected to produce a truncated BOULE protein missing both its RNA binding domain and the remaining C-terminal portion of the protein . Four chimeric mice were derived and correct homologous recombination as well as germline transmission of the Boule mutation was confirmed . Homozygote mice recovered from matings among heterozygotes were identified by genotyping and confirmed by Southern hybridization ( Figure 4B ) . We next determined if the mouse Boule mutation was a complete loss-of-function mutation . We performed Northern blot hybridization on RNA from the testes of wild-type , heterozygote , and homozygote mice with Boule cDNA as a probe and found that there are three Boule transcripts present in wildtype testes , all of which are absent in homozygous Boule mutants . Instead , a single novel transcript corresponding to the size of the predicted chimeric transcript consisting of the truncated Boule and beta-geo ( a transcript containing exon 1 , 2 and lacZ , Figure 4C ) is present . We further confirmed the absence of wild type Boule transcripts by the more sensitive RT-PCR and did not detect exon 3 in the mutant transcript . Instead we only detected a much larger PCR product spanning exon 2 and 4 in homozygotes ( Figure 4D ) . This large PCR product is absent in wildtype and contains the lacZ gene from the knockout vector . Hence we conclude that Boule expression is completely disrupted in the mutant , and we have established a loss-of-function allele in the mouse Boule homolog . Homozygote Boule mutants exhibited normal viability , growth and mating behavior ( Figure 5A ) . We recovered the expected number of homozygotes from heterozygote matings ( wild type∶ heterozygotes∶ homozygotes = 44∶79∶43 ) , indicating that there was no effect on survival . We next tested the fertility of mice homozygous for the Boule mutation to determine whether the Boule mutation affected male and/or female reproduction . Six homozygote males were each mated with two wild type females and individually produced no pups after four months . In contrast , wild type males sired at least two litters each during that time , suggesting that the homozygous Boule males were sterile . Female Boule homozygotes showed no obvious defects and were fertile , producing an average of 8 . 3 pups per litter ( 8 . 3±1 . 4 , n = 12 ) with heterozygote males , similar to wild type or heterozygote females ( 7 . 7±2 . 1; n = 7 for wildtype; 7 . 6±2 . 4; n = 18 for heterozygotes ) . Homozygote females continued to be fertile up to the oldest age tested ( 12 months ) . Thus , mutation of mouse Boule disrupts male reproduction but does not affect normal development , growth or female fertility , suggesting that mammalian Boule is required only for male reproduction , similar to fly boule but different from worm daz-1 . Similar physiological requirements of Drosophila and mouse Boule homologs suggest possible conservation of an ancient male gametogenic requirement . However , it is also possible that such similarity is a mere coincidence since out of hundreds of bilaterian species , only homologs from three species are functionally characterized , with one out of the three being functionally divergent . We reasoned that if mouse and Drosophila Boule function is conserved , then the specific reproductive defects of the loss-of-function mutations in both species should be more likely to be similar than if they had evolved independently by chance . We therefore determined whether mouse Boule and fly boule function within similar processes of male reproduction . In both flies and mice , the major reproductive organs are clustered together in the male reproductive tract . The tract consists of a pair of testes for sperm production; a pair of sperm storage/maturation organs ( the epididymis in mice and seminal vesicle in flies ) ; accessory glands for providing proteins , other nutrients and seminal fluid that accompany sperm migration and fertilization ( prostate , seminal vesicles and coagulating glands in mice and accessory glands in flies ) ; and a sperm transport duct used for sperm transportation and maturation ( vas deferens and urethra in mice and ejaculatory duct in flies ) ( Figure 5B ) [30] . While the major components of the male reproductive tract in mammals and insects appear to serve similar reproductive functions , it is not known if any components are evolutionarily related between vertebrates and invertebrates . In mouse Boule mutants , all the components of the male reproductive tract are present and intact ( compare Figure 5B and 5C ) , similar to that of the fly boule mutant [39] . Compared with wild type mice , the male reproductive tracts of Boule homozygous mutant mice were morphologically indistinguishable except for the testes , which are smaller by weight [61] . Hence the sterility defect is the result of a defect in the testis , similar to the sterility defect associated with the Drosophila boule mutation [39] , [61] . Further characterization of the reproductive defects revealed that mouse Boule mutant epididymides lacked mature sperm ( Figure 5E ) , and instead contained degenerating cells that were not seen in the wild type ( compare Figure 5D and 5F to Figure 5E and 5G ) . Therefore , the observed male sterility of the Boule homozygous mutant mice appears to be due to a complete absence of sperm in the epididymis . Next , we examined the developmental impact of the mouse Boule mutation on sperm production . While the overall testicular structure was normal and all the somatic cell types were present in mouse Boule mutants , the effect of mouse Boule mutation on sperm development was dramatic , with a complete halt of spermatogenesis inside all seminiferous tubules of the testis . Both mature sperm and developing elongating spermatids were entirely absent from the lumen of individual seminiferous tubules ( compare Figure 5H and 5J to Figure 5I and 5K ) . This indicates that the failure to produce sperm resulted from a major block in sperm production due to a global arrest of spermatogenesis prior to spermatid differentiation , similar to the spermatogenic defect seen in the testes of boule mutant flies [39] , [61] . Such similar global , spermatogenic-specific impacts of mutations in orthologs in divergent phyla is surprising and unprecedented , given the vastly different organization of testicular structure , type of spermatogenesis and differences in the contribution of hormonal control in mammals and insects . In Drosophila , spermatogenesis is cystic , where a single spermatogonial cell and its clonal descendants are encapsulated in a somatic cyst throughout sperm development . Though mouse spermatogenesis is acystic , we observed prominent ball-like structures containing degenerating cells in the mouse Boule mutant , resembling the degenerating cysts seen in the fly boule testis ( Figure 5I and 5K , arrowheads ) . Although the cyst structure is not present in mammalian spermatogenesis , descendant cells from a single spermatogonial stem cell remain connected with each other through cytoplasmic bridges during mouse sperm development , similar to that in Drosophila spermatogenesis [62]–[65] . This phenomenon could lead to the merging of multiple interconnected arrested spermatogenic cells in Boule mutant testes , resulting in such giant “cysts” with multiple nuclei . Despite the distinct modes of spermatogenesis in mice and Drosophila , the mouse Boule and fly boule mutations caused a remarkably similar and specific global arrest of spermatogenesis . Though further characterization of the developmental and cellular defects in mouse Boule mutant testes is needed to determine the full extent of similarity in developmental and cellular defects between mouse and Drosophila Boule mutants [61] , our data demonstrate a key physiological requirement of Boule in sperm development and conservation of its male reproduction function between two distant lineages of a protostome ( Drosophila ) and a deuterostome ( mouse ) .
Evolution of reproductive traits and genes is of the utmost interest to our understanding of the central questions in evolutionary biology such as speciation . However , the relatively rapid divergence of sex-biased reproductive genes in comparison with somatic cell proteins or non sex-biased reproductive proteins during evolution has made it difficult to study the evolution of sex-specific reproductive systems across extended evolutionary distances . Even though some reproductive genes are conserved beyond a given phyla , they are often also involved in other developmental processes . Such broad functionality compounds studies of their reproductive evolution because the selective pressures driving their evolution may be due to critical somatic functions , and not a reproduction-related function . The human DAZ family of reproductive genes , with homologs in diverse species , many of which are specifically expressed in reproductive tissues , are ideal candidates for the study of reproduction-specific gene evolution [35] , [38] , [40] , [41] , [46] , [53] , [60] . In particular , Boule , the ancestral gene member , is reproductive specific in flies and worms . We identified homologs of Boule in the major phyla of metazoans , reconstructed the evolutionary history of Boule , and began to determine its functional divergence . We found that Boule , unlike other reproductive proteins , has been maintained in all major phyla of bilaterian animals as well as in Cnidarians , but are absent in the most primitive animals ( the placozoan Trichoplax ) , fungi and plants ( Figure 1 ) . We found that Dazl homologs are only present in vertebrates , supporting the hypothesis that Boule is the ancestral member of the DAZ family [35] . Dazl homologs were absent in representative species of non-vertebrate deuterostomes and cartilaginous fish ( elephant shark ) , but were present in bony fish and tetrapod animals ( Figure 1 ) . This places the origin of Dazl after the divergence of bony fish from cartilaginous fish but before the arrival of tetrapod animals ( Figure 6 ) . On the other hand , the widespread presence of Boule in eumetazoan animals indicates that the ancient Boule gene was present as early as 600 million years ago in the Precambrian era , in the common ancestors of Bilaterians ( often called Urbilateria ) as well as eumetazoans ( Figure 6 and Figure 7 ) [66] , [67] . Interestingly , human BOULE has previously been shown to be able to function in Drosophila testes , and can even rescue meiotic defects of boule mutant flies , suggesting a conservation of a spermatogenesis-specific function [35] , [44] . However , the C . elegans boule homolog daz-1 is required only in oogenesis [40] , making it unclear whether such a transgenic replacement in the fly actually represents a legitimate functional conservation . Furthermore , both C . elegans and Drosophila are protostomes , so whether Boule is even required for reproduction , let alone restricted to spermatogenesis , in any deuterostome species was not known . Using mice as a representative deuterostome , we generated a Boule null allele to address this question ( Figure 4 ) . Boule is required only for male reproduction in mice ( Figure 5 ) , similar to insect boule , revealing not only a conserved function , but suggesting an ancient requirement of Boule in gametogenesis . Furthermore , the requirement of mouse Boule for male reproduction and its dispensability for female fertility suggests that low level expression of Boule in embryonic germ cells and adult ovaries is not essential for either the development of germ cells or the production of female gametes . Similarly , Drosophila boule , initially thought to be testis-specific , has also been found to be alternatively spliced and expressed in the ovary and even some somatic tissues at a low level , though loss-of-function similarly only causes male sterility [59] , [68] . Interestingly , this result shows that Boule has a spermatogenesis-specific requirement conserved in at least two distant lineages of bilateral animals , making it a strong candidate for a conserved male gametogenesis factor between Drosophila and mammals . Given that mouse Boule is required for sperm production like fly boule but different from C . elegans daz-1 , we propose that Urbilaterian Boule had an ancestral function in male gametogenesis which was lost during the evolution of the nematode lineage ( Figure 7 ) . This is consistent with the higher sequence divergence of the C . elegans daz-1 RNA binding motif than most other bilaterian Boule homologs ( Figure 2 ) . While we can not rule out the possibility that the similar male gametogenic requirement in mice and Drosophila is a coincidence and both evolved independently , the striking similarity in the reproductive defects of loss-of-function mutants of Drosophila and mouse Boule homologs ( male specific infertility , global arrest of spermatogenesis , absence of elongating spermatids and mature sperm ) argue against such a possibility . Furthermore , the predominance of testis-biased expression of Boule homologs among distinct bilaterian species ( Figure 3 ) , supports a model of an ancient male gametogenic function ( Figure 7 ) . It is important to note , however , that this model does not exclude the possibility of an additional ancestral ovarian function of Urbilaterian Boule . Since ovary expression of Boule is also prevalent among diverse animals , and C . elegans daz-1 is required in females , the ancestral Boule gene may have also played a role in oogenesis , which may have been subsequently lost in specific lineages . Our data does not rule out this possibility , and such an ancestral oogenesis function of Boule could be in addition to our proposed ancient spermatogenesis function . Further functional analysis in other lineages , including medaka where strong ovarian Boule expression has been observed , could help determine the more likely scenario [53] . Additionally , characterization of Boule homolog ( s ) in the sea anemone , an outgroup to the bilaterian lineage , could provide further insights into the ancestral roles of Boule . Whether or not an ovarian function of Boule is also conserved , our discovery that mammalian Boule is required only for sperm development like its fly counterpart is the first such demonstration of a conserved spermatogenesis-specific function in both lineages . While spermatogenesis occurs in the testes of different animal lineages , it is not known if either spermatogenesis or the testis itself is evolutionarily related between vertebrates and invertebrates . The fly testis , which is a single tube with a linear progression of spermatogenesis , appears different from the mouse testis , which is composed of many seminiferous tubules with a concentric progression of spermatogenesis from the periphery of the seminiferous tubules towards the lumen . However , if we focus on a single cycle of spermatogenesis within a segment of a mouse seminiferous tubule and compare it with a fly testis tubule , we see similar spermatogenic cell types present inside the fly testis tubule and the mouse seminiferous tubule segment [69] . Spermatogenesis in both species starts with spermatogonial stem cells located in a specific position of the tubule , attached to the apical end in fly and to basement membrane in mice , which move and progress into later stages of cell types in one direction , towards the basal end of the testis tubule in fly and towards the lumen in mice . All the major stages of sperm development appear to be present in both species and arranged in a similar spatial and temporal pattern . If both developmental processes evolved from an ancient primitive spermatogenesis prototype , one would predict the presence of at least some common male gametogenesis-specific regulators in both lineages . Yet no such common male gametogenesis factor has been demonstrated to be required exclusively for sperm production in both lineages . The lack of a universal male reproductive factor among all animal lineages , while consistent with rapid evolution of male reproductive genes , is in contrast to the prevalence of sexual reproduction and in particular to the similarity in male gametogenesis among metazoan animals [34] , [69] . This paradox led to the question of whether such similarity in the reproductive traits arose from convergent evolution or from conservation of an ancient prototype in the common ancestor . Furthermore , male reproductive traits and genes undergo rapid adaptive evolution in diverse lineages such as Drosophila , fish , rodents and primates [5] , [9]–[12] , [16] , [17] . Male-biased genes exhibit a higher divergence of expression among closely related species than female-biased genes or genes expressed in both sexes [7] , [13] . Additionally , testis-biased genes have the highest rate of extinction and species-specific de novo gene formation during evolution [13] , [70] , [71] . For example , the most widespread testis-specific proteins among both vertebrates and invertebrates appear to be sperm nuclear basic proteins ( SNBP ) . Many organisms replace histones with a set of small basic structural proteins ( SNBP ) or protamines to establish a highly compact sperm chromatin structure [72] , [73] . Although all metazoan SNBP homologs share their common ancestry with somatic histone H1 protein , the testis-specific SNBPs in different lineages have undergone extensive lineage-specific loss and dynamic evolution , including adaptive evolution [5] , [73] . Furthermore it remains unclear if vertebrate and invertebrate protamine homologs are functionally conserved . Loss of one copy of either mouse Protamine-1 or Protamine-2 leads to male sterility , but in contrast , fly sperm carrying a deletion of both protamine-like homologs appears to be functional [74] , [75] . Sexual selection has been proposed to be the major force driving this fast divergence of male reproductive traits , gene sequences , and their expression patterns [9] , [10] , [12] . Given that sexual reproduction is widespread among animals and sperm production appears to be present in all major phyla of metazoan animals , it raised a question whether any male-biased reproductive gene could be exempt from such selective pressure and remain conserved through extended evolutionary distances . However , Boule homologs have been maintained throughout all major lineages of animals from a common eumetazoan ancestral gene and are required only for sperm development in both Drosophila and mice . We have shown that Boule proteins have resisted sexual selective pressure , and instead evolved under purifying selection . Though ancestral Boule may have also functioned in oogenesis , our findings that bilaterian Boule homologs tend toward male-biased expression , taken together with the similar spermatogenesis arrest phenotypes in both Drosophila and mouse mutants , supports the model of a common origin of bilaterian spermatogenesis . While it remains to be seen if Boule homologs are restricted only to spermatogenesis or also function in the ovary , we have shown a clear case of conservation of a reproduction-specific gene across Bilateria . We found that among a broad representation of bilaterian animals , Boule expression was restricted to the gonads ( Figure 3 ) , indicating that it has remained reproduction-specific throughout evolution . In addition , DNA sequence analysis of multiple Drosophila and mammalian Boule homologs revealed that , unlike other reproductive proteins [11] , [16] , [17] , Boule evolution has been driven not by positive selection , but by purifying selection . This establishes an unambiguous case of a reproduction-specific gene being driven predominantly by purifying selection , in two distinct animal lineages , suggesting a strong functional constraint . Interestingly , our in-depth analysis of the developmental defects in Boule null mice revealed a novel requirement in spermatid differentiation [61] . Such a postmeiotic function for boule is also likely present in Drosophila , though its requirement for spermatid differentiation would not have been revealed in the boule mutant flies due to an earlier block at meiosis [39] , [61] . The previously established function of Boule in meiotic progression in both Drosophila and nematodes [39] , [40] may also be conserved in mice , despite the lack of a similar meiotic defect in Boule null mice [61] . We proposed that Dazl and Boule may redundantly regulate meiosis , and that Dazl may compensate for Boule loss during meiosis in mice [61] . Yet despite this possibility of a partial redundancy of function with Dazl , mouse Boule has been maintained under purifying selection , further indicating that the presence of other DAZ family genes has had little impact on the functional constraint of Boule . While meiosis is fundamental to sexual reproduction and key components of meiotic machinery for chromosomal synapses and recombination are conserved from yeast to mammals [2] , [76] , the absence of Boule homologs in fungi together with the requirement of Boule homologs in only one sex of animals suggest that conservation of Boule is unlikely due to the same functional constraint that keeps components of meiotic machinery conserved . Another main functional constraint on metazoan reproduction appears to be associated with germ cell specification and maintenance . Mutations disrupting those conserved germ cell components , such as Vasa or Piwi , often result in a failure to form germ cells or a loss of germ cells before meiotic stages . Furthermore , the resulting infertility sometimes affects both males and females of the same species [21]–[23] , [77]–[79] . These phenotypes differ from the sex-biased infertility of Boule mutations in all species examined , and the gametogenesis defects in Boule mutants are much less variable than those from either Vasa or Piwi mutants across species [21]–[23] , [61] , [78] , [79] . Further characterization of the subcellular expression and molecular function of Boule will help to discern the relationship between Boule and these other highly conserved germ cell proteins . We've shown the widespread presence of Boule homologs throughout bilaterian animals and the functional conservation of a reproductive-exclusive requirement among Drosophila , worm and mouse . This has revealed an ancient reproductive requirement in the Urbilaterian , the common ancestor of all bilaterian animals and highlights a fundamental reproductive function associated with Boule protein conserved over six hundred million years of evolution . With the identification of Boule and possibly more reproductive genes conserved across such large evolutionary distances , we can begin to compare the impact of sexual selection on the molecular evolution of the same components of reproductive traits in different animal lineages at both the microevolution and macroevolution levels .
For known Boule homologs , DNA sequences from various species were retrieved from the literature and the Genbank database . For species where the presence of Boule or Dazl homologs was unknown , we first searched the EST and cDNA database in Genbank using consensus RRM sequences of either Boule or Dazl using Tblastn and positively identified the homologs using our established criteria . The homolog sequences were further confirmed by the presence of Boule/Dazl homologs with high sequence similarity in other species within the same taxon . In the absence of EST or cDNA information , we then focused on a representative species from the same phylum whose genome had been completely sequenced . The specific genome databases were searched using the consensus Boule RRM sequences and Tblastn , and the top hits were analyzed to determine if they were Boule homologs based on criteria described above . The identified homologs were verified by BLASTing against the human protein database , which should identify human BOULE as the top hit sequence with highest similarity . New Boule and Dazl homologs we have identified as well as known homologs from previous publications are summarized in Table S3 . Sequence alignment of RRMs and entire proteins was performed using ClustalW2 and ClustalX programs [80] . The parameters for alignment were protein Gap open penalty = 10 , protein extension penalty = 0 . 2 , and other parameters at default settings . Phylogenetic analysis was done using Mega 4 . 0 [81] . Ka and Ks were calculated as described [44] . RNA was extracted by Trizol from tissues and reverse transcribed for amplification of Boule cDNA . For tissues collected in RNA later ( Applied Biosystems/Ambion , Austin TX ) , samples were stored at 4°C overnight , solution was removed , and the tissues were stored at −80°C for later RNA extraction . A minimum of two pairs of primers spanning the RRM region and other regions were used to confirm the expression of the Boule gene ( Table S4 ) . All the Boule amplicons were confirmed by sequencing . Dmrt1 ( Doublesex and mab-3 related transcription factor 1 ) was a testis-specific positive control in chicken [82] and Bnd ( Bindin ) was a testis-specific positive control in the sea urchin [83] . We used multiple sets of primers covering exons 2 , 3 , 4 , 5 and 6 , and determined that the main transcript is testis-specific ( Table S4 ) . Mice ( Mus musculus ) were housed and bred in a barrier facility according to the guidelines approved by the ACUC committee at Northwestern University . Boule mutant mice were created in the mixed background of C57B6 and 129svj . Ripe purple sea urchins ( Strongylocentrotus purpuratus ) in spawning season ( May , 2009 ) were collected from Pacific Ocean off Carslad , California ( M-REP , Carslad , California ) and shipped overnight on ice to Chicago . The sex of sea urchins was determined by the presence of eggs ( often a milky spill on the outside of female urchins upon arrival ) and by the presence of distinct gametes in the gonad tissue biopsies . Gonads , guts and ampullae from at least three male and three female purple sea urchins were collected and either stored in Trizol for immediate RNA extraction or snap-frozen in liquid nitrogen and stored at −80°C . Chicken gonadal and other tissues were collected from euthanized White leghorn chickens at the completion of a research project approved by UIUC animal committee at the University of Illinois at Urbana-Champaign Veterinary School . Tissues from two four-year old roosters and two three-year old hens were snap-frozen in liquid nitrogen or stored directly in RNA later . Tissue histology was performed as described previously [35] . Testes were fixed in Bouins' solution overnight and sectioned at 5-µm thickness for hematoxylin/eosin staining . Bright field images were captured using a Leica DM 5000B compound microscope with a DFC320 camera and the Leica image capture suite software . We replaced exon 3 with lacZ-neo using the NZTK2 vector ( Richard Palmiter , University of Washington , Seattle , WA ) . We designed primers with built-in SalI sites to amplify a 2-kb left arm next to exon 3 , and primers with built-in XhoI and NotI sites to amplify a 5 . 9-kb right arm from 129svj mouse genomic DNA . High fidelity platinum PCR kits ( Invitrogen ) were used to amplify the fragment with minimal PCR error . The amplified fragments were cloned into Topo vectors and later released with appropriate enzymes for subcloning into the NZTK2 vector . The clones with correct orientation of left and right arm insertions were chosen for sequencing . Sequencing of the genomic arms in the selected clones indicated that both arms had greater than 99% sequence identity with the genomic sequence . Gene targeting was performed on 200 ES cell clones ( 129svj E14 feeder cell-less ES cells ) and four positive ES clones ( 1D5 , 1H5 , 2A4 and 2D7 ) were identified by the presence of both the 2-kb and 6-kb arms using primers outside each arm and on the vector . The 1D5 clone was used to inject blastocysts at the Northwestern University Transgenic Core Facility . Four chimerical mice produced lacZ-positive progeny and mice from two independent founders were used to generate mutant mice for the analysis . The phenotypes were identical among the mutant mice from two independent lines and we did not distinguish our analyses between the two lines . | While sexual reproduction is widespread among animals , it remains enigmatic to what extent sexual reproduction is conserved and when sex-specific gametogenesis ( spermatogenesis and oogenesis ) originated in animals . Here we demonstrate the presence of the reproductive-specific protein Boule throughout bilaterally-symmetric animals ( Bilateria ) and the conservation of its male reproductive function in mice . Examination of Boule evolution in insect and mammalian lineages , representing the Protostome and Deuterostome clades of bilateral animals , failed to detect any evidence for accelerated evolution . Instead , purifying selection is the major force behind Boule evolution . Further investigation of Boule homologs among Deuterostome species revealed reproduction-specific expression , with a strong prevalence of testis-biased expression . We further determined the function of a deuterostomian Boule homolog by inactivating Boule in mice ( a representative mammal , a class of Deuterostomes ) . Like its counterpart in Drosophila ( a representative of the opposing Protostome clade ) , mouse Boule is also required only for male reproduction . Loss of mouse Boule prevents sperm production , resulting in a global arrest of spermatogenesis in remarkable similarity to that of Drosophila boule mutants . Our findings are consistent with a common origin for male gametogenesis among metazoans and reveal the high conservation of a reproduction-specific protein among bilaterian animals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"evolutionary",
"biology/animal",
"genetics",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics",
"evolutionary",
"biology/sexual",
"behavior",
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/population",
"genetics",
"genetics",
"and",
"genomics",
"evolutionary",
"biology/developmental",
"evolution"
] | 2010 | Widespread Presence of Human BOULE Homologs among Animals and Conservation of Their Ancient Reproductive Function |
Identifying genomic elements required for viability is central to our understanding of the basic physiology of bacterial pathogens . Recently , the combination of high-density mutagenesis and deep sequencing has allowed for the identification of required and conditionally required genes in many bacteria . Genes , however , make up only a part of the complex genomes of important bacterial pathogens . Here , we use an unbiased analysis to comprehensively identify genomic regions , including genes , domains , and intergenic elements , required for the optimal growth of Mycobacterium tuberculosis , a major global health pathogen . We found that several proteins jointly contain both domains required for optimal growth and domains that are dispensable . In addition , many non-coding regions , including regulatory elements and non-coding RNAs , are critical for mycobacterial growth . Our analysis shows that the genetic requirements for growth are more complex than can be appreciated using gene-centric analysis .
Mutagenesis has long been a powerful tool for understanding the roles of genes and other chromosomal elements . Recently , high-density transposon insertion mutagenesis coupled with deep sequencing has enabled comprehensive identification of the required genes in many important bacterial pathogens [1]–[6] . Defining the protein-coding genes required for bacterial growth identifies both key biological processes and potential targets for drug development . However , protein-coding genes are not the only genetic elements that code for required functions . In densely packed bacterial genomes , many regulatory regions are required for appropriate expression of genes [7] . Moreover , all organisms produce large numbers of non-coding RNAs that can be important under a variety of growth conditions [8]–[10] . Gene-oriented analyses also look past cases wherein a single gene encodes several differentially important protein domains . Here , rather than focusing on genes , we take an unbiased approach to create a comprehensive understanding of genomic requirement in Mycobacterium tuberculosis ( Mtb ) . We model the Mtb genome as made up of “functional units” , a term that encompasses both genes and other genetic elements , many of which have yet to be annotated . By not limiting our analysis to whole-gene regions , we can find otherwise unidentified functional units while also gaining a more nuanced view of the genes required for mycobacterial growth , including critical domains within proteins and non-protein-coding regions that play important roles . We find approximately 300 protein-coding genes wherein only portions of the coding sequence are required . These include genes , such as ppm1 and fhaA , where we demonstrate that one domain is required for optimal growth whereas other domains are not . Our unbiased analysis also revealed required genomic elements in regions sitting between protein-coding genes . These include two RNAs , the tmRNA and the RNA component of RNaseP , which are required for optimal growth . In addition , we find a number of other regions that influence viability by uncharacterized mechanisms , but whose effects have previously been overlooked by gene-centric analyses .
To perform a comprehensive assessment of Mtb's genetic requirements for growth , we used two ∼100 , 000-clone Mtb libraries generated through high-density transposon mutagenesis of the H37Rv strain [11] . We generated a library of single-insertion mutants by phage delivery of the Himar1 transposon , which randomly inserts into the genome at sites recognized by the TA dinucleotide ( Figure 1A ) . We then created transposon-mapping probes by selectively amplifying and sequencing transposon-genome junctions using an Illumina Genome Analyzer 2 . Using genome sequences adjacent to the transposon genomic sequences , we were able to map the insertion site of mutants in the library ( Figure 1B ) and count the reads mapped to each insertion site ( insertion count , Table S1 ) . We reasoned that the insertion count should reflect the number of corresponding mutants in the library . To demonstrate this , we picked twelve individual transposon mutants and added each at a known quantity to a manually constructed library . Insertions were again mapped and counted by deep sequencing , and the insertion counts for each site was compared to the known relative quantities of each mutant in the pool ( Figure 1C ) . Insertion counts were highly correlated with the known relative amount of each mutant ( Pearson R = 0 . 880 , p-value<0 . 0001 , n = 14 ) . Additionally , we confirmed that insertion counts accurately reflected the library's genome composition by counting the genome-transposon templates represented in our Illumina reads . Since random shearing events create the genome-transposon templates for amplification , the distance between the transposon and the sheared end represents a unique identifier for each template . We assessed the relationship between estimates of unique template molecules for each TA site and the read count for that site ( Figure 1D ) , revealing excellent correlation ( Pearson R = 0 . 945 p-value<0 . 0001 n = 36 , 488 ) . In our Mtb library , transposon insertions occurred at 36 , 488 of the 72 , 927 possible insertion sites ( TA dinucleotides ) . Each library generated an average of 2 . 3 million reads , resulting in a mean insertion count of 64 per hit-site . We counted the number of sequencing reads from each site in the two libraries and compiled the counts correcting for each library's total insertion count . Having demonstrated that insertion counts faithfully represent mutant numbers ( Figure 1C ) , we used insertion counts to comprehensively assess the relative importance of selected genomic regions . We defined a region as required for optimal growth if total regional insertions were statistically underrepresented compared to genomic controls ( Figure 2A ) . Required regions , therefore , are those in which mutations result in a statistically validated growth defect . We employed a non-parametric test to assess statistical underrepresentation and regions with a p-value of less than 0 . 01 and a false discovery rate ( Benjamini-Hochberg ) of less than 0 . 1 were defined as required for optimal growth in vitro . Instead of assessing the requirement for growth of genomic regions based on predetermined gene coordinates , we divided the genome into contiguous overlapping windows to assess a comprehensive set of potential functional units . Our non-parametric test was powered to find significant regions containing at least 7 TA sites ( 6 or fewer precluded confident rejection of a null hypothesis of variation by chance alone ) . Thus , we focused on regions of sizes likely to contain 7 or more TA sites . The mean number of TA sites in windows of 400 , 500 , and 600 bp was 6 . 75 , 8 . 45 and 10 . 12 , respectively , and were thus used for our sliding window analysis of functional requirement for growth . Intergenic ( IG ) regions are relatively AT-rich in the Mtb genome , allowing us to add a 250 bp sliding window ( mean of 6 . 20 TA sites in IG regions ) to the analysis of IG regions . To lower the computational demands of this analysis , we chose to analyze every tenth window , reasoning also that functional units were unlikely to be smaller than 10 base pairs . Thus , we assessed the requirement for growth of every tenth 400 , 500 , and 600 bp window in the genome , along with every tenth 250 bp window in regions between protein-coding genes ( Figure 2B , Table S4 ) . We overlaid the coordinates of known genes on the generated results to find those that contained regions required for optimal growth . Of the 3 , 989 annotated genes , 742 contained required functional units , 3 , 089 contained no required functional units , while 158 did not sustain insertions but also did not contain enough TAs to meet statistical requirements ( Figure 3A , Table S2 ) . As a screen for genes with multiple functional units of varying requirement , we searched for genes that contained both required and non-required regions . A total of 317 genes met these criteria ( Figure 3A , Table S2 ) . Our finding that many genes contain both required and non-required regions suggested that using only whole genes for analysis could misrepresent their importance for growth . Either the entire gene could appear required for optimal growth or the entire gene would be considered dispensable , leaving no room for the possibility that only a segment of the gene might be required . To determine how our results would compare to a gene-centric analysis , we calculated the requirement for growth of each gene by applying the non-parametric test to the gene as a whole . As expected , genes with required segments had a wide range of p-values when assessed using annotated boundaries instead of unbiased overlapping windows ( Figure 3B , dark blue bars ) . A total of 170 out of the 317 ( 53 . 6% ) had p-values above 0 . 01 , demonstrating that our sliding window strategy accounted for a significant number of required functional units that would be ignored by gene-only strategies ( Figure 3B ) . A transposon insertion into the 5′ end of a gene will often block production of the encoded protein , either by attenuating transcription or disrupting ribosome binding sites and initiation codons . Thus , we expected that regions required for optimal growth would tend to be found at the 5′ ends of predicted genes . Surprisingly , a plot of the likelihood of discovering an required functional region as a function of the intragenic location revealed a symmetric curve , demonstrating that the required regions discovered have an equal likelihood of residing on either end of the gene ( Figure 3C ) . We hypothesized that this may be because the transposon contains a promoter that can direct downstream transcription . To test this , we took two strains that contained transposon insertions and measured mRNA expression upstream and downstream of the transposon ( Figure S2A ) . In both cases , expression upstream of the transposon did not significantly change , while downstream expression increased ( Figure S2B ) . This is consistent with the observation that downstream genes can be transcriptionally activated by transposon insertions [12] . In addition , mycobacteria are able to use several initiation codons thus making it more likely that truncated but functional proteins can be produced from internal start sites . While we were able to use this analysis to make many novel observations , we also found that our results supported previous findings . The majority of genes ( 63% ) described as fully required for growth were similarly required in microarray-based studies using transposon site hybridization ( TraSH ) ( Figure S1A , Table S2 ) [13] . The increased resolution from deep sequencing demonstrated that genes with fewer than 7 TAs resulting in an undersampling that prevented statistically confident requirement assessments ( a separate category for genes with 6 or fewer TAs that did not contain insertions is noted in Figure 3A and Table S2 ) . Since this was not known previously , we predicted that the microarray-determined set of required genes would be biased towards small genes . This proved to be true . In genes predicted to be required by TraSH but not in this study , the average number of TAs was 9 . 90 ( Figure S1B ) . In contrast , the average number of TAs in fully required genes from this study was 19 . 84 , a fair representation of the average of all genes assessed ( 19 . 47 ) . In fact , of the genes only determined to be required in TraSH and not in this study , 43% had 7 or fewer TAs , accounting for much of the discordance between the two methods . A more nuanced analysis of Mtb transposon insertion maps defined essential genes as those that contained “gaps , ” any statistically significant runs of potential insertion sites lacking transposon insertions [3] . As expected , genes found in our sliding window analysis to have both required and non-required regions were more concordant with essential genes found by sequencing using this gap analysis than with microarray approaches or whole-gene analyses of insertion counts . Of genes described in our approach as fully required , 97 . 1% were described as “essential” by Griffin et al ( Figure S1A ) , a remarkable level of agreement given the differences in growth media between the two studies . The increased concordance extended to genes containing both required and non-required regions . Griffin et al . described 151 of these genes as essential , while microarray methods only deemed 81 to be essential . The search for required regions within genes , a feature of both analyses , allowed for the discovery of these regions in longer genes , as evidenced by the increase in average number of TAs within these genes ( Figure S1B ) . We find that , in some genes , encoded domains have different effects on growth , accounting for the varying degrees of requirement across the gene's open reading frame . For example , the gene encoding Ppm1 , an enzyme in the lipoarabinomannan ( LAM ) synthesis pathway , encodes a protein with two distinct domains . The region encoding the carbon-nitrogen hydrolase domain of Ppm1 sustained many insertions , while the region encoding the C-terminal glycosyl transferase was required for optimal growth ( Figure 4A ) . While the specific requirement of the glycosyl transferase is a novel finding , it resonates with a previous report that only the glycosyl transferase was required for the synthesis of LAM , thought to be an essential cell wall component [14] . Another study revealed that Ppm1 has N-acyltransferase activity , which could be the non-required function of this two-domain protein [15] . To confirm that the lack of insertions in this domain was due to a functional requirement and not to insertional bias or the generation of toxic fusions or truncations , we created transposon libraries in the presence of a second copy of ppm1 . We reasoned that a second copy would render the endogenous gene non-required and thus permissive for transposon insertion . We designed footprinting PCR primers upstream of the original ppm1 to specifically generate amplicons containing transposon insertions into the endogenous copy ( Figure 4B ) . Footprinting of the original library confirmed our sequencing results , as no insertions were found in the region encoding the glycosyl transferase . However , in the complemented library , that region did contain insertions , suggesting the glycosyl transferase is functionally required for growth . We further reasoned that only sense insertions—that is , insertions wherein the transposon's internal promoter is oriented in the same direction as the disrupted gene—would be tolerated in the 5′ end of ppm1 to allow for the expression of the C-terminal required domain . To assess this , we used primers specifically designed to amplify sense and anti-sense insertions , and noted only sense insertions in the 5′ end ( Figure S2C ) . In addition , we confirmed that many in-frame internal start sites exist between 5′ transposon insertion sites and the beginning of the 3′ domain . A recent report showed that FhaA was required for optimal growth of Mycobacterium smegmatis and postulated that the importance of the interaction of FhaA with the essential protein MviN for appropriate regulation of growth and peptidoglycan synthesis [16] . These processes are essential for mycobacterial cell division and cell wall biosynthesis . This work further demonstrated the C-terminal forkhead associated ( FHA ) domain of FhaA was required for MviN-binding , while an N-terminal domain of unknown function was dispensable for this interaction . In agreement with these findings , we show here that the region of fhaA encoding the FHA domain cannot sustain insertions , while the remainder of the gene is dispensable ( Figure 4C ) . We used insertion footprinting to confirm these results , and found that the C-terminal insertion mutants were rescued for growth in the presence of a second copy of fhaA ( Figure 4D ) . Notably , both ppm1 and fhaA , which we predict to be required for optimal growth based on the presence of a required region within these genes , were classified as non-essential in a previous microarray-based screen [13] . In fact , 247 of the 328 genes containing both required and non-required regions were not previously described as necessary for growth , likely due to the decreased spatial resolution of microarray-based methods . Microarrays limited the resolution of requirement testing to genes , and each gene received a single metric describing its requirement for growth . In addition , as our approach is not confined to gene boundaries , we have the additional resolution to identify domains within genes , as exemplified by ppm1 and fhaA . Because we are not limited to annotated regions we were also able to probe the importance of intergenic regions . By scanning the genome for required 250 , 400 and 500 bp regions , we found 25 intergenic regions required for optimal growth ( Figure 2B and Table S3 ) . These required intergenic regions contained many components of known essential cellular functions to be required for in vitro growth . These included 10 tRNAs as well as the RNA catalytic unit of RNaseP , which has been shown to be required for tRNA processing in other bacteria ( Figure 5A ) . Additionally , one required intergenic region contained the tmRNA , a molecule required to release stalled ribosomes and to tag polypeptides for proteolytic degradation through an essential protease ( Figure 5B ) [17]–[18] . Of the intergenic segments containing functionally required regions , 11 had annotated functions and an additional 6 were adjacent to genes assessed as required for growth and , therefore , might contain promoters or other transcriptional regulatory elements . The remaining 19 required segments are situated between two non-required genes and , as yet , have no ascribed function .
Finding genetic loci that are required for optimal growth under specific conditions helps inform the basic understanding of bacterial physiology and efforts to develop new therapeutics for pathogens . Previously , we and others have used transposon mutagenesis to infer the requirement for genes under different growth conditions by utilizing the information provided by genome annotations [1]–[6] . Deep sequencing , which allows us to map precisely the insertion site of every mutant , affords a higher resolution assessment of genetic requirement , beyond just genes . Here , we demonstrate that an unbiased sliding window approach harnesses the full potential of this increased resolution . This approach identified not only whole genes required for optimal growth but also other required elements , such as non-protein coding RNAs and protein domains within insertion-containing genes , which would otherwise obscured by gene-centric analysis . An alternative analysis that uses significant gaps in insertion—rather than quantitative insertion counts—was also able to assess the requirement of protein domains ( DeJesus et al . , unpublished data , submitted ) . This analysis likely identifies regions absolutely essential for viability rather than all regions required for optimal growth . We found that many genes contain elements that are important for growth even though other regions are not required . In at least two cases , ppm1 and fhaA , published data have shown that the required regions encode specific protein domains . However , in other cases , these might represent non-protein-coding RNAs or cis regulatory elements . Bacteria encode many small RNAs many of which could be required for optimal growth and some of which are embedded within genes [8]–[10] . In addition , most genes have been annotated computationally , an uncertain pursuit that clearly can lead to misannotated start sites [19] . Genes with only 5′ insertions could fall into this category . Similarly , important non-protein-coding regions could have multiple roles . In some cases , we found that known RNAs , such as rnpB , the catalytic RNA component of RNase P , and the tmRNA were required for optimal growth , supporting previous speculation [20]–[21] . Again , some other required regions might encode as yet unidentified non-coding RNA molecules . Still others might be promoters or other regulatory regions . In this study , our resolution was limited by the specific properties of the Himar1 transposon in mycobacteria . Our previous studies have shown that insertions are randomly distributed apart from the desired selection against insertion in essential regions [11] , [22] . Despite this , we cannot assume that all sites lacking insertions represent required regions since unknown insertional biases of the transposon may exist . Thus , we defined a required region as one with a statistically underrepresented insertion count using a non-parametric test to account for such potentially unique biases within these data ( Figure 2A ) . This allowed us to exclude , for example , windows with 6 or fewer TA sites , which demonstrably lacked power to distinguish a region as essential for growth relative to background variation . In GC-rich protein-coding regions , this limited our scope to windows of greater than 400 bp; less GC-rich intergenic regions allowed the assessment of windows greater than 250 bp . Thus , while we were able to identify many required protein domains and RNAs , it is certainly possible that smaller elements required for growth were missed due to these size constraints . This is a particular problem for non-coding RNAs that are often very small . For example , while we found 10 tRNAs required for growth , the remaining tRNAs reside in non-coding regions that did not have the requisite number of TA sites to determine requirement . Using the Himar1 transposon in organisms with less of a GC bias , or in organisms in which a less restricted transposon exists , should result in increased resolution [4] . The analysis we used provides a powerful tool to perform functional genome analysis . Importantly , this type of approach is useful not only for single conditions , as we described but can also be used to identify elements critical under one growth condition but not another [23]–[25] . This is particularly important in organisms like Mtb , an obligate pathogen that never grows under conditions precisely comparable to those we use in vitro . Coupling high-density insertion libraries with deep sequencing and analytic methods such as that described here provides a powerful experimental tool for functional genome annotation .
Two independent libraries of 100 , 000 mutants were generated in the Mtb strain H37Rv as previously described on 7H10 agar [11] . Independent libraries were also generated in Mtb strains overexpressing ppm1 and fhaA . Genomic DNA was isolated from each library and randomly fragmented to 400–600 bp pieces by sonication with a Covaris E220 . Nicked ends were repaired ( Epicentre end repaired kit ) , and A-tails were added with Taq polymerase to allow the ligation of T-tailed adapters . Transposon-junctions were amplified for 30 cycles ( 94 degrees , 30 seconds; 58 degrees , 30 seconds; 72 degrees , 30 seconds ) using a primer recognizing the transposon end ( 5′-AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCCGGGGACTTATCAGCCAACC-3′ ) and one recognizing the adapter ( 5′-CAAGCAGAAGACGGCATACGAGATCGGTCTCGGCATTCCTGCTGAACCGCTCTTCCGATCGTCCAGTCTCGCAGATGATAAGG-3′ ) . Primers used during amplification contained all the requisite sequence for binding to the Illumina sequencing platform . A 250–400 bp fragment of the amplicon was isolated from a gel and sequenced on an Illumina GA2 instrument with a custom sequencing primer ( 5′-TTCCGATCCGGGGACTTATCAGCCAACC-3′ ) . Reads from the Illumina sequencing run were first screened for the presence of sequence from the end of the Himar1 transposon . The following 35 bases were mapped to the Mtb genome , allowing for 2 mismatches . Reads that mapped to the genome at a TA site were designated as mapped insertions . Reads that mapped to multiple sites were randomly assigned to one of the mapped sites . For each library , the number of reads mapping to each site ( insertion counts ) was counted . Insertion counts were plotted on IGV and CGViewer [26]–[27] . For every possible region size containing x potential insertion sites , a null distribution of mean read counts was generated by calculating the mean read counts from a set of 10 , 000 randomly selected sets of x sites . The 10 , 000 randomly generated means were sorted and the rank of the test region's mean insertion count within the ordered null distribution was determined . The p-value was calculated as the rank of the test mean divided by the size the null distribution ( 10 , 000 ) . Multiple test correction was performed by calculating the Benjamini-Hochberg false discovery rate over all regions tested . Regions containing 7 TA sites with no insertions had a p-value of 0 . 008 and an FDR of 0 . 06 , while regions containing 6 TA sites with no insertions had a p-value of 0 . 018 and an FDR of 0 . 12 . In order to power our study to detect required regions containing at least 7 TAs , we determined a region to be required for optimal growth if it had a p-value less than 0 . 01 and an FDR less than 0 . 1 . Footprinting of transposon insertion sites was performed by PCR using a primer recognizing the Himar1 ITR sequence ( 5′-CCCGAAAAGTGCCACCTAAATTGTAAGCG-3′ ) and primers recognizing a genomic segment just upstream of the gene of interest . For directional footprinting , we used one primer to amplify sense insertions ( 5′-TTTTCTGGATTCATCGACTGTGGC-3′ ) —where the kanamycin resistance gene on the transposon was oriented in the same direction as the disrupted gene—and another for antisense insertions ( 5′-CAGCTCATTTTTTAACCAATAGGCCG-3′ ) . Standard PCR conditions were used for long amplification with Phusion polymerase ( 94 degrees , 15 seconds; primer-dependent annealing temperature , 30 seconds; 72 degrees , 2 minutes ) . | The significant rise in drug resistant strains of Mycobacterium tuberculosis has highlighted the need for new drug targets . Here , we present a novel method of defining genetic elements required for optimal growth , a key first step for identifying potential drug targets . Similar strategies in other bacterial pathogens have traditionally defined a set of essential protein-coding genes . Bacterial genomes , however , contain many other genetic elements , such as small RNAs and non-coding regulatory sequences . Protein-coding genes themselves also often encode more than one functional element , as in the case of multi-domain genes . Therefore , instead of assessing the quantitative requirement of whole genes , we parsed the genome into comprehensive sets of overlapping windows , unbiased by annotation , and scanned the entire genome for regions required for optimal growth . These required regions include whole genes , as expected; but we also discovered genes that contained both required and non-required domains , as well as non protein-coding RNAs required for optimal growth . By expanding our search for required genetic elements , we show that Mycobacterium tuberculosis has a complex genome and discover potential drug targets beyond the more limited set of essential genes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods",
"and",
"Materials"
] | [
"genome",
"analysis",
"tools",
"functional",
"genomics",
"genetic",
"screens",
"microbial",
"pathogens",
"biology",
"genomics",
"microbiology",
"microbial",
"growth",
"and",
"development",
"genetics",
"and",
"genomics"
] | 2012 | Global Assessment of Genomic Regions Required for Growth in Mycobacterium tuberculosis |
The appearance of the notochord represented a milestone in Deuterostome evolution . The notochord is necessary for the development of the chordate body plan and for the formation of the vertebral column and numerous organs . It is known that the transcription factor Brachyury is required for notochord formation in all chordates , and that it controls transcription of a large number of target genes . However , studies of the structure of the cis-regulatory modules ( CRMs ) through which this control is exerted are complicated in vertebrates by the genomic complexity and the pan-mesodermal expression territory of Brachyury . We used the ascidian Ciona , in which the single-copy Brachyury is notochord-specific and CRMs are easily identifiable , to carry out a systematic characterization of Brachyury-downstream notochord CRMs . We found that Ciona Brachyury ( Ci-Bra ) controls most of its targets directly , through non-palindromic binding sites that function either synergistically or individually to activate early- and middle-onset genes , respectively , while late-onset target CRMs are controlled indirectly , via transcriptional intermediaries . These results illustrate how a transcriptional regulator can efficiently shape a shallow gene regulatory network into a multi-tiered transcriptional output , and provide insights into the mechanisms that establish temporal read-outs of gene expression in a fast-developing chordate embryo .
The transcription factor Brachyury plays a paramount role in mesoderm formation in animals with widely diverse body plans [1] , [2] . In chordate embryos ranging from sea squirts to mice , Brachyury is required for the formation of the notochord from axial mesoderm [3] , [4] . In addition to its prominent function in notochord development , Brachyury has recently been shown to induce epithelial-mesenchymal transition when over-expressed in human carcinoma cells [5] and it has been described as a causative agent of chordomas , human tumors of presumed notochordal origin [6] , [7] . Brachyury is believed to exert its multifaceted role by controlling the transcription of a large number of downstream effectors [8]–[10] . This point is proven by studies in various systems , including ascidians , zebrafish , mouse , and chordoma cell lines , all showing that Brachyury binds hundreds of genomic loci [11]–[15] . Although genome-wide studies have added numerous candidates to the list of mesodermal genes whose activity is influenced by Brachyury , the specific cis-regulatory mechanisms through which this factor performs its crucial function in notochord formation are still in need of elucidation . This is mainly due to the fact that detailed studies of cis-regulatory modules ( CRMs ) are complicated in vertebrate model systems by a number of intrinsic experimental limitations , including the early pan-mesodermal expression of Brachyury , genomic complexity , scarce accessibility of the notochord , slow embryonic development , and laborious transgenic protocols . However , in the ascidian Ciona intestinalis , an invertebrate chordate , expression of the single-copy Brachyury ( Ci-Bra ) gene is restricted to notochord cells by the action of the transcriptional repressor Ciona Snail [16] . In addition , compared to other chordates , the Ciona model system is characterized by a compact , fully sequenced genome , a readily distinguishable notochord , fast development , and ease of transgenesis [17]–[20] . The specificity of Ci-Bra expression in notochord cells has provided a unique experimental advantage for the initial identification of over 50 validated Ciona genes controlled by this transcriptional activator [11] , [21]–[25] . More recently , the number of potential Ci-Bra target genes has surged to over 2 , 000 following genome-wide studies of chromatin occupancy by this factor in early embryos [13] . These observations have led to the assumption that Ci-Bra presides over a “shallow” gene network and controls the majority of its targets directly . In support of this view , the notochord CRMs associated with two early-onset Ci-Bra downstream genes , Ci-tropomyosin-like ( Ci-trop ) and Ci-leprecan , have been found to be controlled by Ci-Bra directly , through non-palindromic binding sites that share the consensus sequence TNNCAC [21] , [26] . Remarkably , however , even though transcripts for Ci-Bra appear in notochord cells from the 64-cell stage and persist throughout embryogenesis [27] , [28] , many of its bona fide target genes are sequentially activated at various developmental stages [22] , [25] , [29] . The sequential deployment of Ci-Bra targets is crucial to ensure that the morphogenetic steps that lead to notochord formation progress seamlessly . As a result , within ∼8 h , notochord cells transition smoothly through invagination , convergent extension , cell-shape changes , lumen matrix secretion , and tube formation [30] . We sought to uncover how Ci-Bra establishes this sequential transcriptional output , and to this aim we analyzed the architecture and functional requirements of the notochord CRMs associated with a representative suite of Ci-Bra downstream genes expressed at different stages of notochord development . Upon completion of these analyses , the newly discovered cis-regulatory mechanisms were used to identify notochord CRMs from still uncharacterized Ci-Bra downstream genes and to predict their temporal onset of expression . The in vivo occupancy of the CRMs directly controlled by Ci-Bra was assessed by chromatin immunoprecipitation ( ChIP ) assays . This investigation yielded an evolutionarily conserved consensus sequence shared by functional Brachyury binding sites , led to a classification of the direct Ci-Bra target CRMs into different groups , and uncovered a relay mechanism that ensures the activation of late-onset Ci-Bra targets . We propose that these cis-regulatory strategies concertedly create a differential temporal read-out of the steady transcriptional input provided by Brachyury .
The expression patterns of numerous notochord genes that are likely controlled by Ci-Bra have been described in previous studies [11] , [22] , [23] , [25] , [31] . In this study we used whole-mount in situ hybridization ( WMISH ) to precisely determine the onset of notochord gene expression for a subset of bona fide Ci-Bra target genes for which this information was missing or incomplete ( Figures 1A–1O and S1 ) . The results of this analysis and of previous reports are summarized in Figure 1P and plotted against a time-course of the main developmental events that punctuate notochord formation in Ciona . From these comparisons , it is evident that the genes controlled by Ci-Bra fall within different classes , which we define here as early , middle , and late onset . Early-onset genes are detected in notochord precursors from gastrulation and include Ci-prickle ( Ci-pk ) , the gene with the earliest onset [22] , Ci-thrombospondin3 ( Ci-thbs3 ) ( Figure 1A–1E ) , Ci-fibrillar collagen ( Ci-FCol1 ) ( Figure S1A–S1D ) , Ci-Noto5 ( Figure S1E–S1H ) , and Ci-ezrin-radixin-moesin ( Ci-ERM ) ( Figure S1I–S1L ) . Middle-onset genes begin to be expressed in the notochord by late gastrulation , when the neural plate becomes distinguishable and is composed of ∼six rows of cells abutting the notochord precursors [32]; these genes include Ci-Noto1 [22] , Ci-Noto8 ( Figure 1F–1J ) , Ci-Noto4 ( Figure S1M–S1P ) , and Ci-Noto9 ( Figure S1Q–S1T ) . The late-onset genes include Ci-ATP citrate lyase ( Ci-ACL ) , which is first detected at the late neural plate stage ( [22] and our unpublished results ) and Ci-β1 , 4-Galactosyltransferase ( Ci-β4GalT ) , which is first detected at the neurula stage ( Figure 1K–1O ) . We aimed at identifying the cis-regulatory mechanisms responsible for the differences observed in the developmental onset among Ci-Bra-downstream genes . To accomplish this goal , we employed a position-biased cloning strategy to identify notochord CRMs located within the genomic loci of bona fide Ci-Bra transcriptional targets representative of the early , middle and late-onset groups . Genomic fragments ranging from 1 to 3 . 7 kb were PCR-amplified from the 5′-flanking regions of 17 Ci-Bra target genes ( Figures S2 and S3 ) , accounting for a total of ∼43 kb of C . intestinalis genomic DNA . The size and genomic location of the initial fragments were selected on the basis of previously published work , which has shown that Ciona CRMs are compact sequences that frequently lie either in the 5′-flanking region or in the first intron of the gene with which they are associated [33] . Each fragment was cloned into the pFBΔSP6 vector upstream of the Ci-FoxA-a basal promoter region fused to LacZ [24] and tested in vivo for cis-regulatory activity by electroporation in Ciona zygotes [34] . Ten of the 17 genomic fragments were able to activate gene expression in notochord cells; of these , five , Ci-Noto1 , Ci-Noto8 , Ci-Noto9 , Ci-β4GalT , and Ci-ERM , map to larger genomic regions that have been reported to display notochord activity by a parallel study from another group [35] , [36] . In the majority of cases , the CRMs that we identified also directed expression in other tissues in addition to the notochord ( Figure S2 ) . Of the remaining seven genomic regions , five , two of which from the same locus ( Ci-Noto3 ) , were found to contain CRMs active in tissues other than the notochord; the patterns of activity exhibited by some of these fragments partially recapitulated the expression of their neighboring genes ( Figure S3 ) . In sum , of the 17 loci that were surveyed , eight harbored a notochord CRM near their 5′-end . In addition to these , one gene , Ci-Noto5 , was found to contain a notochord CRM within its coding region , ∼9 kb downstream of its transcription start site , while the minimal functional sequences for Ci-ACL were found within its third intron . The characterization of minimal notochord CRMs was achieved through the analysis of sequence-unbiased serial truncations of the initial genomic fragments . Here we define as “minimal CRMs” those enhancer sequences , usually ranging from 65 to ∼300 bp , that are still able to direct a consistent , clearly detectable notochord staining in vivo . Once identified , the minimal CRMs were subjected to sequence-biased ( i . e . , binding site-specific ) individual and combined point-mutation analyses . Putative binding sites were identified by scanning the sequences of each CRM with previously published consensus binding sites for Ciona notochord transcription factors , as well as using available databases . In particular , we used the consensus sequence TNNCAC to identify putative Ci-Bra binding sites , on the basis of previously published observations in Ciona [21] as well as in other chordates [37] , [38] . We first attempted the dissection of the Ci-pk upstream region ( Figure S2 ) . We found that the 1 . 2-kb Ci-pk notochord CRM that we had identified is enriched in putative Ci-Bra binding sites with the generic TNNCAC core sequence ( 15 versus <5 expected by random occurrence ) . However , this notochord CRM relies not only upon the numerous putative Ci-Bra binding sites that are found in its distal region , but also on additional sequences located in its proximal region ( unpublished data ) . This structural feature prevented the isolation and further dissection of minimal sequences required for the function of this CRM . In the case of Ci-thrombospondin 3 ( Ci-thbs3 ) [25] , through serial truncations , we were able to reduce the original 1 . 96-kb genomic region ( Figure S2 ) to a 116-bp minimal notochord CRM ( Figure 2A ) . The minimal 116-bp Ci-thbs3 notochord CRM contains four Ci-Bra binding sites ( numbered T1–4 in Figure 2A ) , which do not affect notochord activity when individually mutagenized ( Figure 2A–2D ) . However , when three of these sites ( T2 , T3 , and T4 in Figure 2A ) are mutated in conjunction , their synergistic activity becomes evident ( Figure 2E–2G ) . These results are quantified in the graph in Figure 2H . In particular , the double mutations of the T2 and T4 sites , which share the same core sequence TCGCAC , and of the T2 and T3 sites ( Figure 2E , 2H , and unpublished data ) leave only very little notochord activity ( red arrowheads in Figure 2E ) . However , only the triple mutation of sites T2 , T3 , and T4 completely inactivates the CRM in notochord cells ( Figure 2G and 2H ) . In general , we observed that mutations attacking the “CAC” sequence were usually more effective than the mutations that changed the “T” in the TNNCAC core ( unpublished data ) ; hence these mutations were used for all Ci-Bra binding sites analyzed . Another subset of notochord CRMs were also found to require multiple Ci-Bra binding sites for their activity ( Figure 3 ) . These notochord CRMs are associated with the genomic loci of Ci-FCol1 and Ci-Noto5 and contain Ci-Bra binding sites of various core sequences . The Ci-FCol1 notochord CRM was originally identified as part of a larger cis-regulatory region , spanning 2 . 2 kb ( Figure S2 ) , which also harbors muscle and endoderm CRMs [39] . Truncation analyses allowed the identification of a 65-bp minimal CRM , which contains three Ci-Bra binding sites that were individually mutagenized to assess their respective roles ( Figure 3A–3F ) . Mutation of the distal-most site , with a TTTCAC core , had little or no effect on cis-regulatory activity ( Figure 3C and 3G ) , but when similar mutations were introduced in the centrally located Ci-Bra binding site , with a TATCAC sequence , a reduction of notochord activity was detected ( Figure 3D and 3G ) . A stronger effect on notochord staining was seen when the proximal TAACAC site was mutated ( Figure 3E and 3G ) . Finally , the combined mutation of the central and proximal Ci-Bra binding sites was able to completely abolish notochord activity ( Figure 3F ) . The Ci-Noto5 notochord CRM , which appeared to be located in the 5′-flanking region of this gene according to outdated gene models , is instead contained in an intron of gene model KH . L153 . 32 ( Figure S2 ) [40] . The minimal 83-bp Ci-Noto5 notochord CRM contains two Ci-Bra binding sites at its extremities and a centrally located putative Fox binding site ( Figure 3H ) with a TRTTTAY core . This sequence was examined because it is shared with the functional Ci-FoxA-a site that is required to activate the Ci-tune notochord CRM synergistically with Ci-Bra [41] . Interestingly , none of the individual mutations of either Ci-Bra or Ci-Fox binding sites had any detectable effect on the Ci-Noto5 CRM notochord activity ( Figure 3I–3L ) ; however , the combined mutation of both Ci-Bra binding sites completely abolished staining in notochord cells , leaving the activity in mesenchyme cells intact ( Figure 3M ) . These observations were confirmed by quantitative measurements of a statistically representative number of embryos ( Figure 3N ) . Ci-Noto1 becomes detectable in notochord precursors at the late gastrula/early neural plate stage ( Figure 1 and [22] ) . Its notochord expression pattern is recapitulated by a 2 . 1-kb genomic fragment from its 5′-flanking region ( Figure S2 ) . We reduced this fragment through progressive truncations ( Figure 4A and 4B ) and found that the 170-bp minimal notochord CRM contains putative binding sites for proteins of the Fox , Ets , and ROR transcription factor families . However , differently from the minimal CRMs described thus far , the 170-bp Ci-Noto1 CRM contains only one putative Ci-Bra binding site . Since previous work had shown that in addition to Ci-FoxA-a and other Fox genes , various members of the Ets and ROR transcription factor families are expressed in the notochord [42] , we individually mutagenized these putative binding sites within the 170-bp CRM; however , no reduction of notochord staining was observed ( Figure 4A ) . Instead , the mutation of the single Ci-Bra binding site caused complete loss of notochord activity ( Figure 4A and 4D ) . Considering the fact that Ci-Noto1 was identified , along with numerous other genes , in a subtractive screen between wild-type and Ci-Bra-overexpressing embryos , we co-electroporated this construct along with Ci-FoxA-a>Bra , the construct that was used to induce mis- and over-expression of Ci-Bra in endoderm , central nervous system ( CNS ) , and notochord through the Ci-FoxA-a promoter region [11] , to test whether the 170-bp CRM was responsive to the ectopic expression of Ci-Bra . These experiments demonstrated that the 170-bp CRM is ectopically activated by the misexpression of Ci-Bra ( Figure 4C ) , and that the mutation of the Ci-Bra site abolished this response ( Figure 4D and 4E ) . To further validate these results , the same mutation of the Ci-Bra binding site was introduced into a longer version ( 1 . 1-kb ) of the CRM , which contains additional Ci-Bra binding sites and directs a more robust staining ( Figure 4F ) . Even within this broader context , the mutation of the single Ci-Bra site was still sufficient to completely obliterate notochord activity , leaving the staining in the CNS and in both papillary and tail neurons unchanged ( Figure 4G ) . This result rules out the possibility that Ci-Bra binding sites found in the longer sequence might be able to compensate for the mutation of the main Ci-Bra binding site , and reinforces the observation that the Ci-Noto1 CRM relies for activity on a single site . A Ci-Bra binding site with a core sequence identical to the one identified in the Ci-Noto1 CRM ( TGGCAC ) was also found to be necessary for the notochord activity of the CRM associated with Ci-Noto9 ( Figures 4H–4L and S2 ) . Ci-Noto9 encodes an ortholog of a transcriptional regulator , FUSE-binding protein [43] and is first detected in notochord cells at the neural plate stage ( Figures 1 and S1 ) , slightly later than Ci-Noto1 . Mutation of the Ci-Bra site inactivated the Ci-Noto9 CRM in notochord cells , leaving intact its ability to sporadically stain mesenchyme and a few muscle cells ( Figure 4I–4L ) . In addition to the previous cases , another notochord CRM associated with a middle-onset gene , Ci-Noto4 ( Figure S2 ) , was found to be dependent upon an individual Ci-Bra binding site , with core sequence TGACAC ( Figure 4M and 4N ) . Mutation of this site similarly obliterated notochord activity in both the minimal 144-bp CRM and in a longer fragment spanning 0 . 88 kb ( Figure 4M ) . None of the other putative binding sites identified in the minimal 144-bp CRM , which included putative Fox and Zn-finger binding sites , was found to substantially contribute to the notochord activity upon individual mutagenesis ( Figure S4 ) . The mutation of the Ci-Bra binding site was sufficient to abolish both notochord staining and the response to ectopically expressed Ci-Bra ( Figure 4N–4Q ) . Interestingly , a Brachyury binding site with an identical core sequence has been previously described in the enhancer region of Xenopus Bix4 , a target of Xbra [38] . Finally , the analysis of the weak 972-bp Ci-Noto8 notochord CRM ( Figure S2 ) identified two Ci-Bra binding sites clustered in tandem arrangement within a 25-bp interval at the 3′-end of this CRM , with core sequences TCACAC and TAACAC ( Figure S5 ) . Truncation of the TAACAC site resulted in the inactivation of the CRM in the notochord ( Figure S5 ) . We conclude that this site is mainly responsible for the notochord activity of this CRM . The experiments described here allowed us to gather a set of minimal CRM sequences and of functional Ci-Bra binding sites , which are shown in Table 1 along with previously published Ci-Bra binding sites . From this comparison , it is evident that among the 16 possible combinations , some core sequences are preferentially represented in the Ciona notochord CRMs identified thus far ( Table 1 ) , and that some core sequences are more frequently encountered in functional Ci-Bra binding sites ( highlighted in bold in Table 1 ) . Of note , 7 out of 16 ( ∼44% ) of the possible core sequences are yet to be found to be required for notochord activity in any CRM . We have aligned the functional core Ci-Bra TNNCAC sequences and their flanking regions and we have compared them to the published binding sites for Brachyury proteins identified in other organisms . The most informative alignments are shown in Table 2 , where we used as a reference the consensus binding site previously identified in Drosophila for the Brachyury ortholog Brachyenteron , which was shown to also be bound by mouse Brachyury [44] . The vast majority of the functional Ci-Bra binding sites identified thus far display a considerable homology with the consensus sequence identified in Drosophila , with the mismatches occurring almost exclusively in the outermost flanking nucleotides ( highlighted in red in Table 2 ) . In particular , the Ciona provisional consensus is richer in pyrimidines at both its 5′ and 3′ ends . We then estimated the distance of the functional Ci-Bra binding sites from the putative transcription start sites of their neighboring genes , in order to identify spatial constraints that might modulate the activity of the functional Ci-Bra binding sites within their genomic context . We referred to the 5′-end of the updated evidence-based KH gene models ( http://ghost . zool . kyoto-u . ac . jp/SearchGenomekh . html#CDNA ) [40] as the transcription start sites ( Table S2 ) . Even taking into account the position bias that characterized the approach used for the identification of the majority of the CRMs , which targeted the 5′-flanking regions , we observed that the Ci-Bra binding sites of single-site CRMs are predominantly located within <600 bp of the respective putative transcription start sites , with the exception of the Ci-ABCC10 Ci-Bra binding site , which lies at position +2 . 3 kb . The cooperatively acting binding sites can be found , on average , at higher distances from the transcription start sites , with the Ci-Noto5 notochord CRM being located >9 kb downstream of the Ci-Noto5 transcription start site ( Table S2 ) . These findings do not reveal evident recurring intervals or other architectural constraints , suggesting that Ci-Bra might be able to activate transcription from its target CRMs regardless of their location within the genomic loci . The Ci-ACL CRM was first identified as a 2 . 15-kb fragment from the 5′-flanking region of the Ci-ACL gene ( Figure S2 ) . This region was subsequently reduced through serial truncations to a 215-bp notochord-specific CRM , which differs from the CRMs previously described since it is devoid of apparent Ci-Bra/T-box TNNCAC binding sites ( Figure 5A and 5B ) . Among the recognizable putative binding sites that were found by scanning this sequence were two Fox sites , a Krüppel-like site and a generic homeodomain site , all of which are clustered within 40 bp at the 3′-end of the 215-bp sequence . Mutations of the putative Fox sites , both individually and combined , did not decrease notochord staining , nor did the mutation in the putative Krüppel-like binding site ( Figure 5A and unpublished data ) , although the truncation of the region containing both Fox sites reduced the notochord activity ( Figure 5A and 5C ) . However , a mutation of 4 bp , which changes the AATTAA core binding site for homeodomain proteins to TTTTGC , was sufficient to abolish over 90% of the notochord activity ( Figure 5A and 5D ) . Finally , the truncation of the whole 40-bp 3′-end region of the CRM was able to completely obliterate notochord activity ( Figure 5A and 5E ) . These results suggest that this 215-bp notochord CRM primarily requires a sequence that resembles the binding site for transcription factors of the homeodomain family for its activity; this sequence might function synergistically with sequences contained in the 40-bp 3′-end region , as well as with sequences found at the 5′-end region of the 215-bp CRM , as truncations of this region also weaken the notochord staining , although to a lesser extent ( unpublished data ) . Similar results were obtained through the dissection of the 2 . 47-kb notochord CRM identified in the 5′-flanking region of the late-onset gene Ci-β4GalT ( Figures 1 and S2 ) . The minimal Ci-β4GalT CRM is also devoid of Ci-Bra binding sites , although it does not apparently rely upon a homeodomain or any other clearly identifiable binding site ( unpublished data ) . These results bring forth the possibility that Ci-Bra might also be controlling the Ci-β4GalT CRM through a transcriptional intermediary , which is likely distinct from the factor that regulates the Ci-ACL CRM . This “relay” mechanism is consistent with the late developmental onset of expression of the genes associated with these CRMs , which begin to be expressed at the late neural plate and at the neurula stage , respectively ( Figure 1 and [22] ) . Once different categories of notochord CRMs were identified , we noticed that single-site minimal CRMs had , on average , the same qualitative “strength” as the multiple-site CRMs , i . e . , they were able to direct intense notochord staining in a large number of embryos . Therefore , we investigated whether these different cis-regulatory mechanisms rather influenced the developmental onsets of the CRMs . We selected representative CRMs associated with direct Ci-Bra targets of the early-onset and middle-onset groups , and to precisely determine their developmental onsets we re-cloned these CRMs upstream of their endogenous promoters . This strategy was employed to avoid possible interference from the early Ci-FoxA-a basal promoter [45] and to recapitulate the natural context of each CRM . We prepared these endogenous promoter constructs for the Ci-thbs3 and Ci-FCol1 CRMs ( Figure 1 ) , which are associated with early-onset genes , for Ci-Noto1 , which is linked to a typical middle-onset gene [22] , and for Ci-β4GalT as a representative late-onset gene . Time-course experiments were carried out for these CRMs , which were all driving the LacZ reporter , in parallel with the 434-bp Ci-Bra>LacZ CRM [27] , which provided a control for temporal onset and notochord-specific staining ( Figure 6A and 6B ) . The results for the Ci-FCol1 and the Ci-Noto1 time-courses are shown in Figure 6C–6G . Embryos electroporated with these constructs were allowed to develop until the 110-cell ( Figure 6A , 6C , and 6E ) , early gastrula ( Figure 6B , 6D , and 6F ) , and mid-gastrula stages ( unpublished data ) , then subjected to WMISH using an antisense RNA LacZ probe and scored for LacZ expression in notochord precursors ( Figure 6G ) . The results revealed a sharp difference between the onset of activity of the Ci-FCol1 CRM , which is first detected in notochord precursors at the 110-cell stage ( Figure 6C and 6G ) and increases at the early gastrula stage ( Figure 6D and 6G ) , and the onset of Ci-Noto1 , whose activity is first detected at the 110-cell stage , weakly and sporadically , in muscle precursors ( Figure 6E , 6G , and inset in 6F ) and in early gastrulae increases in this territory , while remaining absent from the notochord ( Figure 6F and 6G ) . By late gastrulation , ∼10% of total Ci-Noto1 transgenic embryos begin displaying 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside ( X-Gal ) staining ( top right inset in Figure 6F ) . Early onset similar to that of Ci-FCol1 was determined through the time-course of Ci-thbs3 , whose activity was also first detected at the 110-cell stage , as in the case of Ci-FCol1 , although in a lower number of embryos ( 2 . 4% versus 10 . 4% of total stained embryos; unpublished data ) . A late onset was also detected for the single-site Ci-Noto4 CRM , in which case X-Gal staining was detected in 5 . 2% of the stained neural plate embryos ( unpublished data ) . Lastly , we carried out a similar time-course experiment using the Ci-ACL and Ci-β4GalT constructs described in Figures 5 and S2 , and we detected X-Gal staining beginning at the neurula stage ( Figure 6H ) . Comparable results were observed for the Ci-β4GalT CRM cloned on its endogenous promoter ( unpublished data ) . Together , these results indicate that notochord CRMs controlled by Ci-Bra through multiple binding sites display the earliest onset of activity , around the 110-cell stage , while the notochord CRMs controlled by Ci-Bra through a single binding site become active around the mid/late-gastrula stage , and the notochord CRMs controlled by Ci-Bra indirectly are activated around neurulation . The comparison between Ci-FCol1 and Ci-thbs3 shows that the number of cooperative Ci-Bra binding sites in a CRM does not influence the onset of activity . In conclusion , the time-course experiments indicate that the onsets of activity of the notochord CRMs mirror the onsets of expression of the endogenous genes associated with them ( Figures 1 and S1 ) . In an effort to test the general applicability of the cis-regulatory mechanisms identified through this study , we employed various combinations of functional Ci-Bra binding sites to rapidly identify genomic regions with the potential to function as notochord CRMs . We first scanned the genomic loci of known Ci-Bra target genes and identified various candidate enhancer regions , among which the most promising was a 560-bp fragment of the 5′-flanking sequence of Ci-ERM . This region contained four clustered Ci-Bra binding sites , two of which had identical core sequences and arrangement to those identified in the Ci-Noto5 CRM ( Figure 3H ) , although with a narrower spacing ( 45 bp in the case of Ci-Noto5 , 35 bp in the case of Ci-ERM ) . We cloned and tested this region ( Figure 7A ) and found that it was sufficient to direct strong notochord expression in Ciona embryos . Moreover , when the 560-bp CRM was subdivided into two fragments , we found that only the 362-bp proximal construct , containing the two Ci-Bra binding sites identical to those found in the Ci-Noto5 CRM , was sufficient to direct strong notochord staining in vivo ( Figure 7A and 7B ) . Site-directed mutations of the Ci-Bra binding sites showed that ablation of the distal-most site , TAACAC , did not affect notochord activity ( Figure 7C ) , while the disruption of the proximal Ci-Bra binding site , TCACAC , was able to reduce both the intensity and the frequency of the notochord staining ( Figure 7D ) . However , as in the case of Ci-Noto5 , the combined mutation of both Ci-Bra binding sites , TAACAC and TCACAC , completely abolished notochord activity , leaving a residual mesenchyme staining and sporadic muscle staining comparable to the vector background staining ( Figure 7E and unpublished data ) . Through time-course experiments , we found that the Ci-ERM CRM , once transferred to its endogenous promoter region , began its activity by early/mid-gastrulation ( unpublished data ) . As a next step , we attempted to extend these predictions to notochord CRMs which we had previously identified , and whose relationship with Ci-Bra was still unclear . We noticed that another minimal 122-bp notochord CRM , which we had identified through a separate set of experiments in the 5′-flanking region of the Ci-laminin gamma-1 ( Ci-lamc1 ) gene , was enriched in Ci-Bra binding sites and also contained a putative Fox binding site ( Figure 7F and 7G ) . After all these sites were separately mutagenized ( Figures 7F and S6 ) and the results were quantified ( Figure S6 ) , we found that the simultaneous mutation of two Ci-Bra binding sites , with sequences TCACAC and TCGCAC , respectively , was sufficient to completely abolish notochord activity ( Figures 7H and S6 ) , while mutations in the other Ci-Bra binding sites , including another site with an identical TCGCAC core sequence but opposite orientation with respect to the active site , had no visible effect ( Figure S6 ) . These data strongly suggested that Ci-lamc1 could be a notochord gene under the transcriptional control of Ci-Bra . To prove this hypothesis we studied Ci-lamc1 expression in embryos carrying the Ci-FoxA-a>Bra transgene and observed that Ci-lamc1 is highly responsive to ectopically expressed Ci-Bra ( Figure 7I and 7J ) , as is its notochord CRM ( inset in Figure 7J ) . Of note , after this analysis had been completed , Ci-lamc1 was also reported as an early Ci-Bra target by another study [13] . Another notochord CRM had been identified in the genomic locus of Ciona ATP-binding cassette subfamily C member 10 ( Ci-ABCC10 ) through a screen of random Ciona genomic fragments ( unpublished data ) . Interestingly , this is the first evidence , to our knowledge , of the expression of this transporter protein in the notochord . We had previously narrowed the original 2 . 146-kb sequence to a 772-bp fragment ( Figure 7K , 7L , and unpublished data ) via sequence-unbiased truncations , and sequence inspection revealed that this shorter CRM fragment contained six putative Ci-Bra binding sites ( listed in Table 1 ) . We focused the point-mutation analyses on the two sites that had been found to be required for the activity of other notochord CRMs . Surprisingly , the mutation of the TGGCAC site , which is necessary for the Ci-Noto1 and Ci-Noto9 CRMs ( Figure 4 ) , did not affect notochord staining ( Figure 7K and unpublished data ) , while the mutation of the distal TAACAC site was sufficient to completely abolish notochord activity , but left the mesenchyme staining unaffected ( Figure 7K , 7L , and unpublished data ) . On the basis of its dependence upon a single functional Ci-Bra binding site , we predicted this notochord CRM to behave as a middle-onset . To verify this point , we cloned the CRM upstream of its endogenous promoter region and carried out time-course experiments , as previously described . Through WMISH ( unpublished data ) and X-Gal staining , we determined that the onset of activity of this weak CRM is around the neural plate/early neurula stage , when notochord activity is detected in >50% of the stained embryos ( unpublished data ) ; notochord staining increases at the initial tailbud stage ( Figure 7M ) . This is consistent with the timing of Ci-ABCC10 transcript accumulation in the notochord ( Figure S7 ) . To assess the hierarchical relationship of Ci-ABCC10 with Ci-Bra , we studied Ci-ABCC10 expression in embryos carrying the Ci-FoxA-a>Bra transgene and found that this gene is ectopically expressed in response to the ectopic expression of Ci-Bra ( Figure 7N and 7O ) ; a similar behavior was exhibited by the Ci-ABCC10 notochord CRM in Ci-FoxA-a>Bra embryos ( inset in Figure 7O ) . We had tested in previous studies some of the putative Ci-Bra binding sites identified in notochord CRMs for their ability to be bound in vitro by Ci-Bra via electrophoretic mobility shift assays ( EMSA ) ( Table 1 ) ; here we assessed the occupancy of these sites in vivo through ChIP assays on mid-tailbud stage embryos ( Figure 8A ) using a polyclonal Ci-Bra antibody ( Figure 8B ) [46] . To test the specificity of the binding by Ci-Bra , we carried out ChIP over a 10-kb stretch encompassing part of the Ci-FCol1 locus and its neighboring gene ( Figure 8A ) . The results show that the highest peak of Ci-Bra occupancy indeed corresponds to the Ci-FCol1 notochord CRM ( red rectangle in Figure 8A ) ; these results are comparable to those previously published [13] , although not all peaks of Ci-Bra occupancy coincide ( unpublished data ) , likely because our experiments were carried out on embryos at a much later stage ( mid-tailbud versus 110-cell ) . We then proceeded with the ChIP assays of the remaining notochord CRMs , along with adequate controls ( Figure 8C and unpublished data ) . The results demonstrated that in addition to the Ci-tune minimal CRM [41] , which served as one of our positive controls , the Ci-Noto1 , Ci-Noto4 , Ci-Noto5 , Ci-Noto9 , Ci-FCol1 , Ci-lamc1 , and Ci-thbs3 CRMs are also bound in vivo by Ci-Bra . On the other hand , the minimal Ci-ACL notochord CRM , which we have shown to be devoid of Ci-Bra binding sites , was not specifically recognized , similar to the negative control used for these experiments , 18S rRNA gene . These results confirm the findings obtained through the analysis of individual CRMs . No direct relationship was observed between the number of functional Ci-Bra sites and the enrichment of the immunoprecipitated DNA over the input; this might be due to the size of the immunoprecipitated fragments ( ∼200–800 bp on average ) , which likely contain additional sequences that are bound by Ci-Bra but are not required for the activity of the minimal CRMs . We analyzed the VISTA phylogenetic footprints ( http://pipeline . lbl . gov/cgi-bin/gateway2 ) obtained by comparing the sequences of the notochord CRMs identified in this study between C . intestinalis and C . savignyi to assess the extent of their evolutionary conservation . We found that the most highly conserved notochord CRMs are Ci-Noto9 ( single-site ) and Ci-FCol1 ( multiple-site ) ( Figure S8 ) . In the case of Ci-Noto9 , the conservation of the functional Ci-Bra site and its flanking sequences between the two species is complete . In the case of the other single-site CRMs , we found that in the C . savignyi Noto1 genomic region corresponding to the C . intestinalis CRM there was a single change in the TNNCAC core sequence , which did not disrupt the putative Brachyury binding site ( TGGCAC to TGCCAC ) . Sequence comparisons for Ci-Noto4 showed a disruption of the functional Ci-Bra site ( s ) sequence found in the corresponding location ( TGACAC to TCACGC ) but a surprising conservation of the dispensable TCCCAC site , which suggests that this sequence might be of some relevance in C . savignyi . Also in the case of Ci-Noto8 , the main Ci-Bra binding site was not conserved ( TAACAC to TAACAT ) , although some of the other putative binding sites found in the CRM were shared between the two Ciona species ( unpublished data ) . The Ci-thbs3 CRM displays a complete conservation of one of its three cooperative sites , TAACAC , but disruption of the other two sites , TCGCAC and TGGCAC . As for the multiple CRMs , in the case of Ci-FCol1 , the dispensable TTTCAC site ( Figure 3A ) is not conserved , while of the two cooperative sites , one is entirely conserved , but the other has a single nucleotide substitution in C . savignyi ( TATCAC to TCTCAC ) , which does not disrupt the TNNCAC sequence and might be therefore a functional binding site in this species . These observations correlate with the results of the point mutation analysis ( Figure 3B–3E ) , which shows that the mutation of the most conserved site , TAACAC , is more effective than the mutation of the less conserved one . The Ci-ERM and Ci-ACL CRMs showed scattered interspecific conservation , although not in the regions directly corresponding to the functional Ci-Bra binding sites or the putative homeodomain binding site , respectively ( unpublished data ) . Finally , for Ci-Noto5 we did not find any informative sequence alignment in the regions of the C . savignyi locus corresponding to the C . intestinalis CRM , and the minimal Ci-ABCC10 , Ci-β4GalT , and Ci-lamc1 CRMs were poorly conserved overall ( unpublished data ) . Our findings on multiple and single-site CRMs directly controlled by Ci-Bra imply that the conversion of a multiple-site CRM to a single-site should suffice to delay the developmental onset of its activity , thus turning it from an early-onset to a middle-onset CRM . To prove this point , we performed time-course experiments using the Ci-lamc1 notochord CRM ( Figures 7F–7J and S6 ) , after re-cloning it upstream of its endogenous promoter . On the basis of its reliance upon two functional Ci-Bra sites ( Figure 7F ) , we predicted this CRM to display an early onset of activity . In support of this hypothesis , >80% of mid-gastrula embryos electroporated with this construct showed X-Gal staining in notochord precursors ( Figure 9A and 9F ) , indicating that by the early gastrula stage transcription of LacZ directed by the Ci-lamc1 CRM/promoter has already begun . This latter inference was confirmed by LacZ WMISH experiments ( unpublished data ) . These experiments indicate that the developmental onset of the Ci-lamc1 CRM/promoter is comparable to that observed in the case of Ci-FCol1 ( Figure 6C , 6D , and 6G ) . At the neural plate stage , the percentage of Ci-lamc1 transgenic embryos with notochord staining was slightly higher ( Figure 9D and 9F ) . However , when we tested constructs carrying mutations in either one of the Ci-Bra functional sites ( Ci-lamc1-T1M and Ci-lamc1-T4M , Figure 9B and 9C ) we found that the number of embryos showing notochord staining at the mid-gastrula stage had dropped below 30% compared to the wild-type CRM ( Figure 9F ) . By the neural plate stage , the number of embryos showing notochord staining had significantly increased in both mutants ( Figure 9E , 9F , and unpublished data ) , and had practically reached the levels observed in mid-tailbud stage embryos for these mutants ( Figure 9F ) . Figure 10 summarizes our findings and proposes a correlation between the number of functional Ci-Bra binding sites and the developmental onsets of the notochord CRMs and the genes linked to them . Notochord CRMs controlled directly by Ci-Bra through multiple functional sites begin their activity between the 110-cell stage and early gastrula , while direct target CRMs controlled by Ci-Bra through a single site become active between late gastrula and neural plate . Finally , notochord CRMs controlled indirectly by Ci-Bra through transcriptional intermediaries begin their activity at neurulation .
Among the validated Ci-Bra targets expressed in the notochord , the earliest is the planar cell polarity gene Ci-pk , which is required for the establishment of notochord cell polarity [48] and for intercalation [23] . Ci-ERM , which is first detected in the notochord a few cell divisions after Ci-pk , has been shown to be required for notochord elongation [23] and lumen formation [49] . Ci-Noto4 is first detected in notochord cells at the neural plate stage , and it is also required for midline intercalation [50] , [51] , as is the late-onset gene Ci-ACL , which plays a role also in medio-lateral polarization of notochord cells [23] . Ci-lamc1 encodes a putative ortholog of human Laminin gamma 1 , which is also found in the notochord remnants of the intervertebral discs in human embryos [52] . Ci-ABCC10 had not been previously detected in notochord cells , and encodes an anionic pump that might be involved in lumen formation , the terminal step of notochord differentiation [30] . Interestingly , this gene is not in the list of putative Ci-Bra targets identified via genome-wide studies of chromatin occupancy in early embryos [13] , most likely because its expression begins around neurulation . These findings underscore the breadth of Brachyury functions , which encompass all stages of notochord formation , and explain the deleterious effects of its inactivation on the development of this structure in widely different chordates , from ascidians [3] to mice [53] . We sought to shed light on the molecular mechanisms that enable Ci-Bra to sequentially deploy its target genes through the systematic characterization of their notochord CRMs . The dissection of notochord CRMs associated with representative Ci-Bra targets allowed their categorization on the basis of the mechanisms employed by Ci-Bra to control their minimal sequences . We found that four CRMs , Ci-Noto1 , Ci-Noto9 , Ci-Noto4 , and Ci-ABCC10 , are controlled by Ci-Bra through individual binding sites , which are necessary to elicit notochord activity and to mediate the response to ectopically expressed Ci-Bra . These Ci-Bra binding sites have either the core sequence TGGCAC ( Ci-Noto1 and Ci-Noto9 ) or TGACAC ( Ci-Noto4 ) , while the Ci-Bra binding site in the Ci-ABCC10 CRM has a TAACAC core . Of note , our results show that in the case of these CRMs , additional Ci-Bra binding sites that might be present in the vicinity of the single functional site are unable to compensate for its loss . In addition to single-site Ci-Bra target CRMs , we have also identified direct targets that are controlled through two cooperative Ci-Bra binding sites , Ci-FCol1 , Ci-Noto5 , Ci-ERM , and Ci-lamc1 . This class also includes the previously characterized Ci-leprecan notochord CRM ( [26] and unpublished data ) , the Ci-trop notochord CRM , which mainly relies upon a TCGCAC site [21] but also on an adjacent TATCAC site , which alone is not sufficient for activity ( our unpublished results ) . Finally , the Ci-thbs3 CRM is controlled by Ci-Bra through three binding sites and the Ci-pk CRM likely relies upon multiple Ci-Bra binding sites and additional sequences . Through the analysis of the Ci-tune notochord CRM , we had previously identified another class of direct Ci-Bra target CRMs , which are controlled synergistically by Ci-Bra and Ci-FoxA-a [41] . Notably , in the present study , we found that most notochord CRMs contained putative Ci-FoxA-a binding sites in addition to the Ci-Bra binding sites . However , site-directed mutation analyses of the Ci-FoxA-a binding sites that were related to the sites found in the Ci-tune CRM ( TRTTTAY core ) did not reveal an evident role in notochord activity . Nevertheless , it is conceivable that these sites might be used in vivo by Fox proteins , which are known to possess a pioneer chromatin-opening activity [54] , to increase the accessibility of the CRMs within their native genomic context , or that some divergent Fox binding sites might be contributing to notochord activity . Recent genome-wide ChIP-chip studies have elucidated the mesodermal gene regulatory network presided over by one of the zebrafish Brachyury orthologs , No tail , leading to the identification of an in vivo binding site for this transcription factor , TCACACCT [12] , [55] , which matches the half-site previously identified for mouse Brachyury [56] and the Xbra binding site identified 936 bp upstream of the promoter region of Xenopus eFGF [37] . The present study revealed a considerable heterogeneity in the functional sequences found in direct Ci-Bra targets , as well as the lack of considerable homology in the sequences flanking Ci-Bra binding sites with identical cores . Nevertheless , most of the Ci-Bra functional sequences identified by this and our previous studies conform , albeit to a different extent , to the 12-bp consensus identified for Brachyenteron and vertebrate Brachyury proteins by [44] , with the Ci-Bra binding sites seeming prone to higher variability in the nucleotides more distant from the central 6-bp core sequence . Similarly to its orthologs characterized in other model systems , Ci-Bra is able to bind palindromic sites , possibly in the form of a dimer [21] , with the dimerization likely being mediated by the evolutionarily conserved PDSPNF amino acid motif within its T-domain [44] . However , like previously characterized Ci-Bra target CRMs , the CRMs reported here predominantly rely upon half-sites . Only the Ci-Noto1 functional Ci-Bra binding site , one of the cooperative Ci-Noto5 sites and one functional Ci-Bra site found in the proximal region of the Ci-lamc1 CRM display an incomplete palindromic arrangement , which nevertheless seems dispensable for their function , as indicated by the results of individual mutations . We tested the general validity of the molecular mechanisms identified through these studies by using the sequences of the functional Ci-Bra binding sites to scan either the loci of other bona fide Ci-Bra target genes , or other notochord CRMs identified through cloning of random genomic sequences . Among the interesting clusters of putative Ci-Bra sites that we identified within these sequences , a region found upstream of Ci-ERM showed a striking similarity with the Ci-Noto5 notochord CRM and was therefore cloned and tested in vivo , and resulted capable of directing strong notochord staining . Mutation analyses showed that also this predicted CRM relies upon two Ci-Bra binding sites , whose core sequences and arrangement are identical to those of the functional Ci-Noto5 sites . Of note , the expression patterns of Ci-Noto5 and Ci-ERM in the notochord are remarkably similar ( Figure S1 ) . The spacing between the Ci-Bra binding sites varies by 10 bp ( 35 bp for Ci-ERM and 45 bp for Ci-Noto5 ) , a full helical turn [57] , which seemed suggestive of some flexibility in their spacing . However , when we decreased the distance between the Ci-Bra binding sites in the Ci-Noto5 CRM to match the spacing found in the Ci-ERM CRM , we observed a loss of notochord activity ( unpublished data ) . When we attempted to predict the functional sites of notochord CRMs identified either through random testing of genomic sequences ( Ci-ABCC10 ) or by position-biased cloning ( Ci-lamc1 ) , again we observed a striking context-dependent difference in the functional relevance of binding sites with identical cores . In fact , in the case of the Ci-ABCC10 notochord CRM the site with the TGGCAC core , which alone is necessary for the activity of both Ci-Noto1 and Ci-Noto9 CRMs , turned out to be fully dispensable , while the mutation of a TAACAC core site was sufficient to abolish notochord activity . Similarly , within the Ci-lamc1 CRM , the mutations of two identical TCGCAC core sites had very different results , with the distally located site being completely dispensable and the proximal site being highly relevant for notochord activity . Despite the context dependence and the variability in the sequence and relevance of the core Ci-Bra binding sites , a common consensus sequence for functional Ci-Bra binding sites is beginning to emerge from these results; this sequence displays evolutionary conservation with the 12-bp sequence bound in vitro by both chordate and protostome Brachyury orthologs [44] . A considerable level of structural flexibility was also observed when the distances of functional Ci-Bra binding sites from the respective transcription start sites were compared . Moreover , our preliminary analyses indicate a remarkably wide degree of variability in the evolutionary conservation of the minimal CRM sequences between the two Ciona species . However , it is still possible that some of the Ci-Bra binding sites that are not evidently conserved might be functionally replaced by related sites found in the C . savignyi sequences , as we previously hypothesized in the case of the functionally conserved Ci-tune CRM [41] . In addition to identifying notochord CRMs directly targeted by Ci-Bra , this study has also led to the discovery of two notochord CRMs that are physically associated with bona fide Ci-Bra target genes but are devoid of Ci-Bra binding sites in their minimal sequences . These CRMs are associated with two late-onset Ci-Bra targets , Ci-ACL , which is activated in notochord at low levels beginning around the late neural plate stage ( [22] and our unpublished results ) and Ci-β4GalT , which is detected in notochord cells starting at the neurula stage . Therefore it seems conceivable that these minimal CRMs might be controlled by Ci-Bra indirectly through transcriptional intermediaries . We have previously shown that transcription factors of different families are expressed in the Ciona notochord following the onset of Ci-Bra expression , and that the expression of several of these genes is controlled by Ci-Bra [25] , [31]; these reports , along with the results of genome-wide screens [13] , [42] , [58] provide a number of candidate activators for these CRMs . The lack of binding by the anti-Ci-Bra antibody that we observed through the ChIP assays suggests that in addition to lacking canonical Ci-Bra binding sites , the Ci-ACL notochord CRM is not likely to contain low-affinity , non-canonical Ci-Bra binding sites such as the “type B” sites described in Drosophila , which are occupied by Brachyenteron but are unable per se to activate transcription [44] . Rather , this might be the case for the Ci-β4GalT CRM , which yielded a relatively noisy ChIP signal ( unpublished data ) . These latter results are consistent with reports in Drosophila that show that transcription factors can bind up to thousands of genomic regions which often are not involved in transcriptional events and might be non-functional [59] , [60] . The results of this study indicate that the notochord CRMs controlled by Ci-Bra through multiple sites are frequently linked to genes that begin to be expressed early during notochord development , around the early/mid-gastrula stage , and are usually expressed at high levels in these cells . Instead , the notochord CRMs controlled by Ci-Bra through single sites are usually associated with genes that are expressed in the notochord at low levels starting around the late gastrula/neural plate stage . This hypothesis is confirmed by the observation that the removal of one of the functional Ci-Bra binding sites from the multiple-site Ci-lamc1 CRM results in a delay in the developmental onset of its notochord activity . One possible scenario arising from these observations suggests that by the 110-cell stage , at a time when a substantial accumulation of Ci-Bra in the nuclei is detected ( our unpublished results ) , the synergistic binding of Ci-Bra molecules to the multiple-site CRMs might trigger faster changes in the chromatin state and in the rate of transcriptional activation , as compared to the binding to an individual main site . Hence , transcripts for genes controlled through multiple-site CRMs are detected earlier and in larger amounts than those linked to single-site CRMs . These results might enable predictions of the expression patterns of Brachyury target genes in Ciona and possibly in other chordates . To our knowledge , this represents the first report of a correlation between the number of functional Brachyury binding sites found in minimal CRMs and the onset of transcription of their target genes in a chordate . In non-chordate model systems , it is noteworthy that another transcriptional activator , the ortholog of FoxA2/HNF3beta , is able to modulate the onset of gene expression in the pharynx of the nematode Caenorhabditis elegans through binding sites with different affinity [61] . High-affinity binding sites are associated with early expression and can be activated by low levels of FoxA2/HNF3beta during early organogenesis [61] . This mechanism therefore resembles the spatial read-outs observed in the Drosophila embryo in response to the anterior-posterior and dorsal-ventral gradients of the morphogens Bicoid [62] and Dorsal [63] , respectively . Our findings on the context-dependent behavior of Ci-Bra binding sites with identical TNNCAC core sequences suggest that additional sequences , rather than the core sequences per se , might be modulating the binding affinity of Ci-Bra for its sites . Preliminary evidence suggests that a subset of the TNNCAC Ci-Bra binding sites found in single-site notochord CRMs share short flanking sequences , which could be responsible for their indispensable function ( DSJ-E and ADG , unpublished observations ) . We conclude that even though Ci-Bra controls several of its target genes via direct binding , this shallow branch of its gene regulatory cascade is modulated at the level of individual CRMs by the differential mechanisms identified through this study , to ultimately result in a gradated developmental response .
Adult C . intestinalis were purchased from Marine Research and Educational Products ( M-REP ) and kept in an aquarium in recirculating artificial sea water at ∼18°C . Fertilization and electroporations were carried out as previously described [24] . After incubation with X-Gal , only well-developed embryos that showed β-galactosidase staining in any tissue were counted . Each construct was electroporated and scored in triplicate , using embryos obtained from different batches of animals collected from the same location . All genomic fragments were PCR-amplified from C . intestinalis genomic DNA purified from sperm of a single individual and cloned into the pFBΔSP6 vector [24] , either directly or after an intermediate cloning step into the vector pGEM-T ( Promega ) . A list of the oligonucleotides employed for PCR amplifications of the initial fragments is provided in Table S1; sequences of the oligonucleotides employed for the construction of the truncations and point mutation constructs are available upon request . The Ci-ACL and Ci-ABCC10 notochord CRMs were identified by screening random genomic regions for cis-regulatory activity , essentially as previously described [64] . Digoxigenin-labeled antisense RNA probes were synthesized in vitro from existing expressed sequence tags ( ESTs ) generated and kindly distributed by the Satoh lab [65] . Clone GC29n22 was used for Ci-Noto8; GC07k01 for ABCC10; GC42d12 for laminin gamma-1 chain precursor ( aka laminin B2; abbreviated as Ci-lamc1 ) ; GC01e23 for Ci-FCol1; GC10b11 for Ci-Noto5; GC28e19 for Ci-ERM; GC33g19 for Ci-Noto4 and GC28p14 for Ci-Noto9 . EST 84P15 ( Beckman Coulter Genomics; kindly made available by the Lemaire lab ) was used for Ci-β4GalT . The Ci-thbs3 probe was cloned as published previously [25] . Probe labeling and WMISH experiments were performed as previously reported [24] . Immunohistochemistry was performed on either wild-type or transgenic C . intestinalis embryos carrying the Ci-Bra>GFP construct [27] , essentially as previously described [66] , using the published anti-Ci-Bra polyclonal antibody [46] . After an overnight incubation at 4°C with a goat anti-rabbit Alexa Fluor 546 fluorescent antibody ( Invitrogen ) in PBS , the embryos were washed six times for 10 min in PBS , mounted with Vectashield mounting medium containing DAPI ( Vector Laboratories ) and imaged using a Leica DMR microscope . ChIP assays were carried out as previously published [46] . qPCR was performed in triplicate on each of the biological replicates , using SYBR green ( USB ) in an Applied Biosystems ( ABI ) Prism 7700 Real-Time qPCR thermocycler . The biological replicates were either two or three in each ChIP-qPCR experiment . To obtain standard curves , duplicates of 5- , 50- , 500- , and 5 , 000-fold diluted Ciona genomic DNA samples were used , starting from 20 ng . The percent input and standard deviation were averaged from immunoprecipitated/input WCE scores . p-Values were calculated using a two-tailed Student's t test . The negative control was the 18S rRNA gene ( National Center for Biotechnology Information [NCBI] accession number: AJ250778 [http://www . ncbi . nlm . nih . gov]; JGI gene model: gw1 . 7761 . 2 . 1 ) . | Transcription factors control where and when gene expression is switched on by binding to specific stretches of DNA known as cis-regulatory modules ( CRMs ) . In this study , we investigated the architecture and composition of CRMs that direct gene expression in the notochord—a transient rod-like structure found in all embryos that belong to the phylum chordata , which includes humans . Here we used the sea squirt Ciona , a simple chordate , and analyzed how the transcription factor Brachyury ensures the appropriate deployment of its target genes at specific times during the sequential steps of notochord formation . We compared CRMs found in different notochord genes downstream of Brachyury , expecting to find genes associated with greater numbers of Brachyury binding sites to be expressed at higher levels . To our surprise , we found instead that a higher number of functional Brachyury binding sites is typical of CRMs associated with genes that are expressed early in notochord development , while single-site CRMs are characteristic of genes that are turned on during the intermediate stages of this process . Finally , CRMs associated with genes expressed late in notochord development do not contain functional Brachyury binding sites but are controlled by Brachyury indirectly , through the action of intermediary transcription factors . These differences explain how a transcription factor that is present at all stages in a certain cell type can generate a sequential transcriptional output of gene expression . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2013 | Functional Brachyury Binding Sites Establish a Temporal Read-out of Gene Expression in the Ciona Notochord |
Muscle morphogenesis is tightly coupled with that of motor neurons ( MNs ) . Both MNs and muscle progenitors simultaneously explore the surrounding tissues while exchanging reciprocal signals to tune their behaviors . We previously identified the Fat1 cadherin as a regulator of muscle morphogenesis and showed that it is required in the myogenic lineage to control the polarity of progenitor migration . To expand our knowledge on how Fat1 exerts its tissue-morphogenesis regulator activity , we dissected its functions by tissue-specific genetic ablation . An emblematic example of muscle under such morphogenetic control is the cutaneous maximus ( CM ) muscle , a flat subcutaneous muscle in which progenitor migration is physically separated from the process of myogenic differentiation but tightly associated with elongating axons of its partner MNs . Here , we show that constitutive Fat1 disruption interferes with expansion and differentiation of the CM muscle , with its motor innervation and with specification of its associated MN pool . Fat1 is expressed in muscle progenitors , in associated mesenchymal cells , and in MN subsets , including the CM-innervating pool . We identify mesenchyme-derived connective tissue ( CT ) as a cell type in which Fat1 activity is required for the non–cell-autonomous control of CM muscle progenitor spreading , myogenic differentiation , motor innervation , and for motor pool specification . In parallel , Fat1 is required in MNs to promote their axonal growth and specification , indirectly influencing muscle progenitor progression . These results illustrate how Fat1 coordinates the coupling of muscular and neuronal morphogenesis by playing distinct but complementary actions in several cell types .
Neuromuscular morphogenesis involves complex tissue interactions simultaneously governing the generation of skeletal muscles and the production of the somatic motor neurons ( MNs ) that innervate them . Both processes independently rely on the execution of a generic regulatory program sequentially leading to cell fate determination , differentiation , and functional diversification [1–3] . These regulatory events are coupled with dynamic morphogenetic events leading to the definition of multiple muscle shapes , and the simultaneous topographical wiring of motor axonal projections . During muscle morphogenesis , myogenic progenitors migrate collectively from their origin in either somites or cephalic mesoderm to their final position in the limb , trunk , or face [4] . Trunk and limb connective tissues ( CTs ) , which derive from lateral plate mesoderm , provide instructive signals for incoming somite-derived myogenic cells [5] . Reciprocal signals exchanged by muscle progenitors and the mesenchymal environment pattern muscle shapes by allowing the definition of fiber orientation , muscle attachment sites , and tendon development [5 , 6] . Muscle progenitors subsequently engage in a complex regulatory process , through which they give rise to differentiating cells called myocytes , in charge of producing the contractile apparatus [2 , 3 , 7] . Myocytes then fuse with each other to form multinucleated muscle fibers . The process of muscle growth is determined by a tightly regulated balance between progenitor expansion and production of myocytes and differentiating muscle fibers [2 , 3] . In parallel with muscle morphogenesis , MNs emit axons , which grow in peripheral tissues , selecting a trajectory by responding to multiple guidance cues , allowing them to find their target muscles , within which they ultimately establish a selective pattern of intramuscular arborization [8] . During these processes , axons of MNs and migratory myogenic progenitors follow converging trajectories , along which they simultaneously probe the environment and respond to instructive cues , as evidenced by classical embryological studies [5 , 9–11] . Multiple signals emitted by peripheral tissues to instruct MNs’ specification and axonal pathfinding have been identified [12] . Likewise , some recent discoveries have started shedding light on how non-myogenic CTs exert their influence on muscle patterning [13] . In spite of such advances , what controls the coordinated behavior of the two cell types to orchestrate neuromuscular morphogenesis has not been studied . In the present study , we have examined the possibility that the Fat1 cadherin , a planar cell polarity ( PCP ) molecule involved in tissue morphogenesis , could contribute to coordinate muscular and neuronal morphogenesis . We recently identified Fat1 as a new player in muscle morphogenesis that influences the shape of subsets of face and shoulder muscles , in part by polarizing the direction of collectively migrating myoblasts [14] . Fat1 belongs to the family of Fat-like atypical cadherins [15] . Together with their partner cadherins Dachsous , Fat-like cadherins are involved in regulating coordinated cell behaviors , such as planar cell polarity ( PCP ) in epithelia [15–17] , collective/polarized cell migration [18–20] , and oriented cell divisions [21 , 22] . Through these actions , Fat/Dachsous signaling modulates cell orientation , junctional tension [23 , 24] , and microtubule dynamics [25] , thereby influencing the mechanical coupling between cell behavior and tissue shapes [17] . Aside from their canonical role in regulating the PCP pathway [16] , Fat-like cadherins also control tissue growth via the Hippo pathway [26 , 27] and were recently found to contribute to mitochondria function and metabolic state by interacting with the electron transport chain [28 , 29] . In vertebrates , the most studied Fat homologue , Fat4 , plays multiple functions in development to coordinate kidney [22 , 30–32] , skeletal [21] , heart [33] , or neural morphogenesis [19 , 34 , 35] . The other family member , Fat1 , is known for playing complementary functions during kidney [36 , 37] , muscle [14] , and neural [38] morphogenesis . Here , to explore the mechanisms underlying the coupling of neural and muscular morphogenesis and to assess how Fat1 contributes to this process , we focused on a large flat subcutaneous muscle , the cutaneous maximus ( CM ) , which expands under the skin by planar polarized myoblast migration . This muscle is linked to its cognate spinal MN pool through the selective production by the CM muscle of glial cell line-derived neurotrophic factor ( GDNF ) , a secreted growth factor required to control specification of the corresponding MNs [39] . Unlike limb-innervating MNs , which are born with an intrinsic molecular program specifying their anatomical characteristics [9 , 11] , CM-innervating MNs are incompletely specified at the time they first send their axons and are dependent on extrinsic signals from peripheral tissues [39–41] . GDNF , produced first by the plexus mesenchyme and subsequently by the CM and latissimus dorsi ( LD ) muscles , is perceived by axons of a competent population of MNs when they reach the plexus and as they continue growing along the expanding muscle [39 , 42] . GDNF acts through the Ret tyrosine kinase receptor in a complex with a GPI-anchored co-receptor Gfra1 [43–45] by inducing expression of the transcription factor Etv4 ( Ets variant gene 4 , also known as Pea3 ) in the MN pools innervating the CM and LD muscles [39] , in synergy with another mesenchyme-derived factor , hepatocyte growth factor ( HGF ) [46 , 47] . Etv4 in turn influences MN cell body positioning , dendrite patterning , intramuscular axonal arborization , and monosynaptic reflex circuit formation [40 , 41] . Whereas MN cell body positioning is thought to involve the regulation of the Cadherin code by Etv4 [41 , 48] , patterning of sensory-motor connections is accounted for by the Etv4-regulated Sema3E , acting as repellent cue for sensory neurons expressing its receptor PlexinD1 [49 , 50] . GDNF can also directly influence axon pathfinding , as demonstrated in the context of dorsal motor axon guidance [51–54] or of midline-crossing by commissural axons [55] , and is subsequently required for survival of subsets of MNs [56 , 57] . We found that inactivation of the Fat1 gene has a profound impact on the assembly of the CM neuromuscular circuit , affecting not only the rate of subcutaneous expansion of the CM muscle by progenitor migration and the subsequent rate of differentiation but also the acquisition of identity and projection patterns of their cognate MNs . Intriguingly , in addition to its function in myogenic cells [14] , Fat1 is also expressed in muscle-associated mesenchymal cells and in the MN subset corresponding to the Fat1-dependent CM muscle . Through a series of genetic experiments in mice , we have selectively ablated Fat1 functions in the distinct tissue types in which it is expressed along this neuromuscular circuit and assessed the impact on muscular and neuronal morphogenesis . We uncovered two novel Fat1 functions in the mesenchymal lineage and in MNs , which synergize to coordinate the development of the CM neuromuscular circuit . Fat1 ablation in the mesenchymal lineage causes severe non–cell-autonomous alterations of CM morphogenesis , disrupting expansion of the GDNF-expressing CM progenitors and the subsequent processes of muscle fiber elongation , CM motor innervation , and the acquisition of MN pool identity . This identifies the mesenchymal lineage as a source of Fat1-dependent muscle- and MN-patterning cues . The neural consequences of mesenchyme-specific Fat1 ablation partially mimic the effects of Gdnf or Etv4 mutants and can be aggravated by further reducing Gdnf levels genetically . In parallel , we find that MN-Fat1 is required cell-autonomously for motor axon growth and MN specification . Unexpectedly , MN-specific Fat1 ablation also influences myogenic progenitor spreading in a non–cell-autonomous manner , demonstrating a reverse influence of MNs on muscle morphogenesis . Collectively , these data show that Fat1 exerts complementary functions in several tissue types along the circuit , each of which contributes to neuromuscular morphogenetic coupling through distinct mechanisms , coordinating the adaptation of MN phenotype to muscle shape .
In this study , we have used two phenotypically equivalent constitutive knockout alleles of Fat1 ( summarized in S1 Table ) : the first allele ( Fat1- allele , also known as Fat1ΔTM ) derives from the conditional Fat1Flox allele by Cre recombinase ( CRE ) -mediated excision of the floxed exons encoding the transmembrane domain , thus abrogating the ability of the Fat1 protein to transduce signals [14] . The second allele ( Fat1LacZ allele ) is a genetrap allele of Fat1 , in which the inserted transgene results in expression of a Fat1-β-galactosidase chimeric protein , in the endogenous domain of Fat1 expression [14] . Both alleles cause comparable phenotypes [14] . We focused on one of the muscles affected by constitutive loss of Fat1 functions , the CM muscle ( Fig 1A and 1B ) , a flat muscle emerging from the forelimb plexus ( also called brachial plexus ) . Completing our previous analysis [14] , we first followed the establishment of the CM muscle and its evolution during development by using a transgenic line ( the Mlc3f-nLacZ-2E line , later referred to as MLC3F-2E , S1 Table ) expressing a nuclear LacZ reporter in differentiating muscle cells , thus behaving as a reporter of sarcomeric gene expression [58 , 59] . This line was combined to the constitutive Fat1- allele , and wild-type and mutant embryos carrying the transgene were stained with X-gal . We previously reported that this approach reveals in Fat1-/- embryos ( 1 ) myocytes dispersed in the forelimb region and ( 2 ) a supernumerary muscle in ectopic position in the upper part of the forelimb ( see Fig 6 in ref [14] , S1A and S1B Fig ) . Both phenotypes can be quantified ( S1B Fig ) and can also be observed on histological sections using additional markers of muscle development , such as Pax7 , to label myogenic progenitors , and the sarcomeric protein myosin heavy chain 1 ( Myh1 ) , to visualize muscle fibers ( S1C and S1D Fig ) . The evolution of the CM muscle follows a very specific growth pattern . Muscle fibers can be viewed as “chains” of MLC3F-2E positive nuclei . In the CM , these fibers appear to originate from the brachial plexus , just posterior to the forelimb bud , and to extend posteriorly from this point . Individual fibers spread in a radial manner under the skin to form a fan-shaped structure , ranging from dorsally directed fibers to ventrally directed fibers , the median direction being approximately horizontal . As development proceeds , the length of such fibers ( yellow arrows , Fig 1A and 1B ) increases posteriorly , and the overall area covered by β-galactosidase-positive fibers ( red dotted area , Fig 1A and 1B ) expands . The rate of CM expansion can be measured by following the area containing differentiated fibers , plotted relative to the area of body wall ( BW ) muscles ( yellow dotted area on upper pictures in Fig 1A and 1B ) , used as a proxy for the embryo stage , as these muscles are not affected by the mutation . In wild-type MLC3F-2E+ embryos , the CM area follows a positive linear evolution , strongly correlated with expansion of the BW muscle area ( Fig 1C , left plot ) . In Fat1-/-; MLC3F-2E+ embryos , CM expansion appears severely delayed , although not abolished , with a growth rate reduced by more than 2-fold compared to control embryos ( Fig 1C , left plot ) . As a result , at comparable stages , the differentiated CM area is systematically smaller in absence of functional Fat1 . Given the highly dynamic nature of CM expansion over just one day , in order to pool data from all embryos examined , the ratio between the CM and BW areas was calculated and normalized to the median ratio of control littermate embryos , corresponding to 100% ( Fig 1C , right plot ) . Overall , loss of Fat1 functions causes the area of differentiated CM to be reduced to a median value of about 32% compared to control embryos . Interestingly , observation of older embryos ( Fig 1B ) reveals that fiber length and LacZ-positive nuclei density appear more drastically reduced in the ventral part of the CM than in the dorsal part . We previously documented that at later stages , in this ventral area , occurrence of misoriented fibers crossing each other can frequently be observed [14] . We next took advantage of the fact that the CM is a selective source of GDNF , thus offering an excellent marker to follow development of this muscle [39 , 42] . Alterations of the CM muscle shape resulting from disrupted Fat1 functions can be visualized by following β-galactosidase activity in embryos carrying a GdnfLacZ allele ( Fig 1D and 1E , S1 Table ) . We therefore produced embryos carrying one copy of the GdnfLacZ allele in wild-type or Fat1-/- contexts and performed staining with Salmon-Gal , a substrate more sensitive than X-gal , adapted to the low level of Gdnf expression ( see Materials and methods ) . GdnfLacZ expression can be detected as early as E11 . 5 , prior to the emergence of the CM , at the level of the plexus mesenchyme ( at fore- and hind limb levels ) , where it serves to guide motor axons and instruct them of their identity [39 , 42 , 54] . Gdnf expression is then detected in the CM and in the underlying LD muscles as they emerge ( around E12 . 0 ) from the brachial plexus [14 , 39] . The LD muscle is not visible on our pictures because it is hidden by the CM , but it can be recognized on embryo sections . From that stage onward , these muscles progress by migrating under the skin in a posterior direction , radiating from their point of origin . As development proceeds , the area occupied by CM progenitors expands ( and can be viewed through the skin by transparency in whole embryos ) . We focused on the time window when most of the subcutaneous progression is occurring ( E12 . 0–E12 . 75 ) . To analyze the rate of expansion , the GdnfLacZ-positive area ( white dotted area in Fig 1D and 1E ) was plotted relative to the trunk length , which is used as a value that increases regularly as the embryo grows , thus reflecting the stage of development . At any stage examined , the area covered by GdnfLacZ-positive cells is smaller than in control embryos ( Fig 1D–1F ) . The rate of CM expansion is significantly reduced in Fat1-/- embryos , with a median area reduced to 66% of controls . As seen with MLC3F-2E , this effect also appears more pronounced in the ventral part of the CM . Furthermore , staining intensity in the GdnfLacZ-positive zone behind the progression front appears reduced in Fat1 mutants ( compare intensity along the vertical blue dotted line in Fig 1E ) . Rather than reflecting a reduction in the level of GdnfLacZ expression per progenitors , this effect appears to result from a reduction in the density of LacZ-expressing cells in this front of migration . This effect on Gdnf expression can also be observed in the context of the other null allele of Fat1 ( Fat1LacZ ) by following Gdnf expression by in situ hybridization on embryo sections ( S2A Fig ) . In this context , reduced thickness of the Gdnf-expressing cell layer is observed in posterior CM sections of Fat1LacZ/LacZ embryos , reflecting a reduced number of Gdnf-expressing cells . The reduced density in muscle progenitors in constituting the CM is also visualized by following markers of myogenic cells or subsets of migratory muscle populations such as MyoD , Six1 , and Lbx1 on sections at posterior levels ( S2B Fig ) . A reduced Gdnf expression level was also detected at the level of the plexus mesenchyme in Fat1LacZ/LacZ embryos ( S2A Fig ) . Overall , we confirm using two independent alleles that Fat1 is required for the development of the CM muscle . The CM muscle represents an excellent example in which to study the coupling between muscle morphogenesis and neuronal specification . Given the strong effect of Fat1 loss-of-function on expansion and differentiation of the CM muscle , we next wondered if changes could also be observed in the pattern of innervation . E12 . 5 control and Fat1 mutant embryos were therefore stained by whole-mount immunohistochemistry ( IHC ) with an anti-neurofilament antibody and visualized after clearing in benzyl-benzoate/benzyl-alcohol ( BB-BA ) ( Fig 2 ) . Embryos were cut in half and flat-mounted for imaging . Motor axons innervating the CM muscle can be recognized on the embryo flank by their horizontal progression , as they intersect the vertically oriented thoracic nerves . After initial imaging of flat-mounted embryos halves ( Fig 2A and 2B , top and middle images ) , all inner structures , including thoracic nerves , were manually removed to better distinguish CM axons ( Fig 2A and 2B; bottom pictures ) . CM motor axons cover an area with a shape very similar to that occupied by GdnfLacZ-expressing cells . This shape was affected by Fat1 loss-of-function in a similar way as was the GdnfLacZ-expressing region . At comparable stages , the area covered by CM axons was smaller in Fat1-/- embryos , which exhibited shorter CM axons than wild types ( Fig 2A and 2B ) . As seen with muscle markers , the dorsal part of the muscle appears less affected . In the ventral muscle , mutant motor axons were shorter , with an apparent lower density of axon bundles than controls ( inserts in Fig 2A , bottom images ) . Throughout the period considered , the subcutaneous expansion of CM-innervating axons is fast and dynamic . It is therefore best represented by showing two consecutive stages , and quantified by measuring the area covered by CM axons , plotted relative to a reference structure ( such as the length of the 10th thoracic nerve [T10] , black dotted line ) used as an indicator of developmental age ( Fig 2C ) . In wild-type embryos , the area covered by CM axons expanded steadily , covering the embryo flank in little more than half a day . In contrast , the rate of progression of CM innervation is reduced by approximately 2-fold in Fat1-/- embryos compared to controls . Similar observations can be made with the other null allele ( Fat1LacZ , S3 Fig ) . In both mutants , in contrast to the abnormal behavior of CM-innervating axons , most other limb-innervating nerves are preserved and appear unaffected in Fat1-/- embryos ( Fig 2 ) . There was no obvious ectopic nerve corresponding to the supernumerary muscle in the scapulohumeral region . Thus , loss of Fat1 functions appears to predominantly affect the development of axons innervating the CM , matching the pronounced effect on muscle spreading and differentiation . A number of important features emerge when carefully comparing MLC3F-2E+ embryos and GdnfLacZ embryos during the developmental progression of the CM muscle in wild-type embryos . For clarity in the following description , it is necessary to recall a few notions of orientation . Because the muscle progresses from anterior to posterior ( see Figs 1 and 3A ) , the front of muscle progression is located posteriorly , whereas the rear corresponds to the point of origin of the muscle , at the brachial plexus , located anteriorly . When comparing MLC3F-2E+ embryos and GdnfLacZ embryos at similar stages ( compare Fig 1A with 1D and 1B with 1E ) , the area occupied by GdnfLacZ cells ( marked with a white dotted line ) appears larger than the area occupied by MLC3F-2E+ muscle fibers ( marked with the red line ) . This MLC3F-2E+ CM differentiation has a specific fan-like shape , in which multinucleated muscle fibers extend from a narrow zone at the anterior origin to a posterior side distributed along a wider dorsoventral extent ( Fig 1A and 1B ) . Because progression occurs at this posterior front ( called front of fiber elongation ) , this indicates that multinucleated fibers elongate by adding new nuclei at the posterior side . This posterior front of fiber elongation appears to be situated approximately in the middle of the muscle , at a regular distance from the even more posterior front of progression of the GdnfLacZ+ area , likely composed of migrating muscle progenitors ( Fig 3A ) . Nevertheless , when carefully observing X-gal–stained MLC3F-2E+ embryos , one can also distinguish , beyond the front of multinucleated fiber elongation , some LacZ+ nuclei expressing β-galactosidase at lower levels , within an area matching in size and shape the GdnfLacZ+ area ( white dotted line , Fig 1 ) . This suggests that the posterior half of the CM muscle ( between the two fronts ) is essentially occupied by GdnfLacZ-positive migrating myogenic progenitors and a few scattered mononucleated MLC3F-2E+ myocytes . In contrast , the anterior half of the muscle contains elongating MLC3F-2E+ fibers and GdnfLacZ-positive cells . This was confirmed by analyzing serial sections of GdnfLacZ/+ embryos by IHC following β-galactosidase , the sarcomeric protein Myh1 in muscle fibers , and the muscle progenitor marker Pax7 ( Fig 3B and 3C ) . Throughout the extent of the CM muscle ( not considering the plexus mesenchyme region ) , we found that GdnfLacZ was co-expressed with Pax7 , confirming that it labels progenitors ( Fig 3B ) , whereas it is not co-expressed with the sarcomeric protein Myh1 ( Fig 3C inserts ) . Thus , GdnfLacZ can be used as a specific marker of CM ( and LD ) progenitors . When following the CM muscle on serial sections ( Fig 3C ) , we confirmed that only the anterior and middle sections contained a mixture of Pax7+/GdnfLacZ+ progenitors and Myh1+ fibers , whereas in the posterior sections , the CM only contained Pax7+/GdnfLacZ+ progenitors , but no fibers . Thus , there is a physical separation between the front of progenitor migration and the front of differentiation , where new differentiating myocytes are being added to growing muscle fibers on their posterior side . Interestingly , the shape of the area covered by CM-innervating motor axons ( Fig 2 ) also appears more similar to the shape covered by GdnfLacZ-expressing progenitors than to the shape of the area covered by differentiated fibers ( as visualized with the MLC3F-2E transgene ) ( Fig 1 ) . The population of spinal MNs innervating the Gdnf-producing muscles CM ( Fig 3A ) and LD is characterized by expression of the transcription factor Etv4 [39 , 41] . We therefore took advantage of an Etv4-GFP transgene ( S1 Table , [60] ) , in which GFP expression reproduces that of Etv4 and enables detection of the corresponding axons , to follow CM-innervating motor axons on serial sections of GdnfLacZ; Etv4-GFP+ embryos ( Fig 3A and 3C ) . GFP-positive axons can be seen throughout the extent of the CM , initially running as large bundles located along the interior side of the muscle in anterior sections and progressively detected as smaller bundles , posteriorly ( Fig 3C ) . Even in the posteriormost sections , in which Myh1+ fibers are no longer detected in the GdnfLacZ-positive CM sheet , GFP-positive axon bundles are found intermingled with GdnfLacZ progenitors . Thus , Etv4-GFP+ motor axons cover the entire zone enriched in GdnfLacZ-expressing progenitors , as previously observed [39] . These observations imply that the front of migration contains GdnfLacZ progenitors and distal tips of Etv4-GFP+ motor axons , which appear to progress hand in hand , but not differentiated fibers . Thus , there is a significant topographical separation between the front of progenitor migration and axonal elongation , and the front of progression of muscle fiber elongation , where new myocytes are added to growing muscle fibers on their posterior side . These observations raise the interesting possibility that the speed and direction of CM muscle fiber elongation could be influenced by the direction/speed of progenitor migration or axon elongation , or by a combination of both . In addition , the simultaneous progression of CM progenitors and Etv4-GFP+ motor axons raises the issue of determining whether axons or muscle progenitors influence more the progression of CM expansion and the subsequent expansion of muscle fibers . We next asked in which cell type Fat1 is required to exert the function ( s ) underlying this complex event of CM muscle and nerve morphogenesis . Although we previously showed that Fat1 is required in the myogenic lineage to modulate myoblast migration polarity , the consequences of Fat1 ablation driven in trunk myoblasts by Pax3cre were milder than in constitutive knockouts [14] , suggesting that Fat1 might be required in other cellular components of the circuit for its muscle-patterning function . We therefore first analyzed Fat1 expression during CM development , focusing on all the cell types involved , including muscles , surrounding CTs , and MNs . Fat1 expression was followed either with anti-β-galactosidase antibodies on serial sections of Fat1LacZ/+ embryos ( Fig 4A and 4C , S3 and S4 Figs ) , by anti-Fat1 IHC on sections of GdnfLacZ/+ embryos ( Fig 4B and 4D , in which the β-galactosidase pattern reproduces Gdnf expression ) , or by in situ hybridization with a Fat1 RNA probe ( Fig 5 ) . As previously reported [14] , in addition to its expression in migrating myogenic cells ( with Fat1LacZ expression detected in both Pax7-positive progenitors and Myh1-positive muscle fibers , Fig 4C ) , we also found that Fat1 was expressed in mesenchymal cells surrounding the CM muscle , thus constituting a sheet of Fat1-expressing cells , through which the CM muscle expands ( Fig 4C and 4D; see also [14] ) . Similarly , Fat1 protein is detected not only in GdnfLacZ-positive muscle precursor cells ( Fig 4D ) but also in cells surrounding the layer of GdnfLacZ-expressing progenitors ( Fig 4D ) . β-galactosidase staining intensity in Fat1LacZ/+ embryos appears higher in the subcutaneous mesenchymal layer at the level of posterior CM sections than it is at anterior levels ( Figs 4C and S3B ) . This expression was preserved in genetic contexts leading to depletion of migratory muscles , as evidenced by the robust Fat1LacZ expression detected in embryos in which myoblast migration is abrogated , such as in mutants of the HGF receptor gene Met [61 , 62] or of Pax3 [63–65] ( S5A and S5B Fig , S1 Table ) . In the latter case , following cells derived from the Pax3cre lineage using the R26-YFP reporter ( in which expression of the yellow fluorescent protein , driven by the ubiquitous Rosa26 locus , is conditioned by cre-mediated removal of a stop cassette , S1 Table , [66] ) ( S5B Fig ) reveals that most of the Fat1-β-galactosidase fusion protein detected in the dorsal forelimb remains unchanged , even in absence of the YFP+ myogenic component in Pax3cre/cre; Fat1LacZ/+ compared to Pax3cre/+; Fat1LacZ/+ embryos ( the remaining YFP+ cells correspond to Schwann cells along the nerves ) . Together , these findings indicate that a large part of Fat1 expression surrounding or within the CM muscle corresponds to mesenchymal cells ( or CT ) . The population of spinal MNs innervating the Gdnf-producing muscles CM and LD is characterized by expression of the transcription factor Etv4 ( Fig 5A ) [39 , 41] . Interestingly , in addition to its peripheral expression , we also detected Fat1 expression in groups of MNs at brachial levels encompassing the pools of Etv4-expressing MNs ( Fig 5C–5E ) . Aside from expression in neural precursors in the ventricular zone all along the dorsoventral axis ( Fig 5B , S6C Fig ) , this brachial MN column was the main site of high Fat1 expression in the spinal cord ( Fig 5C ) . Fat1 expression was also detected in a column of ventral neurons at thoracic levels ( S6A Fig ) and , with later onset , in subsets of lumbar and sacral MNs ( Fig 5C [orange arrowheads] and S4B Fig ) . The overlap between Etv4 and Fat1 expression in brachial MNs was maximal at E11 . 5 ( Fig 5D ) , whereas additional groups of Fat1-positive; Etv4-negative ( Fat1-only ) neurons become detectable in dorsal positions at E12 . 5 ( Fig 5D and 5E ) . At that stage , the CM MN pools have completed their shift in cell body position in the spinal cord [41] , resulting in the dorsoventral split of Fat1 expression domain at C7–C8 levels into a dorsal Fat1-only pool ( Fig 5D and 5E , asterisk ) and a ventral Fat1+/Etv4+ pool ( Fig 5D and 5E , arrowheads ) , matching position of CM MNs . Given the Fat1/Etv4 co-expression , we asked whether Fat1 could be a transcriptional target of Etv4 or whether its expression was dependent on factors acting upstream of Etv4 , such as GDNF and HGF [39 , 41 , 47] . Fat1 expression only appeared modestly reduced in shape in Etv4-/- and Gdnf-/- spinal cords and unchanged in Metd/d spinal cords at E11 . 5 ( S6A and S6B Fig ) . These data indicate that Fat1 induction occurred independently of HGF/Met , GDNF , and Etv4 , in spite of the subtle changes in shape of Fat1-expressing columns in Etv4 and Gdnf mutants . Nevertheless , at E12 . 5 , after the dorsoventral split into two Fat1-expressing columns , the ventral Fat1-expressing pool was missing in the Etv4-/- , Gdnf-/- , and Metd/d spinal cord ( S6A and S6B Fig ) , consistent with previously reported changes in the fate of CM MNs [39 , 41 , 47] . In contrast , the dorsal column appeared increased in the Etv4-/- and Gdnf-/- spinal cord ( S6B Fig ) , also consistent with the altered positioning of some CM neurons [39 , 41] . This dorsal column appeared reduced in Metd/d spinal cord , possibly resulting from the onset of enhanced MN death in the absence of the target muscle [46 , 47 , 60] . Altogether , these data are consistent with Fat1 being expressed in CM motor pools . The reiterated use of Fat1 expression in several components of the GDNF/Etv4 circuit and the altered shape of CM muscle and innervation pattern in Fat1 mutants raise the possibility that loss of Fat1 functions might influence development of this neuromuscular circuit , either by acting directly in MNs or as an indirect consequence of its role in muscle patterning . We next asked whether the CM muscle phenotype and the changes in motor axon patterns observed in Fat1 mutants were also associated with molecular defects in the corresponding spinal MN pools . The co-expression of Fat1 with Etv4 in MNs and the selective alteration in shape and nerve projections to the CM prompted us to focus on specification of the CM motor pools and to examine whether Etv4 expression was altered in Fat1 mutants . This analysis revealed that Fat1 is dispensable for the establishment of Etv4 expression domain in Fat1-/- spinal cord ( Fig 6A–6C and S7A Fig ) . Using the Etv4-GFP transgene ( S1 Table , [60] ) also allowed detecting a near normal appearance of the GFP-positive motor columns in Fat1-/- spinal cords ( Fig 6D ) . Nevertheless , analysis of signal intensity of Etv4 mRNA detected a modest but significant lowering of Etv4 signal intensity of around 20% ( Fig 6B and 6C ) . We next asked whether such modest changes in Etv4 levels may be sufficient to impact expression of some of its transcriptional targets ( for example , affecting low-affinity but not high-affinity targets ) . We therefore analyzed expression of several Etv4 target genes expressed in subsets of Etv4+ neurons , some of these markers being also deregulated by loss of Met , a situation that we showed partially alters Etv4 expression [41 , 47] . We first studied Sema3E and Cadherin 8 , two known Etv4 targets in the CM motor pool [41] . Consistent with previous reports [41] , their expression is absent in Etv4-/- spinal cords , whereas it was reduced but not lost in Metd/d spinal cords ( S7E Fig ) . Sema3E and Cadherin 8 expression appeared unaffected in Fat1-/- spinal cords ( S7E Fig ) , with these two genes behaving as expected for robust “high-affinity” Etv4 targets . We next studied expression of Clusterin and runt related transcription factor 1 ( Runx1 ) , two genes that we selected for their expression in the CM motor pool . Runx1 is a transcription factor expressed in a rostral column of ventrally located neurons spanning C1–C6 [67] and in a separate pool at C7–C8 , matching the CM subset of Etv4+ MNs ( Fig 6E and S5C Fig ) , where its expression was shown to require Met signaling [60] . Clusterin is a glycoprotein known to accumulate in neurons after axotomy , injury , or in neurodegenerative diseases ( such as amyotrophic lateral sclerosis [ALS] , Alzheimer ) that was proposed to modulate cell death , autophagy , and clearance of protein aggregates or cell debris [68–71] . Clusterin expression in the developing spinal cord was restricted to a subset of MNs matching the position of the Etv4+ CM pool ( Fig 6H ) . In contrast to Sema3E and Cadherin 8 , expression of both Clusterin and Runx1 was completely abolished in Etv4 and Met knockout spinal cords ( Fig 6E–6J ) . Clusterin and Runx1 signal intensities were severely reduced in the C7–C8 MN pool in Fat1-/- spinal cords , although not as severely as in Met and Etv4 mutants ( Fig 6E–6J ) . The severe effect on Clusterin and Runx1 expression and mild effect on Etv4 expression detected at E12 . 5 in Fat1 mutant spinal cords is unlikely to result from increased cell death . First , Runx1 is co-expressed with Sema3E ( S6C Fig ) , the expression of which is unaffected by loss of Fat1 ( S7E Fig ) , indicating that the neurons that fail to express Clusterin/Runx1 in Fat1 mutants are still present and retain Sema3E expression . Second , naturally ongoing MN death only peaks at E13 . 5 in mice , one day after the stage analyzed here . Absence of muscle development in mutants of genes such as Pax3 or Met was shown to have a minor effect on MN numbers prior to the establishment of the trophic dependency of MNs for muscle-derived factors [47] . Thus , at E12 . 5 , the stage of our current analysis , cell death is unlikely to contribute to the effects of Fat1 loss-of-function on Clusterin/Runx1 expression . Thus , their sensitivity to the absence of Fat1 is consistent with Clusterin and Runx1 expression being low-affinity targets of Etv4 , affected by subtle changes in Etv4 levels . Altogether , these data confirm that loss of Fat1 compromises the correct acquisition of the molecular identity of the CM-innervating MN pool , in addition to its effect on CM muscle growth and differentiation . There is a strong topological connection between altered CM muscle morphology observed in Fat1 mutants and the selective defects in the corresponding MN population . This topology is even more puzzling when considering that Fat1 is expressed in several cell types , including Etv4+ MN , along the GDNF/Etv4 circuit . Neural and muscular aspects of the phenotypes could represent the consequences of one single primary phenotype resulting from Fat1 deletion in one single cell type . Given the known dependency of Etv4 induction on GDNF [39 , 41 , 47] , the neural phenotypes could arise as a consequence of the reduced amount of GDNF-producing cells . However , the discovery of Fat1 selective expression in Etv4+ MNs raises the alternative possibility that some aspects of these phenotypes may result from its ablation in MNs . Thus , the overall knockout phenotype could represent the cumulated consequences of phenotypes resulting from ablation of independent Fat1 assignments in distinct cell types . To tackle this question , we used a conditional approach to selectively interfere with Fat1 functions in a tissue-specific manner . This allowed assessing in which cell type it was necessary and sufficient to ablate Fat1 functions to reproduce each aspect of the muscular and neural phenotypes observed in constitutive mutants . We combined our Fat1Flox conditional allele [14] with several Cre lines matching the different sites of Fat1 expression ( see S1 Table for full descriptions of each mouse line ) : Pax3-cre knock-in for ablation in the premigratory myogenic lineage [64] , Olig2-cre knock-in for MN precursors [72] , Prx1-cre transgenic line for limb mesenchymal cells [73 , 74] , and Wnt1-cre transgenic line in neural crest derivatives [75] . We first focused on the strong alterations of CM muscle appearance observed in the constitutive knockout . We followed muscle differentiation using the MLC3F-nLacZ-2E transgene . Although we previously described significant alterations of myogenesis resulting from Fat1 ablation in the premigratory myogenic lineage ( driven by Pax3-cre [14] ) , the appearance of the CM muscle in E14 . 5 Pax3-cre; Fat1Flox/Flox embryos was grossly normal , albeit with reduced density of differentiated muscle cells . This contrasts with constitutive Fat1 knockouts , in which the severe effect on CM shape persists at later stages , when the muscle has further extended to cover most of the trunk [14] . This discrepancy indicated that the effect on muscle growth caused by premigratory ablation in myogenic cells was attenuated at later stages . Thus , Fat1 function underlying proper CM development must therefore also be exerted in a distinct cell type . Given the selective co-expression of Fat1 and Etv4 in the CM MN pool and the fact that the front of CM expansion is led by both muscle progenitors and axons , we next asked if Fat1 deletion in the corresponding MNs had an impact on CM muscle development . However , MN-specific Fat1 deletion in Olig2-cre; Fat1Flox/Flox; MLC3F-2E embryos did not cause any detectable change in the appearance ( S8A Fig ) and expansion rate of the CM , as assessed by measuring the CM area in X-gal–stained embryos carrying the MLC3F-2E-LacZ transgene ( S8B Fig ) . MN-specific mutants also lacked all the other muscle phenotypes observed in constitutive knockouts , with no ectopic muscle in the scapulohumeral region ( S8A Fig ) , no myocyte dispersed in the forelimb ( S8A and S8B Fig ) , and an overall normal appearance at E14 . 5 ( S9 Fig ) . Together , these observations establish that the progression of myogenic differentiation in the CM is not significantly influenced by Fat1 activity in the pool of MNs innervating this muscle . In contrast , we found that Fat1 ablation driven in the limb and trunk mesenchyme by Prx1-cre led to a severe and robust change in the appearance of the CM muscle ( Fig 7A and 7B ) and of upper forearm muscles ( S8A and S8B Fig ) . This phenotype was already visible at E12 . 5 during the phase of CM posterior extension , with a significant reduction of the rate of CM expansion assessed by X-gal staining in Prx1-cre; Fat1Flox/Flox; MLC3F-2E embryos ( Fig 7A and 7B ) . This leads to a phenotype as severe as that observed in constitutive knockouts , as measured when comparing the growth rate of the CM differentiation area to that of BW muscles ( Fig 7B ) . Again , the ventral half of the CM was severely shortened , whereas spreading in the dorsal part appeared less affected ( Fig 7A ) . Moreover , Prx1-cre; Fat1Flox/Flox; MLC3F-2E embryos consistently exhibited changes in forearm muscles , such as the appearance of ectopic muscles in the scapulohumeral area ( S8A Fig ) or a mild but significant increase in the number of dispersed myocytes in the forelimb ( S8B Fig ) . At E14 . 5 , the stage at which the CM is fully extended in the trunk , up to the point of emergence of the hind limbs , Prx1-cre; Fat1Flox/Flox embryos exhibited a pronounced hypoplasia of the ventral cutaneous maximus ( vCM ) , with only very few MLC3F-2E-positive fibers having successfully extended in random orientations ( S9A and S9C Fig ) , thus strongly recapitulating the changes seen in the germ line deletion . Again , the dorsal cutaneous maximus ( dCM ) appeared less affected in E14 . 5 Prx1-cre; Fat1Flox/Flox; MLC3F-2E embryos , although the density of LacZ+ nuclei nevertheless appeared reduced compared to Fat1Flox/Flox; MLC3F-2E embryos ( S9A Fig ) . Constitutive Fat1 knockouts also exhibit abnormal morphology of subcutaneous muscles in the face , mostly visible at E14 . 5 [14] . Unlike the CM muscle , this group of facial subcutaneous muscles appeared unaffected in E14 . 5 Prx1-cre; Fat1Flox/Flox; MLC3F-2E embryos ( S9 Fig ) , consistent with the lack of Prx1-cre activity in craniofacial mesenchyme [73 , 74] . Ablation of Fat1 functions in the craniofacial mesenchyme was achieved using Wnt1-cre ( S1 Table , [75] ) , which drives cre expression in all neural crest cells , including the cephalic neural crest , from which most craniofacial mesenchyme derives [76] . This led to profound morphological alterations in the appearance of facial subcutaneous muscles in Wnt1-cre; Fat1Flox/Flox; MLC3F-2E embryos , with a severely reduced fiber density and drastic changes in fiber orientation and position of fiber origins ( S9 Fig ) . The same severe effect on morphology of facial subcutaneous muscles was also observed in E14 . 5 Pax3cre/+; Fat1Flox/Flox embryos ( S9 Fig ) . This reflects the known activity of Pax3-cre in the neural crest as well . By contrast , the observation of intact morphology of scapulohumeral muscles , normal rate of progression of the CM muscle , and lack of myocyte dispersion in the forelimb in Wnt1-cre; Fat1Flox/Flox; MLC3F-2E embryos ( S8 and S9 Figs ) indicates that Fat1 activity in the trunk neural crest is dispensable for its influence on trunk muscle development . Altogether , these data identify mesenchyme as a cell type in which Fat1 signaling is required for muscle morphogenesis , with the trunk mesenchyme deriving from Prx1-cre lineage and the craniofacial mesenchyme deriving from the Wnt1-cre/neural crest lineage . In contrast , these results establish that Fat1 activity in MNs and the trunk neural crest is dispensable for myogenic differentiation . So far , we have established that Fat1 is required in the mesenchyme for the progression of differentiation and fiber elongation in the CM . Because the delay in CM expansion observed in constitutive knockouts was associated with a reduced rate of expansion of the sheet of GdnfLacZ-expressing progenitors , we next asked if this progenitor progression was also affected , by bringing the GdnfLacZ allele in the mesenchyme-specific mutant context and performing Salmon-Gal staining on whole-mount embryos . As in knockouts , we observed an important reduction in the rate of progression of the area occupied by GdnfLacZ-expressing progenitors ( related to the evolution of trunk length ) in Prx1-cre; Fat1Flox/Flox; GdnfLacZ/+ embryos compared to Fat1Flox/Flox; GdnfLacZ/+ embryos ( Fig 7C and 7D ) . When comparing embryos of similar stage , the GdnfLacZ sheet appeared truncated in the vCM and shorter in the dCM , with an apparent reduction in staining density , visible by comparing staining intensity along a dorsoventral line positioned similarly ( blue dotted line in Fig 7C ) . This approach does not distinguish a reduction in expression level from a reduced number/density of cells expressing GdnfLacZ . Therefore , to discriminate between the two options , we next analyzed the level of β-galactosidase protein ( visualized by IHC ) driven by the GdnfLacZ allele on control and mutant embryo sections . At middle and posterior CM levels , there was a clear reduction in thickness of the sheet of GdnfLacZ+ cells detected in a Prx1-cre; Fat1Flox/Flox; GdnfLacZ/+ embryo compared to a control embryo , resulting from a lowered number of stained cells rather than a reduced expression level per cells ( Fig 7E , inserts ) . GdnfLacZ-expressing cells , however , exhibited comparable β-galactosidase staining intensity , ruling out an effect on Gdnf expression levels . Staining with antibodies against Pax7 to mark progenitors confirmed a reduction in the number of Pax7+ progenitors at comparable anteroposterior ( AP ) levels ( Fig 7E ) . Thus , the reduced CM thickness in Prx1-cre; Fat1Flox/Flox; GdnfLacZ/+ embryos results in large part from a reduced amount of Pax7+ progenitors , and from a consequent reduction in the amount of differentiated fibers ( highlighted with anti-Myh1 antibodies ) , compared to Prx1cre; Fat1Flox/+; GdnfLacZ/+ controls . Our histological analysis of serial sections of mesenchyme-specific mutants and controls was done in a genetic context , allowing lineage tracing of Prx1-cre activity with an R26-YFP reporter ( S1 Table , [66] ) to highlight cells in which cre-mediated recombination is occurring ( Fig 8A ) . This context ( Prx1-cre; R26YFP/+; GdnfLacZ/+ ) allowed visualizing both reporters ( β-galactosidase and YFP ) simultaneously by IHC on transverse sections spanning from the brachial plexus to the CM muscle , comparing Fat1Flox/+ ( Fig 8B ) with Fat1Flox/Flox ( Fig 8C ) mutant settings ( see also Fig 7E ) . This analysis confirmed that the Gdnf expression domain is subdivided into two compartments: in posterior sections , GdnfLacZ-expressing CM muscle progenitors detected throughout the length of the muscle do not derive from mesenchymal progenitors ( Figs 7E and 8 ) . These myogenic cells appear to slide along and be surrounded by a territory of YFP-expressing CT mesenchymal lineage composed of Prx1-cre; R26-YFP+ cells , up to a dorsoventral boundary corresponding to the limits of the Prx1-cre lineage [73] ( Figs 7E and 8 ) . In contrast , in anterior sections , at the level of the brachial plexus , GdnfLacZ-expressing cells co-express β-galactosidase and YFP , indicating that the plexus component of GdnfLacZ domain is constituted of mesenchyme-derived cells ( Fig 8 ) . Thus , Gdnf expression domain is constituted of two subdomains of distinct developmental origins ( myogenic and mesenchymal , respectively ) , which are anatomically connected at the position of origin of migration of the CM ( and LD ) progenitors . Among the two subdomains of GdnfLacZ expression , only the myogenic component was affected in Prx1-cre; Fat1Flox/Flox; GdnfLacZ/+; R26YFP/+ embryos ( Fig 8C ) , whereas the mesenchymal subdomain of GdnfLacZ expression appeared unaffected , both anatomically and in β-galactosidase intensity . In this mutant context , Fat1 activity is disrupted in the YFP-positive cells , and not in the GdnfLacZ/Pax7+-expressing progenitors . Thus , mesenchymal Fat1 depletion has no effect on mesenchymal Gdnf expression at plexus level , in contrast to the constitutive knockout ( S2 Fig ) . This indicates that this part of the knockout phenotype did not result from depletion of mesenchymal Fat1 and most likely reflects Fat1 activity in another cell type . In contrast , Fat1 ablation in the Prx1-cre lineage has a drastic impact on the myogenic component of the GdnfLacZ domain not derived from the Prx1-cre lineage . This demonstrates that Fat1 acts in a non–cell-autonomous manner . It is required in YFP-positive cells of the mesenchymal lineage to promote expansion of the sheet of migrating GdnfLacZ/Pax7+ progenitors , possibly by modulating the production by mesenchymal cells of signals controlling progenitor pool expansion and/or migration . Interestingly , at E12 . 5 , the CM and LD muscles are almost entirely surrounded by Prx1-cre-derived mesenchymal cells , with the exception of the dorsalmost tip of the CM , which lies beyond the dorsoventral limit of lateral-plate mesoderm–derived mesenchyme [73] ( Figs 7E and 8 ) . Interestingly , dorsal to this dorsoventral limit , thickness of the CM appears reinforced , suggesting that once the myoblasts reach the non-recombined mesenchymal zone , the unaltered Fat1 activity available in this dorsal environment allows them to resume their normal growth behavior , providing a possible explanation for the apparent sparing of dCM at E14 . 5 ( Fig 5C ) . Altogether , these data support a model in which muscle-associated mesenchymal cells exert a Fat1-dependent positive influence on CM muscle growth/extension . Thus far , we have identified the Prx1-cre lineage as the cell type in which Fat1 is required to control CM expansion and scapulohumeral muscle patterning . However , the Prx1-cre-derived lineage is broad ( Fig 8 ) and includes several distinct subtypes of CT . These comprise specialized CT , such as bones and cartilage , dense regular CT , such as tendons , and dense irregular CT ( here referred to as loose CT ) , such as muscle-resident mesenchymal derivatives or the perimuscular environment [13] . All of these subtypes express Fat1 at relatively high levels ( Fig 4 , S3 and S4 Figs ) . The CM grows towards a subcutaneous layer of CT , in which we observed increasing levels of Fat1 expression . Furthermore , CM extension appears to be affected on the side of its growth towards the skin interface . Altogether , this suggests that this subcutaneous interface with the CM muscle might be where this Fat1 function is taking place . Whole-mount in situ hybridization with the tenocyte/tendon marker Scleraxis highlights sites of intense expression in the limb tendons or at the interface between intercostal muscles and ribs ( visible by transparency , beneath the CM ) , but only shows background Scleraxis levels in the region in which CM is migrating ( S5A Fig ) . When analyzing the expression of other markers of CT subtypes by IHC on embryo sections , we found that this subcutaneous CT expresses high levels of platelet-derived growth factor receptor alpha ( Pdgfrα ) and Tenascin C but not transcription factor 7 like 2 , T cell specific 4 ( Tcf4/Tcf7L2 ) ( S10 Fig ) , which was otherwise detected at other muscle extremities and in subsets of limb muscle progenitors ( S10 Fig ) , as previously reported [77 , 78] . We next asked whether restricting Fat1 ablation to this subtype of mesenchyme could be sufficient to interfere with CM spreading and differentiation . Our characterization prompted us to consider inactivating Fat1 in the Pdgfrα-expressing lineage . However , as Pdgfrα is expressed at early stages in cells from which a large lineage will derive , we chose to use an inducible Pdgfrα-cre/ERT2 transgenic line ( S1 Table , here referred to as Pdgfrα-iCre ) , in which a CRE-ERT2 fusion protein ( iCRE , CRE fused with the estrogen receptor tamoxifen-binding domain ) is expressed in the Pdgfrα domain but remains catalytically inactive unless tamoxifen is provided [79] . This strategy is expected to highlight only a subset of mesenchymal cells expressing Pdgfrα at the time of tamoxifen administration , including the loose CT at the skin–muscle interface ( Fig 9A and 9B ) . To establish the conditions to obtain a reliable excision rate , we first combined the Pdgfrα-iCre line to the R26-YFP reporter [66] and exposed pregnant females to tamoxifen treatment . Optimal excision efficacy was obtained when injecting a first dose of 50 mg/kg at E9 . 75 ( earlier injections frequently lead to developmental arrest ) , followed by a second injection of 100 mg/kg at E10 . 5 ( Fig 9A and 9B ) . This allowed consistent detection of the R26-YFP signal when screening whole embryos just after dissection ( S10B Fig ) . Analysis of YFP expression by IHC on sections of Pdgfrα-iCre; R26YFP/+ embryos confirmed that recombined YFP-positive cells ( “Pdgfrα-iYFP” cells ) were mostly localized in the loose CT surrounding the CM , whereas none of the Myh1-positive fibers exhibited any detectable recombination ( Fig 9A ) . Although our experimental conditions do not allow simultaneous detection of YFP and Pax7 ( the heat-induced-epitope-recovery treatment required to detect Pax7 protein abrogates detection of YFP ) , analysis of neighboring sections was consistent with the Pax7-containing area not exhibiting any YFP activity ( Fig 9A ) . Thus , these experimental conditions allow an approximate excision rate of 30% in the loose subcutaneous CT surrounding the CM , whereas no activity in the myogenic lineage was detected . We next asked whether Fat1 ablation in around 30% of the loose CT surrounding the CM was sufficient to interfere with its expansion and with progression of differentiation . The Pdgfrα-iCre line was combined with the Fat1Flox allele and with either the MLC3F-2E transgenic line , to follow muscle differentiation , or with GdnfLacZ , to follow the progression of progenitor migration ( Fig 9C and 9E ) . Pregnant females were treated with tamoxifen as defined above and embryos collected at E12 . 5 . In all cases , mutants were compared to control embryos from tamoxifen-treated litters . Analysis was performed as previously , by measuring the area occupied by the MLC3F-2E+ fibers in the CM as compared to the area occupied by BW muscles , or by assessing the area occupied by the GdnfLacZ-expressing progenitors as compared to the head area and/or trunk length . Fat1 ablation driven by such restricted recombination paradigm leads to a significant delay in the progression of both CM muscle fiber elongation ( Fig 9C and 9D ) and CM progenitor migration ( Fig 9E and 9F ) . The median differentiated CM area measured in tamoxifen-treated Pdgfrα-iCre; Fat1Flox/Flox; MLC3F-2E embryos ( ratio of CM differentiated area/BW area , normalized to median control ratio ) was reduced by approximately 30% compared to tamoxifen-treated control embryos ( Fig 9C and 9D ) . Similarly , we observed a 30% reduction of the median area covered by CM progenitors observed in tamoxifen-treated Pdgfrα-iCre; Fat1Flox/Flox; GdnfLacZ/+ embryos ( Fig 9E and 9F ) . In contrast , there was no apparent effect on the shape of scapulohumeral muscles , and we did not detect any significant enhancement of myocyte dispersion in the forelimb ( S10C and S10D Fig ) . In the conditions of tamoxifen treatment we used for the Pdgfrα-iCre line , the percentage of cells in which the reporter R26-YFP expression indicates cre-mediated recombination ( Fig 9A ) is much more restricted compared to the Prx1-cre lineage ( Figs 7E and 8 ) . This indicates that Fat1 activity is required in a significant proportion of cells among these recombined “Pdgfrα-iYFP” cells for CM muscle spreading . Overall , these data are consistent with Fat1 being required in the loose CT for its non–cell-autonomous influence on CM progenitor spreading and muscle fiber extension . In conclusion , we have identified the mesenchymal lineage as the place where Fat1 activity is required to promote CM muscle expansion . We have refined our knowledge on this lineage by uncovering that in large part , this function occurs in the Pdgfrα-dependent loose CT , such as the subcutaneous layer in which CM expansion occurs . Finally , we find that this lineage represents a subset of the Pdgfrα-iCre lineage and corresponds to the subset of cells expressing Pdgfrα between E9 . 5 and E10 . 5 . Having refined the knowledge on which CT type is required for the regulation of CM muscle expansion by Fat1 , we next asked whether this activity of Fat1 in the mesenchyme-derived CT was sufficient to explain alterations of CM innervation and changes in MN pool specification , or whether Fat1 is also additionally required in MNs . To determine which of the two sites of Fat1 expression was accountable for the altered pattern of axonal arborization of the target muscle CM seen in constitutive Fat1 knockouts , we therefore performed whole-mount anti-neurofilament IHC on embryos lacking Fat1 either in MNs ( Olig2cre/+; Fat1Flox/Flox ) or in mesenchyme ( Prx1-cre; Fat1Flox/Flox ) and examined the pattern of CM motor innervation . After imaging of BB-BA cleared embryo flanks ( Fig 10A , upper images ) , the thoracic cage and associated thoracic nerves were removed to allow easier visualization of CM-innervating axons ( Fig 10A , lower panels ) . We found that mesenchyme-specific mutants ( Prx1-cre; Fat1Flox/Flox embryos ) exhibited an overall shortening of the area covered by CM-innervating axons compared to stage-matched controls . This effect was very severe ventrally , with a complete absence of motor axon extending in the vCM ( Fig 10A ) , whereas axons extending in the dCM were present but shorter than in control embryos . To quantify this effect , the area covered by CM-innervating axons was plotted relative to a landmark indicating developmental age , using the length of the T9 nerve ( Fig 10B ) . The net effect quantified by following the ratio area of CM axons/T9 length was almost as severe as the effect measured in constitutive knockouts ( Fig 2 ) . This effect was strikingly reminiscent of the profound shortening and shape change of the area occupied by GdnfLacZ+ CM progenitors seen earlier ( Fig 7 ) . Thus , Fat1 ablation in the mesenchyme non–cell-autonomously influences not only expansion of the muscle progenitor area but also extension of motor axons projecting in this muscle . This indicates that even in this context of severe perturbation of CM expansion , there remains a strong coupling between motor axons and muscle progenitors , and the topography of muscle fiber extension . To determine whether growth of the CM-innervating axon was dependent on muscle , genetic depletion of the myogenic lineage was achieved through Myf5-cre-driven [80] conditional expression of diphtheria toxin A ( DTA ) regulated by the R26 locus ( R26Lox-LacZ-STOP-Lox-DTA locus [81] ) . This paradigm resulted in a severe loss of muscle mass , as assessed with the MLC3F-2E-LacZ reporter . In the trunk , X-gal staining shows the loss of myogenic cells corresponding to the CM muscle , to intercostal muscles , and to the trapezius muscle , whereas epaxial muscles , although irregular , were less efficiently deleted ( S11B Fig ) . As previously shown [46 , 82] , limb-innervating nerves were present and had extended in spite of the absence of muscles ( S11B Fig ) . In contrast , the thoracic intercostal nerves , the CM-innervating motor nerve , and the spinal accessory nerve ( or nerve XI ) , which innervates the trapezius muscle , were severely affected by muscle depletion ( S11A and S11C Fig ) compared to stage-matched control embryos . These results indicate that genetic suppression of myogenic cells was sufficient to cause a drastic shortening of CM-innervating axons , in contrast to the lack of effect on the trajectory of limb-innervating nerves ( S11 Fig ) . This supports the idea that the coupling between motor and muscle phenotypes observed in Prx1-cre; Fat1Flox/Flox embryos could be mediated by muscle-derived factors . Although the effect of MN-specific Fat1 ablation appeared less obvious , our method of quantification nevertheless uncovered a mild but significant reduction ( around 20% ) in the rate of expansion of the area covered by CM-innervating axons in Olig2cre/+; Fat1Flox/Flox embryos compared to age-matched controls ( Fig 10A and 10B ) . Overall , the shape of the innervated area does not change qualitatively , but axons are shorter in Olig2cre/+; Fat1Flox/Flox embryos than in age-matched controls . This indicates that in absence of Fat1 , CM motor axons extend at a slower rate . Even though the net effect was small compared to the phenotype of constitutive knockouts , this result supports the notion that Fat1 is required cell-autonomously for CM motor axon growth . Because CM motor axons progress hand in hand with Pax7+/GdnfLacZ+ progenitors , ahead of muscle fiber extension ( Fig 3 ) , we next asked if this mild but significant shortening of axons had any impact on the expansion of the CM progenitor area . We therefore brought the GdnfLacZ allele in the context of MN-specific Fat1 mutants and followed the area covered by GdnfLacZ progenitors ( Fig 10C and 10D ) . Unexpectedly , whereas the shape of the GdnfLacZ+ area appeared qualitatively unchanged , area measurements uncovered a significant lowering ( of about 25% ) of the ratio GdnfLacZ area/trunk length in Olig2-cre; Fat1Flox/Flox; GdnfLacZ/+ embryos compared to Fat1Flox/Flox; GdnfLacZ/+ controls . This result uncovers that Fat1 ablation in MNs non–cell-autonomously interferes with expansion of CM progenitors . This effect may either reflect a direct control by Fat1 in MNs of the process of muscle progenitor spreading , or a Fat1-independent consequence of the slower progression of motor axons . This finding is counterintuitive , as we had previously excluded that MN-specific Fat1 loss had any impact on muscle fiber extension ( S8 and S9 Figs ) . Thus , the shortening of the area covered by GdnfLacZ progenitors is not associated with a reduced progression of myogenic differentiation in the CM . This indicates that the slower migration seen in Olig2-cre; Fat1Flox/Flox; GdnfLacZ/+ embryos does not significantly influence the efficiency of muscle fiber extension . Thus , the alterations in the CM innervation pattern observed in constitutive Fat1 knockouts appear to result in part from removal of Fat1 axon growth-promoting activity in MNs , and to a large extent from the non–cell-autonomous consequences of Fat1 ablation in the mesenchyme , which simultaneously affects expansion of the CM progenitor area and axonal growth . Although these two non–cell-autonomous consequences of mesenchymal Fat1 ablation could represent two independent phenotypes , the fact that axonal extension of CM-innervating axons requires the presence of myogenic cells supports the possibility that the two phenotypes might be causally linked . Given its known axonal growth promoting activity , GDNF is a likely candidate to explain how impaired myogenic progenitor expansion could lead to inhibited axon elongation in Prx1-cre; Fat1Flox/Flox embryos . In contrast , the observation of a non–cell-autonomous impact of reduced axonal elongation on progenitor expansion in Olig2-cre; Fat1Flox/Flox; GdnfLacZ/+ embryos indicate that motor axons also play a role in promoting expansion of the CM progenitor domain . We next asked which tissue-specific Fat1 mutant was most closely reproducing the effects on expression of Runx1 and Clusterin in CM motor pools observed in constitutive knockouts ( comparing Prx1-cre and Olig2-cre ) . We found that expression of both genes in the CM motor pools was significantly reduced not only in mesenchyme-specific Prx1-cre; Fat1Flox/Flox mutants but also in MN-specific Fat1 mutants ( Olig2cre/+; Fat1Flox/Flox ) ( Fig 11A–11D ) . These observations demonstrate that Fat1 simultaneously exerts two complementary functions in MNs and in the mesenchyme , each of which cooperatively contributes to the acquisition of complete CM motor pool identity . Thus , the constitutive knockout phenotype results from the cumulated consequences of abrogating Fat1 functions in both MNs and mesenchyme . Interestingly , the reduction of Runx1/Clusterin expression in Fat1 conditionals is partially reminiscent of the phenotypes of Etv4 ( Fig 6E–6J ) and of Gdnf mutants ( Fig 11A–11D ) . Furthermore , the degree of reduction in Runx1 and Clusterin expression correlates with the degree of reduction of the area covered by Gdnf expressing myogenic progenitors observed in Prx1-cre; Fat1Flox/Flox and in Olig2cre/+; Fat1Flox/Flox embryos , respectively ( Figs 7 and 10 ) . This suggests that the lowering of Runx1 and Clusterin expression could be a consequence of lowering Gdnf levels and of the subsequent changes in Etv4 expression , even though subtle ( Fig 6A–6C ) . However , although comparable , the MN pool specification phenotype of either conditionals or constitutive Fat1 mutants is less severe than the effect of complete Gdnf elimination . In spite of the reduced expression of the myogenic component of Gdnf expression domain in Prx1-cre; Fat1Flox/Flox mutants , the overall GDNF level is maintained , owing to the remaining mesenchymal-Gdnf expression at plexus levels ( Fig 8 ) . This minimizes the impact on induction of Etv4 , the transcription factor acting upstream of Runx1 and Clusterin ( Fig 6 ) . As a result , Etv4 expression is less severely affected in Prx1-cre; Fat1Flox/Flox embryos than in embryos with only one functional copy of Gdnf ( compare Fat1Flox/Flox; GdnfLacZ/+ and Fat1Flox/Flox [Fig 11E and 11F] ) . As a consequence , the residual expression of Etv4 is sufficient to ensure correct mediolateral positioning of MNs in the spinal cord of Prx1-cre; Fat1Flox/Flox embryos ( S12A Fig ) , contrasting with what occurs in Gdnf-/- or Etv4-/- spinal cords ( S12A Fig , [39 , 41] ) . However , genetic lowering of Gdnf exacerbates the effect of mesenchyme-specific Fat1 ablation on Etv4 expression when both conditions are combined ( Fig 11E and 11F ) . Consequently , Etv4 expression is significantly lower in Prx1-cre; Fat1Flox/Flox; GdnfLacZ/+ embryos than in either Fat1Flox/Flox; GdnfLacZ/+ or Prx1cre/+; Fat1Flox/Flox embryos ( Fig 11E and 11F ) . The possibility that Gdnf could itself be involved in regulating the subcutaneous progression and number of CM progenitors ( thereby indirectly influencing the number of other factors secreted by CM progenitors ) is ruled out by the observation of normal expansion of the area covered by GdnfLacZ expressing CM progenitors in GdnfLacZ/LacZ embryos , compared to GdnfLacZ/+ ( S12B Fig ) . Altogether , these data identify Gdnf as an essential mediator of the effect of mesenchyme-specific Fat1 ablation on acquisition of CM motor pool fate , part of which involves the dose-dependent fine-tuning of Etv4 expression . Finally , we sought to evaluate the respective functional relevance of the two sites of Fat1 expression for neuromuscular junction ( NMJ ) integrity in the adult CM muscle in both MN-specific and mesenchyme-specific mutants . We focused on the dCM , where in spite of delayed myogenesis in mesenchyme-specific mutants , muscle fibers and motor axon have extended . In this region , muscle fibers , visualized by F-actin staining with fluorescent phalloidin , have formed in both mutants ( Fig 12A ) and have been innervated by motor axons , leading to a normal intramuscular pattern of NMJ distribution in both MN-specific and mesenchyme-specific Fat1 mutants . We analyzed synapse morphology in NMJ-enriched regions of the dCM by labeling the postsynaptic site with alpha-bungarotoxin ( α-BTX ) , which binds acetylcholine receptors ( AchRs ) and the nerve endings with anti-neurofilament antibodies . As a measurable feature of NMJ integrity , we analyzed the surface area of AchR-rich zones for an average of 23 synapses per mouse , comparing Olig2cre/+; Fat1Flox/Flox and Prx1cre/+; Fat1Flox/Flox mice with Fat1Flox/Flox mice , aged 9–13 months . This analysis uncovered that synapses are reduced in size in both genotypes compared to controls , this effect being more pronounced in Prx1cre/+; Fat1Flox/Flox mice than in Olig2cre/+; Fat1Flox/Flox , with a median synapse area of 420 μm2 ( 81% ) and 220 μm2 ( 42% ) , respectively , compared to 512 μm2 ( 100% ) in controls ( Fig 12B ) . A careful analysis of the distribution of synapse areas in each genotype confirmed a significant lowering in the percentage of large synapses and an increase in the percentage of smaller synapses , with up to 68% of synapses in Prx1cre/+; Fat1Flox/Flox mice , compared to 32% in Olig2cre/+; Fat1Flox/Flox and 12% only in Fat1Flox/Flox mice , being smaller than 300 μm2 ( Fig 12B ) . Analysis on the same sections/samples of the distribution of fiber width revealed no significant change in fiber diameter , arguing that the reduced size does not simply reflect adaptation to a change of the muscle fibers themselves ( Fig 12C ) . Instead , this smaller synapse area is indicative of altered NMJ integrity and was accompanied by spots of local denervation of the synapses or by fragmentation of the AchR-rich synaptic area ( Fig 12A ) . We next thought to determine if there were consequences in terms of the force-generating abilities of affected muscles . A typical assay is the grip test , which measures the grabbing force ( also called prehension force ) permitted by the coordinated activity of limb muscles ( forelimbs and/or hind limbs ) . Although phenotypes in the subcutaneous CM muscle are unlikely to contribute to forelimb movements , this assay allows detecting loss of strength in other limb muscles . The subtle defect in NMJ integrity observed in Olig2cre/+; Fat1Flox/Flox mice is expected to be restricted to target muscles of Fat1-expressing MNs . These include the CM muscle and a second pool of Etv4-negative neurons for which we did not identify the target muscle . Overall , this did not result in any significant loss of forelimb grip strength in Olig2cre/+; Fat1Flox/Flox mice compared to control mice ( Fig 12C ) . Nevertheless , Fat1 RNA , Fat1LacZ activity , and Fat1 protein were detected in subsets of lumbar MNs at E12 . 5 ( Fig 1E ) and E13 . 5 ( S3B Fig ) . We therefore asked if Fat1 ablation in these lumbar MNs had any impact in the force-generation capacity of hind limb muscles . A reliable way of approaching this force is to measure forelimb plus hind limb strength ( and inferring hind limb strength by subtracting forelimb strength ) . Consistently , we observed a significant lowering of forelimb plus hind limb grip strength in Olig2cre/+; Fat1Flox/Flox mice compared to Fat1Flox/Flox controls . This indicates that MN-Fat1 is required for the functional integrity of subsets of lumbar MN pools innervating selected hind limb muscles . In contrast , mesenchyme-specific Fat1 ablation in Prx1cre/+; Fat1Flox/Flox mice leads to significant lowering of both forelimb and the cumulated forelimb + hind limb grip strength , with an effect more pronounced than in Olig2cre/+; Fat1Flox/Flox mice . This is consistent with the possibility that these mesenchyme-specific mice might exhibit myogenesis defects in muscles other than the CM and LD , including muscles in the scapulohumeral and lumbar regions , such defects being likely to affect the amount of muscle-derived factors supporting NMJ maintenance ( Fig 7 ) . Altogether , these results show that Fat1 , by acting both in MNs and non–cell-autonomously in the mesenchyme , is required to shape CM and scapulohumeral neuromuscular circuits , thus ensuring their anatomical and functional integrity .
Ablating Fat1 in the mesenchyme profoundly disrupts the progression of progenitor migration and subsequently interferes not only with myogenic differentiation but also with axonal growth and complete specification of the cognate MN pool . This suggests that Fat1 controls the capacity of mesenchymal cells to influence ( 1 ) the subcutaneous migration of Gdnf+/Pax7+ CM progenitors , ( 2 ) the subsequent elongation of CM muscle fibers , ( 3 ) the parallel progression of CM-innervating axons of the Etv4-expressing MNs , and ( 4 ) final molecular specialization of this MN neuron pool . The fact that MN-specific Fat1 ablation only alters progenitor migration but not the rate of myogenic differentiation , whereas mesenchyme-specific ablation affects both , suggests that the alteration of fiber extension observed in Prx1-cre; Fat1Flox/Flox embryos is not simply the consequence of an impaired progression of migration but rather reflects a distinct independent function of Fat1 in the mesenchyme , which regulates fiber elongation . The mechanism by which Fat1 acts in the mesenchyme and whether this involves bidirectional Fat-Dachsous signaling remain to be clarified . An interesting analogous situation occurs during kidney development , in which a Fat4-Dachsous1/2 cross talk controls mouse kidney morphogenesis [30–32] . In this context , Fat4 acts in the kidney stroma ( analogous to the CT ) , whereas Dachsous1 and 2 are required in the Gdnf-expressing cap mesenchyme progenitors of renal tubules for their expansion ( analogous to progenitors of the CM muscle ) [30–32] . The mechanism identified in our study is also comparable to the feed-forward mechanism occurring in the drosophila wing imaginal disc , allowing wingless-dependent propagation of expression of the selector gene vestigial , to which mechanism Fat-Dachsous signaling contributes several aspects , thereby coupling cell fate and tissue growth [83] . Our data establish that Fat1 activity controls mesenchyme-derived cues acting on myogenesis . HGF and the chemokine cxcl12 are two ideal candidates for mediating such mesenchyme-derived activity . Both factors are known to induce myoblasts’ motility and promote their migration towards the limb bud [62 , 84 , 85] and to act simultaneously on subsets of MNs to modulate their specification and axon guidance [46 , 47 , 86] . We recently showed that mesenchyme-specific overexpression of the Met receptor tyrosine kinase , without interfering with HGF expression and secretion , nevertheless prevented the release of biologically active HGF from mesenchymal cells , thereby leading to the absence of migration in the limb [84] . This phenotype was , however , much more severe than the alterations of CM shape observed in mesenchyme-specific Fat1 mutants , making it unlikely for significant changes in the amount of biologically active HGF to account for the muscle-patterning defects observed in mesenchyme-specific Fat1 mutants . The ultimate identification of mesenchyme-derived factors , the production of which is altered by deficient Fat1 signaling , will necessitate unbiased approaches such as analyses of the secretome or transcriptome of wild-type and Fat1-deficient mesenchymal cells . In addition to seeking secreted factors , future lines of research will also evaluate whether the mesenchymal Fat1 activity modulates the mechanical properties of the CT or of the extracellular matrix it produces . Indeed , recent data showed that tissue stiffening on its own is a driver for the collective migration of neural crest cells [87] . Besides acting on myogenesis , we have established that Fat1 activity in mesenchymal cells influences motor axon growth and MN fate specification . These non–cell-autonomous consequences of mesenchymal Fat1 ablation on MNs and muscles could either represent independent phenotypes or could be causally linked to one another . Limb-innervating MNs are known to be prespecified to find their peripheral targets [10 , 11] and can correctly execute pathfinding decisions in the surrounding limb mesenchyme in a context in which muscles have been experimentally or genetically ablated [46 , 82] . In contrast , we provide experimental evidence supporting the idea that growth of MN axons innervating the CM , the trapezius , or BW muscles is severely impacted by muscle ablation and the subsequent depletion in muscle-derived factors ( S11 Fig ) . This is consistent with our previous work on the parallel activities of HGF/Met on muscle and MN development . Null mutation of the Met receptor abrogates muscle migration and the CM is part of the missing muscles [47 , 61] . This phenotype is associated with a complete absence of CM-innervating motor axonal arborization [60] . The null allele cannot distinguish whether this axonal defect is a consequence of the absence of CM muscle or of ablation of Met in MNs . Comparing changes in axonal patterns in null mutants and in neuronal-specific mutants indicated that Met is specifically required for motor innervation of another muscle ( the pectoralis minor ) but dispensable for the guidance of CM motor axons to and within the CM muscles [60] . Thus , the complete lack of axons matching a CM pattern in Met null mutant embryos is exclusively a consequence of the impaired migration of myogenic progenitors , also implying that CM motor axons along and within the CM are dependent on signals from myogenic cells . Together , these past and present elements support the possibility that altered extension of CM-innervating axons in mesenchyme-specific Fat1 mutants could likewise be a consequence of the impaired CM muscle progression . We show that in large part , MN phenotypes observed in mesenchyme-specific Fat1 mutants mimic the phenotypes resulting from depletion of one essential muscle-derived factor , GDNF , the production of which by CM myogenic progenitors must be quantitatively affected by the reduced subcutaneous spreading . While mesenchymal Gdnf expression by plexus cells is unaffected , Fat1 ablation in the mesenchyme drastically impairs the subcutaneous progression of the sheet of Gdnf+/Pax7+ progenitors ( Fig 8 ) . Our analysis of consecutive cross sections supports the idea of a reduced number of Gdnf+/Pax7+ progenitors , raising the possibility that this effect could result not only from impaired migration but also from impaired proliferation of the progenitor pool . As a result , MN specification defects exhibited by mesenchyme-specific Fat1 mutants partially mimic the effect of Gdnf loss-of-function . Depletion of myogenic-Gdnf is sufficient to impair Runx1 and Clusterin expression , with limited effect on Etv4 induction . Finally , the additive effect of genetic lowering of Gdnf levels and mesenchyme-specific Fat1 depletion indicates that leftover Gdnf is a rate-limiting factor accounting for the residual induction of Etv4 in CM motor pools in Fat1 mutants . These results do not exclude an alternative scenario , according to which other mesenchyme-derived factors regulated by Fat1 activity would directly act on CM-innervating MNs to control their fate and axonal growth . The fact that we did not detect quantitative changes in the mesenchymal component of GDNF expression argues against the possibility of cell-autonomous regulation of Gdnf promoter by mesenchymal Fat1 . As discussed earlier , the likelihood for the HGF amount to be drastically reduced is low , as this would have resulted in much more severe muscle defects . Thus , Fat1-dependent mesenchyme-derived factors influencing MN fate remain to be identified . A recent study described how regionalized mesenchyme-derived signals contribute to the acquisition of muscle-specific adaptation by proprioceptive afferents [88] . In this study , proprioceptive neuron specialization was neither affected by the MNs nor by muscle genetic ablation , pointing instead towards the contribution of mesenchymal cells . The identity of the regionalized mesenchyme-derived signals regulated by patterning activity of the Lmx1b transcription factor also remains unknown [88] . Which subtype of mesenchymal cells mediates such neuromuscular-patterning Fat1 activity ? The importance of lateral plate–derived CT for muscle patterning has been evidenced long ago through classical embryological studies [5] and is currently the subject of renewed interest [13] . The CT subtypes likely to influence muscle development include tendon progenitors at the interface between bones and muscles , but also a number of muscle-associated subtypes . Genetic data supporting a contribution of the non-myogenic CT to muscle development have previously been reported [13 , 77 , 89–91] . Mesenchyme-specific deletions of TCF4 , β-catenin , or Tbx4/5 lead to drastic alterations of muscle patterning reminiscent of the phenotypes of Fat1 mutants [77 , 91] . Another example of a transcription factor acting in the limb mesenchyme to pattern both muscle shapes and neuronal patterning is Shox2 [92] . These studies converge to identify CT subtypes by the combination of transcription factors they express . Among these transcription factors , Tcf4 expression highlights a subset of CT at muscle extremities distinct from tenocytes , in addition to subsets of myogenic progenitors . Tcf4-expressing CT cells were shown to influence muscle growth and patterning during development , but also adult skeletal muscle regeneration [77 , 78] . Cocultures of Tcf4-expressing CT cells with myogenic cells indicated that these cells are capable of producing factors promoting muscle growth and differentiation in vitro [77] . However , we find that the Fat1-expressing subcutaneous layer of CT that delineates the path of CM progenitor migration/extension does not express Tcf4 , but instead expresses Pdgfrα and TenascinC . Taking advantage of an inducible Pdgfrα-cre/ERT2 line , we identify the Pdgfrα lineage as critical for Fat1 to influence CM growth and differentiation , because a rate of ablation of 30% of the cells in this lineage is sufficient to significantly delay progression of both progenitor expansion and myogenic differentiation . The severity of this phenotype is lower when driven by Pdgfrα-iCre than in the context of the broad deletion mediated by Prx1-cre , supporting the idea of a dosage-dependent effect affecting the concentration of secreted factors . It remains possible that some of the muscle-patterning Fat1 activity might be executed in other CT subtypes , as Fat1 ablation in the Pdgfrα-dependent lineage did not reproduce the phenotypes observed in scapulohumeral muscles . It will be interesting in the future to define the muscle-associated CT subtype that most accurately overlaps with the Fat1 expression domain and defines its domain of activity . CT at the interface between the CM muscle and the skin must be distinct from classical tendons , which connect skeletal muscles to bones for force transmission . Candidate transcription factors that could highlight a signature for the Fat1-dependent CT subtype include Gata4 , Tbx3 , Osr1 , and Osr2 [93 , 94] , which were recently described as markers and essential players in defining distinct CT subtypes , respectively controlling morphogenesis of the diaphragm and of subsets of forelimb muscles [89 , 90 , 95] . The fact that Fat1 deficiency did not alter diaphragm development or lead to diaphragmatic hernias such as those occurring in the absence of Gata4 make it unlikely for Fat1 to be required in the Gata4+/diaphragm CT subtype . In contrast , although CM development was not directly assessed in these studies , the changes in limb muscle patterning observed upon deletion of Tbx3 or Osr1 in the lateral plate–derived lineage ( Prx1-cre ) include selective alterations of muscle attachment sites or fiber orientation reminiscent of those occurring in Fat1 mutants [89 , 95] . The role played by the Pdgfrα-dependent lineage in Fat1-driven developmental myogenesis is interesting in light of the emerging role of muscle-resident CT lineages for adult skeletal muscle . Pdgfrα expression distinguishes a population of muscle-resident cells called fibroadipogenic progenitors ( FAPs ) [96 , 97] . These cells do not directly contribute to the myogenic lineage but are bipotential progenitors capable of giving rise to adipocytes and to CT fibroblasts [96 , 97] . Whereas FAPs are quiescent under homeostatic conditions , muscle lesions activate their proliferation and differentiation [97] , in addition to activating muscle stem cells’ proliferation and triggering the production of new muscle fibers [2] . This damage-induced FAP recruitment was shown to contribute to fibrosis and fat infiltrations in the context of either acute lesions or chronic degeneration/regeneration cycles occurring in a mouse model of Duchenne muscular dystrophy ( mdx mice ) [98 , 99] . In cocultures of FAPs and muscle stem cells , FAPs promote myogenic differentiation , indicating that the recruitment and activation of FAPs in degenerating muscles could contribute to the efficiency of muscle repair [97] . Future work will establish whether Fat1 activity is also necessary for the adult FAP lineage to exert its myogenesis-promoting activity during muscle homeostasis and repair . We find that in addition to its mesenchymal contribution , Fat1 plays additional cell-autonomous functions in the MN pool innervating the CM muscle . Fat1 protein is selectively expressed in the CM MN pool , likely distributed along their axons . MN-specific Fat1 deletion negatively influences axonal growth of CM-innervating axons and alters acquisition of MN fate characteristics . Unexpectedly , although MN-specific Fat1 ablation does not significantly influence the rate of myogenic differentiation , our quantitative analysis uncovers a significant delay in the subcutaneous progression of muscle progenitor migration . This uncovers an unsuspected role of motor axons in the rate of progression of muscle progenitor migration . The MN specification and axonal growth defects occurring in absence of MN-Fat1 may reflect a cell-autonomous role of Fat1 signaling in MNs . Future lines of investigation will determine whether such a Fat1 signaling cascade involves interactions with Dachsous cadherins and how it influences axon growth and modulates Runx1 and Clusterin expression . It will be relevant to discriminate whether MN-Fat1 ( 1 ) modulates the response to—or reception of—GDNF by controlling receptor expression , or efficiency of the GDNF/Ret signal transduction; ( 2 ) impinges on Etv4 transcriptional activity by regulating its phosphorylation; or ( 3 ) acts directly on target gene expression via other cascades known to be downstream of FAT cadherins , such as Hippo/YAP or the transcriptional repressor atrophin . As a protocadherin , Fat1 may also refine neural circuit assembly by mediating axonal self-recognition similar to functions of clustered protocadherins [100 , 101] . Alternatively , the two phenotypes could simply reflect the consequence of the reduction in GDNF levels resulting from the altered progenitor progression also resulting from MN-specific Fat1 ablation . Both phenotypes are reminiscent of the consequences of Gdnf ablation ( as illustrated here and in [39] ) . Gdnf is also known for exerting an axon growth promoting role [102] , and in MN-specific Fat1 mutants , the length of CM motor axons remains correlated with the size of the Gdnf-expressing progenitor sheet . We have uncovered that CM motor axons progress hand in hand with Gdnf+/Pax7+ muscle progenitors and shown that disrupting Fat1 activity in MNs interferes with the rate of subcutaneous progression of CM progenitors . This effect is , however , less predominant than the dependency of CM progenitor progression on mesenchymal Fat1 activity . Such an influence of MNs on myogenic progenitors is counterintuitive , because previous studies supported the idea that muscle development can occur normally in absence of motor innervation [103 , 104] . These studies were examining NMJ maturation in the diaphragm muscle at fetal stages after genetic depletion of MNs . They showed that whereas a pre-pattern of AChRs prefiguring future synapses does occur in the diaphragm in absence of motor axons , the latter are required for subsequent NMJ maturation . The maturation-inducing activity elicited by MNs on muscles is mediated by signals such as Agrin , inducing AChR clustering [105] . These studies do not exclude , however , that the dynamics of muscle morphogenesis could have been altered at the primary myogenesis stage in some vulnerable muscles and compensated at secondary myogenesis stages . Such an influence of MNs on the development of another tissue type is also reminiscent of the role exerted by MNs , by way of VEGF secretion , on vascular development [106] . Finally , because axonal elongation is concomitant with spreading of the CM myogenic progenitor , motor axons and myogenic progenitors collectively expose themselves to the mesenchymal landscape and to mechanical properties of the extracellular matrix ( ECM ) , such as its stiffness . Interfering with the progression of one of the two cell types thus exposes the other on its own to the mesenchyme , rendering it less permissive and slowing the collective progression . In spite of this delayed progenitor progression , myogenic differentiation is not impacted in the absence of MN-Fat1 , and the resulting anatomy of the adult CM is normal . MN-specific Fat1 ablation leads to significant functional impact on NMJ integrity in absence of anatomical abnormalities of the muscle and of changes in muscle fiber diameter . This highlights the importance of MN-Fat1 in ensuring NMJ integrity . This effect could be the direct or indirect consequence of ( 1 ) Clusterin or Runx1 lowering in MNs , ( 2 ) reduced GDNF levels in muscle progenitors , ( 3 ) a reduced number of muscle progenitors persisting in the adult muscle , or ( 4 ) the deregulation of other target genes influencing NMJ integrity that remain to be identified . Muscle phenotypes caused by disruption of Fat1 functions are highly regionalized , and the topography is reminiscent of the map of muscles undergoing degeneration in facioscapulohumeral dystrophy ( FSHD ) [14] . FSHD is the second most frequent muscular dystrophy , characterized by the selective map of affected muscles involving facial expression and shoulder muscles [107] . In most cases , FSHD is caused by genomic alterations on chromosome 4q35 , leading to excess production of the double homeodomain protein DUX4 , a transcription factor encoded in a repeated macrosatellite , D4Z4 , normally active only during zygotic gene activation [107 , 108] . This overexpression is permitted by a genetic context simultaneously involving ( 1 ) a polymorphism stabilizing DUX4 RNA by creating a poly adenylation signal [109] and ( 2 ) the loss of epigenetic DUX4 repression [110] . Such epigenetic de-repression results ( a ) in the most frequent form of the disease ( FSHD1 ) , from deletions that shorten the D4Z4 repeat array [110] , and ( b ) in less frequent cases ( FSHD2 ) , from mutations in genes involved in modulating the chromatin architecture , SMCHD1 [111] , or in DNA methylation ( DNMT3A [112] ) . DUX4 exerts a toxic effect on muscles by deregulating the transcription of multiple targets [113–115] . Individuals carrying a pathogenic context exhibit a high variability in symptom severity [116 , 117] , implying that other factors behave as disease modifiers . The similarity between muscular and nonmuscular phenotypes of Fat1 mutants and FSHD symptoms indicates that Fat1 deficiency phenocopies some FSHD symptoms [14] , raising the possibility that FAT1 dysfunction in FSHD-relevant tissue types could contribute to some of these phenotypes . This possibility was supported by the identification of pathogenic FAT1 variants in human patients , in which FSHD-like symptoms were not caused by the traditional genetic causes of FSHD [14 , 118] . Furthermore , FAT1 is located in the vicinity of the FSHD-associated D4Z4 array on 4q35 , and its expression is down-regulated in fetal and adult muscles of FSHD1 and FSHD2 patients [14 , 119] , suggesting that traditional FSHD-causing contexts may not only enhance DUX4 expression but also lead to lowered FAT1 expression in affected muscles . Although exogenous DUX4 can down-regulate FAT1 expression when exogenously expressed in human myoblasts , short hairpin RNA ( shRNA ) -mediated DUX4 silencing in FSHD1 myoblasts did not restore FAT1 expression levels [119] . Thus , FAT1 deregulation by a pathogenic 4q35 context is not attributable to DUX4 in FSHD1 myoblasts and might be caused by DUX4-independent mechanisms . First , the loss of epigenetic repression of DUX4 might be due to changes in the chromatin architecture resulting from deletions of anchoring sites for chromatin folding proteins such as CTCF or SMCHD1 [111 , 120] . Given the proximity of FAT1 with the D4Z4 array , the same changes in chromatin architecture could simultaneously lead to lowered FAT1 levels . Second , among genetic variants in FAT1 , a copy number variant ( CNV ) deleting a putative regulatory element in the FAT1 locus , which has the potential to deplete FAT1 expression in a tissue-specific manner , was enriched not only among FSHD-like patients but also among classical FSHD1 and FSHD2 [14 , 119] . Furthermore , given the chromosomal proximity , genetic alterations in FAT1 have a high probability to co-segregate with a pathogenic FSHD context occurring on the same 4q chromosome . Support for this came from the recent identification of pathogenic FAT1 variants associated with a classical FSHD1 context [121] . Thus , the combination of a pathogenic 4q35 context with pathogenic FAT1 variants , whether functional or regulatory , has the potential to synergize to reduce FAT1 expression and/or activity below a functional threshold , potentially contributing to worsening FSHD symptoms . Other genetic modifiers affecting chromatin methylation [111 , 112] and/or architecture of the 4q35 arm [122] , hence affecting gene regulation , may simultaneously impact DUX4 and FAT1 expression . Therefore , elucidating Fat1 functions during development could provide instructive information to guide research on FSHD pathogenesis or on pathologies associated with FAT1 alterations in humans . Our findings indicate that myoblasts may not be the only relevant cell type in which FAT1 down-regulation must occur to alter muscle development , and identify mesenchyme in developing embryos as another cell type in which Fat1 disruption is sufficient to cause muscle shape phenotypes with an FSHD-like topography . Whether FAT1 is indeed deregulated in CT of FSHD patients , and whether such regulation change is accounted for by DUX4 or by 4q35 chromatin architecture changes , remains to be established . Several FSHD mouse models exist , some of them engineered for inducible and tissue-specific DUX4 overexpression , and will be instrumental to assess to what extent and through which mechanism loss of Fat1 function could synergize with enhanced DUX4 levels [123–126] . Furthermore , our results also identify GDNF as a muscle-derived signal , the lowering of which is associated with muscular phenotypes of Fat1 mutants and has consequences at least on neuronal development and possibly on NMJ integrity in adult muscle . Interestingly , GDNF appeared significantly down-regulated in FSHD biopsies in at least two reports [127 , 128] . Another study suggests instead that FSHD involves enhanced Ret signaling in myoblasts and proposes Ret inhibitors as a possible therapeutic strategy [129] . In view of these reports and together with our present findings , it may be relevant to determine whether and to what extent modulation of GDNF/Ret signaling , or of other mechanisms influenced by Fat1 signaling , can have any benefit to alleviate adult muscle symptoms and improve quality of life of FSHD patients . Finally , developmental alterations such as those occurring in the Etv4/GDNF circuit might constitute a topographic frame in which muscles have deficient functional output . It will therefore be necessary to understand whether and how this renders the muscles more susceptible to degeneration in adults in an FSHD context and whether FAT1 dysfunction at adult stages further destabilizes muscle homeostasis and contributes to degeneration .
All procedures involving the use of animals were performed in accordance with the European Community Council Directive of 22 September 2010 on the protection of animals used for experimental purposes ( 2010/63/UE ) . The experimental protocols were carried out in compliance with institutional Ethics Committee guidelines for animal Research ( comité d’éthique pour l’expérimentation animale–Comité d’éthique de Marseille ) and in compliance with French law , under an agreement ( Number D13-055-21 ) delivered by the “Préfecture de la Région Provence-Alpes-Côte-d’Azur et des Bouches-du-Rhône . ” When collecting embryos , adult mice were euthanized by cervical dislocation . For collection of adult muscle samples , mice were first anesthetized with lethal doses of Ketamin and Xylasine , and euthanized by transcardiac perfusion with 4% PFA . All the mouse lines and genetic tools used here are described in S1 Table , providing for each of them a link to the mouse genome informatics ( MGI ) information page , a reference describing their production and some brief explanations on what they are , how they were made , and how they are used in the present study . The alleles of the Fat1 gene used in this study included a genetrap knockout/knock-in allele Fat1KST249 ( referred to as Fat1LacZ ) [130] , as well as a constitutive knockout Fat1ΔTM allele ( here referred to as Fat1- ) and the conditional-ready Fat1Fln allele from which it derived , that we previously described [14] . As the original conditional-ready Fat1Fln allele still carried a neo selection cassette flanked by FRT sites [14] , we have crossed Fat1Fln/+ mice with mice carrying the Flp recombinase ( ActFlpe mice , official name Tg ( ACTFLPe ) 9205Dym [131] ) , thereby producing a new conditional-ready allele referred to as Fat1Flox , used throughout this study . The Cre lines used for tissue-specific ablation were either transgenic lines ( official MGI nomenclature Tg ( gene-cre ) author ) , or knock-in/knockout alleles ( official MGI nomenclature genetm ( cre ) author ) . These include Olig2-cre ( knock-in/knockout Olig2tm1 ( cre ) Tmj [72] ) ; Wnt1-cre ( transgenic Line Tg ( Wnt1-cre ) 11Rth [75] ) ; Prx1-cre ( transgenic Line Tg ( Prrx1-cre ) 1Cjt [73 , 74] ) , Pax3-cre ( knock-in/knockout line Pax3tm1 ( cre ) Joe [64] ) , and the tamoxifen-inducible Pdgfrα-iCre line ( BAC transgenic line Tg ( Pdgfrα-cre/ERT2 ) 1Wdr [79] ) . To produce embryos or mice Cre+; Fat1Flox/Flox , all cre alleles were transmitted by males only to avoid female germ line recombination by crossing Cre+; Fat1Flox/+ males or Cre+; Fat1FloxFlox males with Fat1FloxFlox females , the latter carrying one of the reporter genes described below ( MLC3F-2E-LacZ , GdnfLacZ , R26-YFP ) . Knock-in/knockout Pea3-lacZ mice ( here referred to as Etv4- ) were used with the permission of S . Arber and T . Jessell and genotyped as previously described [41] . Transgenic mice carrying the BAC transgene Etv4GFP have been described previously [60]; knock-in/knockout GdnfLacZ mice were used with permission of Genentech , [39 , 132]; knock-in/knockout mice carrying the Metd allele produce a signaling dead version of the Met receptor [61] . Additional lines used in this study are: Rosa26-YFP mice ( Gt ( ROSA ) 26Sortm1 ( EYFP ) Cos line; Jackson laboratory mouse strain 006148 [66] ) ; MLC3F-2E-LacZ mice ( transgenic line Tg ( Myl1-lacZ ) 1Ibdml [58 , 59] ) ; HB9-GFP ( transgenic line Tg ( Hlxb9-GFP ) 1Tmj allele [133] ) ; Myf5-cre mice ( Myf5tm3 ( cre ) Sor line; Jackson laboratory mouse strain 007893 [80] ) ; R26Lox-LacZ-STOP-Lox-DTA mice ( also called R26DTA or Gt ( ROSA ) 26Sortm2 ( DTA ) Riet line [81] ) . All lines were maintained through backcrosses in B6D2 F1-hybrid background . Staining for β-galactosidase activity was performed using two alternative methods . To detect LacZ expression in embryos with the MLC3F-2E-LacZ transgene or Fat1LacZ allele , we performed classical X-gal ( 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside ) staining with standard protocols using potassium ferricyanide and potassium ferrocyanide . Embryos were then postfixed in PFA , partially clarified in 1 volume glycerol/1 volume PFA 4% , and mounted in 100% glycerol on glass slides with depression , under coverslips with glass spacers for imaging . When detecting LacZ expression in embryos carrying the GdnfLacZ allele or Fat1LacZ allele on cryosections , we selected the more sensitive method using Salmon-Gal ( 6-chloro-3-indolyl-beta-D-galactopyranoside , Appolo Scientific UK , stock solution 40 mg/mL in DMSO ) in combination with tetrazolium salt ( nitro blue tetrazolium [NBT] , N6876 Sigma ) , stock solution 75 mg/mL in 70% DMF ) , using the method we previously described [46] . Briefly , after initial fixation ( for GdnfLacZ: 15 minutes in 1% PFA; 0 . 2% glutaraldehyde in PBS ) , embryos were rinsed in PBS and incubated overnight , in order to avoid background staining , at 37 °C in a solution containing 4 mM potassium ferricyanide , 4 mM potassium ferrocyanide , 4 mM MgCl2; 0 . 04% NP40 , but without substrate . Embryos were then rinsed 3 times in PBS and incubated in a solution containing Salmon-Gal ( 1 mg/mL ) , NBT ( 330 μg/mL ) , 2 mM MgCl2; 0 . 04% NB40 , in PBS . After staining completion , embryos/sections are rinsed in PBS and postfixed in 4% PFA . For GdnfLacZ , the staining reaction is completed in 30–40 minutes . For staining of cryosections of Fat1LacZ embryos , the staining reaction is completed in 1–2 hours . Embryos were collected in cold PBS , and when necessary , spinal cords were dissected at that stage . Dissected spinal cords and the remaining embryo trunks or whole embryos were then fixed overnight in 4% paraformaldehyde ( in PBS ) , dehydrated in methanol series , and stored in methanol at −20 °C . In some cases , embryos were embedded in gelatin/Albumin and sectioned on a vibratome ( 80 um sections ) . These floating sections were then dehydrated in methanol series and treated like whole-mount embryos . In situ hybridizations were done as previously described [47] . Briefly , after methanol dehydration , tissues ( embryos , spinal cords , or vibratome sections ) were rehydrated in inverse methanol series , treated with proteinase K , hybridized with Digoxigenin-labeled probes , washed with increasing stringency , incubated with anti-Digoxigenin antibodies conjugated with alkaline phosphatase ( Roche ) , and developed with NBT/BCIP substrates . For double in situ hybridizations , hybridization was done by mixing a Digoxigenin-labeled and a Fluorescein-labeled probe . Revelation of the two probes was done subsequently . We first incubated the samples with the AP-conjugated anti-digoxigenin antibody ( 1/2 , 000 , Roche ) , washed , and stained using NBT/BCIP ( blue color ) . Samples were postfixed , cleared with 50% glycerol , and flat-mounted to be photographed with the first color only ( blue ) . To start the second staining , the samples were then transferred back in PBT ( PBS , 0 . 1% Tween-20 ) . The remaining alkaline phosphatase was then inactivated for 10 minutes with 0 . 2 M glycine pH 2 . 2; 0 . 1% Tween-20 . Samples were further rinsed 4 times in PBT , after which they were incubated with AP-conjugated anti-fluorescein antibody ( Roche , 1 in 2000 dilution ) . After washing , this second staining was revealed using INT/BCIP substrate ( Roche , red color ) . Samples were postfixed again , cleared in 50% glycerol , and flat-mounted for visualization . Embryos were collected in cold PBS and fixed 2–4 hours in 4% PFA on ice . Adult mouse tissues were collected after perfusing anesthetized mice with 4% PFA and were postfixed for 2–4 hours in 4% PFA . The samples were rinsed in PBS , cryoprotected in 25% sucrose-PBS solution , embedded in PBS , 15% sucrose/7 . 5% gelatin , and cryostat sectioned ( 10 μm for embryos , 30 μm for NMJ analysis in adult muscle longitudinal sections ) . For fluorescent IHC , slides were thawed in PBS , then incubated in PBS , 0 . 3% triton-X-100 , bleached in 6% H2O2 for 30 minutes , and rinsed again in PBS , 0 . 3% triton-X-100 . For some antibodies ( see S2 Table ) , heat-induced epitope retrieval was performed in 0 . 1 M citrate buffer pH 6 at 95 °C for 5–30 minutes ( specific time depending on the antibody used ) . Slides were rinsed again 3× in PBS , 0 . 3% triton-X-100 and then incubated overnight at 4 °C with primary antibodies in 20% newborn calf serum , PBS , 0 . 3% triton-X-100 in a humidified chamber . Primary antibodies or detection tools used are described in S2 Table . After extensive washing in PBS , 0 . 3% triton-X-100 , slides were incubated with secondary antibodies in blocking solution as above , for 1 hour and 30 minutes at room temperature in a humidified chamber . Secondary antibodies used were conjugated with Alexa-555 , Alexa-488 , or Alexa-647 ( see S2 Table ) . After extensive washing in PBS , 0 . 3% triton-X-100 , slides were quickly rinsed in distilled water and mounted in mounting medium with ProlongGold antifade reagent and DAPI . Slides were kept at 4 °C and imaged with a Zeiss Z1 Microscope equipped with Apotome for structured illumination , using Zen software . Whole-mount anti-neurofilament staining of E12 . 5 embryos was performed as previously [46] , using the mouse monoclonal anti-neurofilament ( 2H3 ) antibody ( bioreactor version , dilution 1/300 ) and anti-mouse-HRP secondary antibody ( Sigma ) . Briefly , embryos were fixed 2–4 hours in 4% PFA ( when relevant , an X-gal staining was also carried out ) , rinsed in PBS , then transferred in Dent’s fixative ( 80% methanol; 20% DMSO ) and kept at 4 °C until use . A bleaching step with hydrogen peroxide was carried out by incubating embryos overnight in a solution with 1 volume 30% H2O2; 4 volumes Dent’s fixative . All washes were made in PBS , 1% triton X-100 . Antibody incubations were done by incubating at least 3 days ( at least one at room temperature ) in a blocking solution containing 79% newborn calf serum , 20% DMSO , 1% triton X-100 , and thimerosal . The final HRP color reaction was performed with DAB tablets ( SIGMA ) and H2O2 , after rinsing in 0 . 1 M Tris pH 7 . 4 . After staining completion , embryos were dehydrated in methanol , cleared in 2 volumes benzyl benzoate/1 volume Benzy-Alcohol mix ( BB-BA mix; once embryos are in 100% methanol , they are transferred in a mix 50% volume methanol; 50% volume BB-BA , for 30 minutes , then twice in 100% BB-BA mix ) , cut in half , and mounted on glass slides with depression , under coverslips with glass spacers for imaging . In cases of crosses with the Pdgfrα-iCre line , tamoxifen induction was performed by intraperitoneal ( IP ) injection in pregnant females . Tamoxifen ( T5648 , Sigma ) was dissolved first in ethanol ( 100 mg/mL ) under agitation ( in a thermomixer ) for 2 hours at 42 °C , then diluted in sterile corn oil ( C8267 , Sigma ) at 10 mg/mL , kept under a chemical hood at room temperature to let the ethanol evaporate , and aliquoted ( 2 mL aliquots ) for storage at −20 °C . One working aliquot was kept at 4 °C for 1–2 weeks . To achieve recombination in mesenchyme-derived loose CT , two successive injections were performed in pregnant females , first at E9 . 75 at a dose of 50 μg/g ( per gram of tissue weight ) , and second the following day at a dose of 100 μg/g . Administration was done by IP injection using a 26G syringe . Efficiency of recombination was assessed using the reporter line R26-YFP , first by following overall intensity YFP fluorescence on freshly dissected embryos , and later by performing anti-YFP IHC on cryosections , combined with markers of tissue types . Quantifications of signal intensity were performed using the Image J software ( NIH ) as previously described [46 , 84] . Briefly , color images were first converted to gray scale and inverted to negative scale ( a scale with white for highest signal intensity [250] and black for the lowest intensity [0] ) . An area of fixed pixel width was cropped , and signal intensity was measured with ImageJ along a horizontal band with the width of the cropped image ( 500 pixels , matching half a flat-mounted spinal cord , along dotted lines in quantified pictures , always ending in a fixed positional landmark , such as the spinal cord midline ) , and height of a few pixels . After background subtraction ( background being the baseline value of the control samples in the same experiment , and baseline being the minimum value in a fixed window , systematically avoiding points such as the midline , where changes in tissue thickness influence background intensity ) and threshold subtraction ( to avoid negative values ) , the values were expressed as a percentage of the mean maximal amplitude of the control samples used in the same experiment ( so that experiments developed independently can be normalized to their respective controls , and pooled ) . A mean distribution along the line is generated by averaging the normalized intensities between several samples of each genotype ( considering left and right sides separately , the number of spinal cord sides used , for each plot being indicated in the corresponding figures ) to generate an average signal distribution plot , shown ± standard deviation ) . The total signal intensity value was also calculated for each sample by measuring the area under the curve and plotted individually as percentage of the mean control value . Area measurements were used to assess CM subcutaneous progression and NMJ size . Whole embryo images were acquired on a Leica Stereomicroscope , using Axiovision software . We used the area/distance measurement tools in the Axiovision software . For each embryo side , we measured the area occupied by the CM , detected with MLC3F-2E-LacZ by X-Gal staining , with GdnfLacZ using Salmon-Gal staining , or with anti-neurofilament IHC . As reference measurements for normalization , we measured structures that grow proportionally with the embryo age . For this , we measured , respectively , the area occupied by BW muscles ( MLC3F-2E-LacZ ) , the length of the trunk between two fixed GdnfLacZ+ landmarks ( shoulder to posterior limit of the hind limb ) , and the dorsoventral length of a given thoracic nerve ( T10 or T9 ) for anti-neurofilament stainings . For all measurements , the first graphs show the raw values , whereas the second plots represent the normalized area , obtained by calculation the ratio CM area/reference value , and dividing this value by the median ratio for a set of control embryos for normalization . For LacZ-stained embryos , measurements were done on embryos irrespective of the storage time and medium ( PBS/PFA or glycerol , with different degrees of tissue shrinkage ) ; normalization was typically done by comparing embryos of the same litter and mounted/imaged at the same time . For anti-neurofilament staining , as all embryos were stored in BB-BA ( where tissue shrinkage is invariable ) , normalization was done by ranking embryos by size range ( according to the thoracic nerve length ) and comparing embryos in the same size range . For NMJ area measurements , IHC was performed on 30 μm thick longitudinal cryosections of the dCM muscle to detect the neuromuscular junctions ( with α-BTX ) , the nerve endplate ( with anti-neurofilament antibodies ) , and the muscle fiber F-actin ( with Phalloidin ) ( see S2 Table for reagents ) . NMJ images were acquired on Zeiss Z1 Microscope equipped with Apotome for structured illumination and processed using Zen Imaging software ( Zeiss ) or Axiovision software . After NMJ Image acquisition , a maximal intensity projection image was generated . The latter was converted from . czi to . zvi file , and we used the area measurement tool in the Axiovision software . For each mouse , a minimum of 20 NMJs were measured in the dCM using pictures acquired at magnifications ranging from 10× to 63× ( keeping the scale adapted to the magnification to measure areas in μm2 ) , and the percentage of NMJs matching a set of area ranges was determined per mouse . As control , on the same set of images , the diameter of each muscle fiber was also measured ( on longitudinal sections , the MIP shows the maximal width across the section thickness , thus accurately measuring fiber diameter ) . Even when no specific staining of the fiber was performed , background levels of the α-BTX staining are sufficient to accurately detect muscle fibers and measure their width . For each mouse , around 20–50 fiber diameters were measured , and the percentage of fibers matching a set of diameter ranges was determined in each mouse . The NMJ area and fiber diameter size distribution graphs in Fig 12 represent mean percentage in each area range for each genotype ± SEM , indicating the number of mice used next to the genotype indication . Control data from Fat1FloxFlox mice are shown twice ( in blue ) in both graphs , to make readability easier . Grip tests were performed on adult mice using the Grip test apparatus ( Bioseb ) , which measures the maximum force exerted by the mouse as it pulls a sensitive device ( grid ) until release of the grip . We measured either forelimb strength only or the cumulated force of forelimbs + hind limbs ( when a mouse uses all 4 limbs to pull the measurement grid ) . As raw force in the measured unit ( Newton ) was significantly different between males and females ( controls ) , the data were normalized for each gender and genotype by dividing the measured value by the average value of the control group of the same gender . This allows expressing the output in percentage of the control value , set as 100% in average , hence allowing us to pool data from males and females . In the results shown in Fig 12 , male and female data have been pooled , as the separate analysis of both genders gave similar results . Three to four measurements were made for each mouse , and an average value was calculated . In both the pooled and non-pooled graphs , each dot represents the average value for one mouse . Statistic tests were carried out on the average values . StatEL ( add-in to Excel , by Ad Science , Paris , France ) was used for statistical analyses . For comparisons between two groups , differences were assessed either using the unpaired Student t test , when data were showing a normal distribution and equal variance , or using the nonparametric Mann Whitney test otherwise . All p-values are indicated in the figures , except for Fig 9 , where * indicates p-value < 0 . 01 , whereas ** indicates p-value < 0 . 001 . Differences were considered significant when p < 0 . 05 . For signal intensity distribution curves , average values ± standard deviations are shown ( in light blue ) . For dot plots , all individual data are plotted ( one per sample ) , and the median value is shown with a black bar . | Fat cadherins are evolutionarily conserved cell adhesion molecules , which play key roles in modulating tissue morphogenesis through the control of collective cell behavior and polarity . We previously identified the mouse Fat1 gene as a regulator of muscle morphogenesis and reported a role for this gene in muscle progenitors to modulate their migration polarity . Recent findings have revealed a potential link between muscle patterning and non-connective tissues . Here we have analyzed the mechanisms that coordinate the behavior of two cell types , mesenchymal cells and brachial spinal motor neurons , during mouse neuromuscular morphogenesis . We show that Fat1 disruption in connective tissue robustly alters muscle morphogenesis of the cutaneous maximus muscle , affecting not only migration of progenitors and expansion of myofibers but also subsequently impairing axon growth and specification of cognate motor neurons . We observe that Fat1 acts in motor neurons in parallel to modulate axonal growth and neuronal specification , modestly influencing muscle morphology . Together , these results show that Fat1 coordinates the coupling between muscle and neuronal development by playing complementary functions in mesenchyme , muscles , and motor neurons . These findings could guide research on muscle pathologies associated with FAT1 alterations in humans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"nervous",
"system",
"population",
"genetics",
"neuroscience",
"gene",
"pool",
"developmental",
"biology",
"nerve",
"fibers",
"population",
"biology",
"embryos",
"immunologic",
"techniques",
"morphogenesis",
"research",
"and",
"analysis",
"methods",
"embryology",
"spinal",
"cord",
"musculoskeletal",
"system",
"muscle",
"differentiation",
"animal",
"cells",
"axons",
"muscle",
"fibers",
"muscles",
"immunohistochemistry",
"techniques",
"cellular",
"neuroscience",
"neuroanatomy",
"cell",
"biology",
"anatomy",
"neurons",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"evolutionary",
"biology",
"histochemistry",
"and",
"cytochemistry",
"techniques"
] | 2018 | Tissue-specific activities of the Fat1 cadherin cooperate to control neuromuscular morphogenesis |
B-cells not only produce immunoglobulins and present antigens to T-cells , but also additional key roles in the immune system . Current knowledge on the role of B-cells in infections caused by intracellular bacteria is fragmentary and contradictory . We therefore analysed the phenotypical and functional properties of B-cells during infection and disease caused by Mycobacterium tuberculosis ( Mtb ) , the bacillus causing tuberculosis ( TB ) , and included individuals with latent TB infection ( LTBI ) , active TB , individuals treated successfully for TB , and healthy controls . Patients with active or treated TB disease had an increased proportion of antibodies reactive with mycobacteria . Patients with active TB had reduced circulating B-cell frequencies , whereas only minor increases in B-cells were detected in the lungs of individuals deceased from TB . Both active TB patients and individuals with LTBI had increased relative fractions of B-cells with an atypical phenotype . Importantly , these B-cells displayed impaired proliferation , immunoglobulin- and cytokine- production . These defects disappeared upon successful treatment . Moreover , T-cell activity was strongest in individuals successfully treated for TB , compared to active TB patients and LTBI subjects , and was dependent on the presence of functionally competent B-cells as shown by cellular depletion experiments . Thus , our results reveal that general B-cell function is impaired during active TB and LTBI , and that this B-cell dysfunction compromises cellular host immunity during Mtb infection . These new insights may provide novel strategies for correcting Mtb infection-induced immune dysfunction towards restored protective immunity .
Human B-cells not only mediate humoral immunity but are also key players in the initiation and regulation of T-cell responses . B-cells can act as professional antigen presenting cells , provide co-stimulatory signals , produce cytokines and can exert immunoregulatory properties . Antigen uptake by B-cells typically occurs via the B-cell-receptor; however , live mycobacteria can also infect B-cells through macropinocytosis , resulting in MHC class II antigen presentation [1–3] . Although less appreciated , B-cells exist in multiple flavours , not unlike the large variety of T-cell subsets . By implication , the type of B-cell that activates T-cells may critically determine the final fate and direction of the ensuing T-cell response . B-cells can be divided into subpopulations based on lineage and differentiation markers , and include naïve B-cells , immature B-cells , plasma cells , regulatory B-cells ( Bregs ) and memory B-cells [4] . Memory B-cells can be further subdivided into classical , active and atypical B-cells , based on the combined expression patterns of CD21 and CD27 or IgD and CD27 [5 , 6] . The role of B-cells in infectious diseases , in particular intracellular bacterial infections such as with Mycobacterium tuberculosis ( Mtb ) has not been investigated in great detail , mostly because B-cell derived immunoglobulins were considered not to play a prominent role in infections with intracellular pathogens [7] . However , B-cells have lately been rehabilitated as important players in the immune response during chronic inflammation irrespective of immunoglobulin production . Nevertheless , in human TB B-cell phenotypes and function have not been extensively investigated . Studies enumerating B-cells in patients with TB disease have yielded conflicting results , not only in patients with active pulmonary TB but also in latently TB infected individuals ( LTBI ) . In active TB , B-cell frequencies have been reported to be unaltered [8]; increased [9]; or decreased [10 , 11] compared to healthy controls . In addition , compared to healthy donors , LTBI individuals have been reported to have decreased B-cell frequencies [10] , whereas those successfully treated for TB had increased B-cell frequencies [8] . In addition , patients with multi-drug resistant ( MDR ) TB were reported to have decreased frequencies of unswitched , IgD+CD27+ B-cells and decreased plasma cell frequencies , which are frequently observed during chronic inflammation [12] . The results described so far are rather conflicting and highly descriptive , without any analyses of the functional capacities of the B-cells . The only functional assessment of B-cells was described for a very small group of only 3 TB patients , which suggested hampered proliferation of B-cells following specific antigenic stimulation but did not take into account absolute B-cell numbers or phenotypes [13] . The contribution of B-cells to TB disease development has been studied in animal infection models , mostly in mice with genetic B-cell deficiencies . Infection of B-cell deficient mice with virulent Mtb resulted in enhanced pathology and increased bacterial loads [14 , 15] , depending on the route of infection , either the lung [15] or systemic [14] . Moreover , these models showed less severe granuloma formation in the lungs and delayed dissemination of bacteria , indicating a role for B-cells in coordination of granuloma formation [15–17] . Indeed , transfer of B-cells , but not immunoglobulins , restored granuloma formation and inflammation [17] . However , not all studies revealed the same phenotypes , including increased bacterial loads , in B-cell and IL-4 deficient mice . Recently , B-cell depletion studies were performed in Mtb infected non-human primates and although disease severity and clinical outcome were not altered , local inflammation and bacterial loads appeared to be critically modulated by B-cells [18] . Cytokines like T-cell derived IL-17 , IL-2 and IL-10 are increased in granuloma’s , whereas IL-6 and IL-10 are decreased [18] . Analysis of murine lungs during Mtb infection revealed an 8-fold increase in the absolute number of B-cells ( CD19+ ) in the lung , representing 6–8% of CD45+ leucocytes ( normally 2% ) , compared to very early stages of infection . These levels remained persistently elevated during chronic Mtb infection [19] . Histological analysis of the lungs showed a strong clustering of B-cells together with dendritic cells ( DCs ) , surrounded by T-cells [19] . These clusters were also observed in human lungs affected by TB disease and recognized as germinal-center like structures [15 , 16 , 20] . Characterization of these clusters suggested lymphoneogenic formation of tertiary lymphoid nodules at the site of disease . B-cells in these ectopic B-cell follicles expressed GL7 ( a classical B-cell germinal center marker ) and CXCL13 ( strongly associated with lymphoid neogenesis ) [15] . In addition to these animal and histological studies , further indications for a potential involvement of B-cells in TB have emerged from unbiased studies . Searches for host biomarkers of TB disease and of curative responses to TB treatment in blood samples revealed decreased mRNA expression of B-cell related genes during active TB disease , which increased again following treatment [21 , 22] . B-cell associated genes were among the strongest differentially regulated genes between time points of diagnosis and end of treatment [21] . Moreover , combined analysis of gene expression data from 8 independent studies also revealed strongly altered expression of B-cell related genes , suggesting that B-cells are significantly involved in TB disease [23] . In another recent study we identified IL13 mRNA expression , a B-cell promoting cytokine , months before actual TB diagnosis as a correlate of risk for the development of TB disease in HIV infected individuals [24] . Therefore , to better understand the role of B-cells during Mtb infection we have analysed B-cell phenotypes and functional properties in peripheral blood cells from patients at different stages of TB infection and disease . We also analysed in situ B cells in lung specimens from deceased TB patients . We find unexpectedly large differences between patients with active TB and controls , indicating impaired B-cell functions in active TB patients . Unexpectedly also individuals with LTBI showed reduced B-cell functionalities . Moreover , our findings demonstrate that normally functioning B-cells , which are present in treated TB patients , are important for optimal activation of human Mtb-specific T-cell immunity .
Plasma samples from Italian controls , individuals with latent Mtb infection , active or treated TB disease were tested for IgG antibodies reactive with PPD , Ag85B or ESAT-6/CFP-10 . We detected IgG antibodies against PPD in patients with active TB disease and those successfully treated for TB at a higher level compared to LTBI or control individuals ( Fig 1 ) . Antibodies reactive with Ag85B were detected in all infected groups compared to controls , albeit at very low levels ( Fig 1 ) . ESAT-6/ CFP-10 specific antibodies were detected in some individuals at a very high level , however no differences were observed between the groups ( Fig 1 ) . We analysed individuals with latent Mtb infection , active or treated TB disease and compared those to healthy uninfected controls . Patients with active TB had significantly increased numbers of leucocytes in their peripheral blood , mostly as a result of increased numbers of circulating neutrophils and monocytes ( S1A Fig ) . The percentage of lymphocytes showed a relative decrease at the time of diagnosis , although the absolute number remained constant ( S1B Fig ) . The frequency of CD19+ B-cells was significantly decreased within the total lymphocyte population in patients with active TB disease compared to healthy controls ( Fig 2A ) . Individuals with LTBI or treated TB individuals had frequencies of circulating CD19+ B-cells similar to controls ( Fig 2A ) . Frequencies of plasma cells and immature B-cells did not differ among the evaluated subjects ( Fig 2A ) ; however , the frequency of naïve B-cells was decreased in patients with active TB and trended to a decreased frequency in individuals with LTBI compared to controls ( Fig 2A ) . The frequency of CD19+ B-cells did not correlate with the levels of IgG antibodies reactive with PPD , Ag85B or ESAT-6/CFP-10 within the same individuals . To assess the quality and specific characteristics of the B-cells from TB patients we investigated the distribution of regulatory B-cells ( Bregs ) and memory B-cell subsets . In contrast to previous studies [25 , 26] we did not observe any differences in CD1dHI Bregs or CD25+CD71+ Bregs between any of the groups ( Fig 1B ) . We also assessed the alternative CD24HICD27+ and CD24HICD38HI Breg subsets , as they have been reported in other infectious diseases [27] , however there were no significant changes observed between ( treated ) TB patients , LTBI individuals or controls ( Fig 2B ) . Taken together , we did not observe significant changes in four different Breg populations . B-cells can be divided in memory subsets based on the combined expression patterns of CD21 and CD27 , or the combined expression of IgD and CD27 . Naive B-cells are identified as CD21+CD27- or IgD+CD27-; classical B-cells as CD21+CD27+ or IgD+CD27+; atypical B-cells as CD21-CD27- or IgD-CD27- and activated B-cells as CD21-CD27+ or IgD-CD27+ ( B-cell differentiation scheme in S4A Fig ) . Analysis of these three markers over the different groups , using concatenate analysis ( which combines expression patterns of all individuals within a group ) revealed an altered expression of CD21 and IgD in active TB patients and individuals with LTBI ( Fig 3A; gating in S2 Fig ) . Analysis of the memory B-cell subsets during Mtb infection/disease showed an increased representation of double-negative atypical memory B-cells both in active TB patients and individuals with LTBI compared to controls and TB treated subjects ( Fig 3B and 3C ) . In addition , active TB patients and individuals with LTBI had a larger proportion of IgD-CD27+ activated B-cells and fewer IgD+CD27- naive B-cells ( summary of data in S4B Fig ) . Atypical , double negative , B-cells are also referred to as tissue-like memory B-cells or exhausted B-cells and have predominantly been described in chronic viral infections such as HIV [5] . They can be further characterized by the expression of exhaustion associated molecules such as FcRL4 , CD85J and CD22 . We have assessed the expression of these markers within the population of circulating atypical B-cells ( IgD-CD27- ) , however we did not observe statistically significant differences between patents with active TB , LTBI , treated TB or controls ( Fig 3D , 3E and 3F ) . Nevertheless , it was remarkable that some individuals with active TB disease or LTBI expressed these markers on a high proportion of their atypical B-cells . This suggests that both TB patients and LTBI subjects have an increased proportion of B-cells with an atypical phenotype . To assess the specific phenotypical and numerical characteristics of B-cells at the site of TB disease , an immunohistochemical analysis on the lung specimens from autopsies of patients with pulmonary TB was performed ( Table 2 ) . Samples from subjects with pneumonia different from TB or subjects who died from causes other than pneumonia , such as liver cirrhosis or heart failure , were evaluated as controls ( Table 2 ) . In particular , the samples from patients that had died from causes other than pneumonia displayed lung parenchyma characterized by fibrin aggregates , hyaline membrane or interstitial fibrosis as seen in respiratory distress syndrome . Therefore , only few inflammatory cells were found in these samples . The histological examination showed alveolar edema and interstitial inflammation consisting of CD14+ macrophages ( Fig 4A ) and CD3+ T-cells ( Fig 4D ) . Rare CD20+ B-cells were observed ( Fig 4L , 4M and 4N ) . Moreover , few Ki67 positive cells were found ( Fig 4G ) . The lung specimens from subjects whom died from pneumonia not related to active TB were characterized by congested alveolar septa , airspaces filled with fibrin and neutrophils , with or without necrosis . Lymphocytes were only focally grouped or dispersed in the interstitial space . Pathological findings showed alveolar damage with neutrophilic infiltrates associated with abundant CD14+ macrophages ( Fig 4B ) ; CD3+ T-cells ( Fig 4E ) were found in the wall of vessel . Few Ki67+ cells ( Fig 4H ) and CD20+ B-cells were seen in the lung parenchyma ( Fig 4L ) . The lung samples from active TB patients showed granulomas characterized by a central region of caseous necrosis surrounded by foamy and epithelioid macrophages , as well as Langhans giant cells . Granulomas contained mainly CD14+ macrophages and CD3+ T-cells ( Fig 4C and 4F respectively ) ; a low Ki67 expression was found ( Fig 4I ) . B cells were rarely found in the lung parenchyma ( Fig 4L ) , as in patients with pneumonia ( Fig 4K ) , and in the patients died from pneumonia other than TB; differently they were found mainly around the granuloma ( Fig 4M ) or in the outermost layer of the granuloma , near the vessels ( Fig 4N ) , organized in pseudofollicles ( Fig 4N ) with a submembranous and membranous staining pattern . Based on the score used to classify the presence of CD20+ cells , 80% of the samples analyzed from the patients with active TB had a scoring higher than 1 ( Table 3 ) whereas only 40% of the controls showed a score higher than 1 ( Table 3 ) ; this difference was not significant . Representative images of CD20+ scores are shown in S3A–S3D Fig . Prompted by the finding of phenotypically atypical ( CD21-CD27- or IgD-CD27- ) B-cells in the circulation of active TB patients , we isolated CD19+ B-cells from all groups and stimulated equal numbers of these cells with a strong polyclonal activation signal , consisting of anti-CD40 combined with anti-IgG/IgM antibodies to assess the functional properties of these B-cells . Proliferation , as measured using a cell cycle tracking dye , was strongly decreased in B-cells from both TB patients and showed a similar trend in B-cells from individuals with LTBI compared to healthy controls and patients after TB treatment ( Fig 5A ) , implying a functional deficit in the B-cells of patients with active TB or LTBI . In addition to proliferation , also intracellular IL-6 was found to be decreased in B-cells from active TB and similar patterns observed in LTBI ( Fig 5B ) . Intracellular IL-10 showed a similar trend ( Fig 5C ) . Finally , as the hallmark of classical B-cell function , immunoglobulin ( Ig ) production was assessed . A reduced total Ig production by B-cells from LTBI subjects compared to controls was found ( Fig 5D ) . Analysis of antibody isotypes revealed the strongest decrease in IgA and IgM levels compared to IgE or IgG ( Fig 5E ) . Moreover , the B-cells from active TB patients expressed less HLA-DR , an activation marker for B-cells , following stimulation , compared to B-cells from the other groups ( Fig 5F ) . Together these and the above results indicate that circulating B-cells from patients with active TB disease and LTBI -but not from treated TB patients- have an atypical phenotype and are functionally impaired . As T- and B-cells closely interact during the adaptive immune response we next investigated T-cell activation across the respective stages of TB infection/ disease . To this end , PBMCs were stimulated with live M . bovis BCG for 6 days and cytokine production as well as CD4+ and CD8+ T-cell subset involvement were assessed . We have previously shown that 6 day live BCG stimulation results in activation of mycobacterium-specific T-cells with regulatory capacity ( Treg ) and therefore assessed Treg frequencies here as well [28 , 29] . Secretion of Th1 ( IFNγ ) or Th2 ( IL-13 ) cytokines in supernatants of BCG stimulated PBMCs was observed in all groups infected with Mtb , with the highest levels detected in individuals successfully treated for TB ( Fig 6A ) . Secretion of IL-10 was also observed in all groups , albeit at low levels , with the highest levels again in treated TB patients . Analysis of T-cell phenotypes focussed on the identification of regulatory T-cells ( Tregs ) and T-cells expressing inhibitory receptors associated with exhaustion , in order to assess the possible association between ‘exhausted’ atypical B-cells and exhausted T-cells ( S4 Fig ) . CD4+ and CD8+ Treg marker expression , defined as CD4 or 8 , CD25 , FoxP3 and LAG3 , were assessed; CD4+ Tregs were induced in all TB groups whereas CD8+ Tregs were significantly increased in TB treated patients ( Fig 6B ) . Intriguingly , the population of CD4+ Tregs was most abundantly detected , and at a very high frequency , in patients previously treated for TB . Similarly , but less dramatically different , also the proportion of CD8+ Tregs was highest in previously treated TB patients , possibly as a result of the Mtb antigenic load which is released during following TB chemotherapy . T-cell exhaustion is typically defined by the expression of the inhibitory receptors PD1 and/or KLRG1 . We mostly observed PD1 expression on CD4+ T-cells across the different Mtb-exposed groups , predominantly in TB treated subjects . Among CD8+ T-cells , we observed PD1 expression in active TB . In contrast , KLRG1 expression was significantly detected on CD4+ T-cells of TB treated subjects and was not increased in LTBI individuals or TB patients compared to controls , whereas CD8+ T-cells showed no differential expression of KLRG1 among the analysed groups ( Fig 6C ) [30] . These data thus strongly suggest that T-cell exhaustion and B-cell exhaustion are independently regulated and do not necessarily occur in the same individuals at the same time . As B-cells are central players in the adaptive immune response , B-cell malfunction may contribute to the decreased T-cell responsiveness observed in subjects with LTBI or active TB compared to those successfully treated for TB . Therefore , T-cell responses were analysed in total PBMCs and compared to B-cell depleted PBMCs . Depletion of B-cells had the strongest effect on T-cell activation in patients previously treated for TB , whom had fully functional B-cells as shown above . In contrast , B-cell depletion had only a minor effect on cytokine responses in active TB patients and individuals with LTBI , both of whom had dysfunctional B-cells . Both Th1 and Th2 cytokine responses strongly decreased following B-cell depletion of PBMCs in treated TB patients ( Fig 7A ) . IL-10 production in supernatants was almost completely abrogated following B-cell depletion in all samples ( Fig 7A ) . Interestingly , B-cell depletion resulted in a significant decrease in the frequency of CD4+ Tregs compared to total PBMCs in treated TB subjects , but not in active TB patients or LTBI individuals ( Fig 7B ) . These data indicate that depletion of B-cells in TB patients does not affect T-cell activation , supporting again the poor functionality of B-cells during active Mtb replication . CD8+ Tregs increased following B-cell depletion in patients with active TB ( Fig 7B ) . Similarly , PD1 expressing CD4+ T-cells and to a lesser extent also KLRG1 expressing CD4+ T-cells decreased following B-cell depletion , to the largest extent in treated TB patients ( Fig 7B ) . Intriguingly , as also seen for CD8+ Tregs , CD8+ T-cells expressing PD1 increased significantly . Thus depletion of functional B-cells such as in treated TB subjects diminished BCG induced cytokine production and T-cell activation , possibly as a result of APC depletion . In contrast , depletion of functionally impaired B-cells in LTBI individuals or TB patients hardly affected T-cell responses , which implies that functional B-cells are a critical component determining the magnitude of the antimycobacterial-specific Th1 and Th2-cell response .
The functional significance of B-cells in the control of Mtb infection in humans has been under studied compared to the large number of studies assessing the contribution of T-cells . Literature on B-cells in TB has suggested that these cells , apart from the antibodies they produce , may be involved in disease pathogenesis but data are scarce and often contradictory . We therefore decided to analyse phenotype and function of B-cells during Mtb infection and disease; active pulmonary TB , successfully treated TB , recent LTBI and control uninfected individuals . We specifically analysed B-cell memory- and exhaustion-phenotypes , B-cell functions ( proliferation , cytokine and immunoglobulin production ) as well as their ability to induce T-cell activation . We demonstrate not only that circulating B-cell numbers are decreased in patients with active TB disease , but more importantly—for the first time to the best of our knowledge- that B-cells from individuals with active disease or recent LTBI are functionally impaired . We identified functionally impaired B-cells in ( recent ) LTBI and active TB patients , but not in treated TB subjects . B-cells thus seem impaired occurs during active Mtb replication , which likely happens also in recent LTBI , however the mechanism responsible for this remains unknown . Alternatively or additionally , inflammatory processes may be involved in down regulation of B-cell functions . Moreover , B-cells from LTBI and active TB patients had also impaired functional capacities following polyclonal activation , including impaired cell proliferation , impaired cytokine production as well as impaired immunoglobulin production . These impairments were more prominent in patients with active TB compared to LTBI individuals , and varied for the different parameters . In particular , individuals with LTBI showed considerable variation in their responses which may reflect the heterogenic “spectral” nature of this group , ranging from active infection to presumably sterile eradication of Mtb infection . The variation in responses observed in all groups is considerable , reflecting the variation in TB disease phenotype or even inter-individual variation . Our data thus highlight abnormal B-cell numbers and memory B-cell distribution in patients with active TB as well as a functional impairment of B-cells during active Mtb replication . Moreover , we show that B-cells are functionally involved in T-cell activation , including activation of regulatory T-cells , because depletion of B-cells significantly affected these cell populations particularly in treated TB patients . In TB disease , the role of immunoglobulins , the primary products of B-cells , is under debate: although antigen-specific immunoglobulins are highly produced by TB patients their contribution to bacterial and/or infection control is less clear [31] . Antibodies from patients with active TB disease have been reported to recognize a different repertoire of Mtb antigens than those from individuals with controlled latent TB , and antibody titers correlated with mycobacterial burden [32] . Here , we also detected antibodies against mycobacterial PPD in patients with active and treated TB disease but not in those with LTBI , antibodies against Ag85B were detected in all infected groups albeit at very low levels . Interestingly , antibody levels did not correlate with total B-cell frequencies within the same sample , patients with very low B-cell frequencies still had average levels of PPD reactive antibodies . Importantly , we did not observe differences in the antibody producing plasma cell frequencies between any of our groups ( Fig 2A ) . Plasma cells differentiate from activated memory B-cells ( S5A Fig ) , a population that is relatively upregulated in LTBI and patients with active TB , suggesting that Mtb does not alter the normal development of antibody producing cells and thus also allow the production of Mtb specific antibodies . We did not observe differences in antibody levels between patients with active TB disease and those successfully treated for TB , supporting that plasma cell differentiation and function of differentiated plasma cells is not affected in a similar matter as total B-cell function in those with active TB disease . The impact of antibodies in TB remains debated [32] , for a long time they have been considered inefficacious because Mtb resides within the macrophages . However , more recently it has been demonstrated that immunoglobulins can contribute to the elimination of intracellular pathogens and thereby reduce tissue damage ( reviewed in [7 , 32] ) . Potential mechanisms that have been implicated include opsonisation , enhanced phagolysosomal fusion mediated through FcγR signalling , and increased intracellular killing of Mtb following Ca2+ flux [32] . A recent paper demonstrated that IgG antibodies directed against the Mtb capsular polysaccharide arabinomannan contributed to enhanced phagocytosis and inhibition of intracellular Mtb growth [33] . Moreover , it has been shown that antibody binding to pathogens may also be detected intracellularly via a cytosolic Fc-receptor called TRIM21 [34 , 35] . Pathogen bound antibody triggering of TRIM21 resulted in immune activation and inflammatory signals [34] . TRIM21 expression was increased in the blood of active TB patients compared to healthy controls , and expression patterns of TRIM21 grouped with those of other Fc receptors [22 , 23] . The decreased frequencies of B-cells in the circulation of TB patients could have been the result of B-cell homing to the site of disease . Our immunohistochemical analyses showed an increased presence of B-cells in lung tissues from TB patients compared to the lungs of patients that died from causes different from pneumonia . These B-cells were associated to the periphery of the granuloma and to ectopic follicles close to the vessels as previously described and explained as cells needed to orchestrate the immune response against Mtb outside the granuloma [20] . However this B-cell increase in the lung appeared modest compared to the B-cell depletion observed in the circulation which appeared severe . Unfortunately , we were not able to analyse paired tissue and peripheral blood samples from the same patients to investigate a causal relation . We do appreciate that those unfortunate patients that succumbed from the active TB may not reflect the normal disease process as they represent the most severe end of the disease spectrum . Yet , analysis of this tissue material has provided us insight in the B-cell pattern during severe TB disease locally in the lung . In addition to their decreased frequencies , also B cell function is severely hampered in patients with active TB disease , with a similar trend in LTBI individuals . Patients that have successfully completed the 6 month treatment regimen for pulmonary TB ( studied time points ranged from 1–72 months post treatment completion ) have fully restored B-cell numbers and functions , comparable with healthy uninfected controls . This suggests that chemotherapeutic elimination or reduction of Mtb replication and metabolism , and consequent reductions in inflammation are related to the functional restoration of the B-cell compartment . Moreover , our data suggest that B-cells with phenotypical characteristics of exhaustion are not necessarily exhausted , but can recover , or at least repopulate , the circulation with normally functioning B-cells . This is in line with what has been reported previously for HIV-infected patients in which treatment with anti-retroviral therapy also resulted in abrogation of FcRL4 expressing atypical B-cells [5 , 36] , as well as for individuals infected with Schistosomes in which the frequency of atypical B-cells was increased during infection and decreased to normal levels following treatment [37] . The B-cell unresponsiveness that we observe here may not only impact immunity towards Mtb itself but also other unrelated antigens and pathogens , as we find a strong and generalized impairment in the functional properties of B-cells following polyclonal stimulation . Individuals with LTBI in our study were recent contacts from active TB patients , although their previous mycobacterial exposure is unknown . It may well be that LTBI individuals exposed and infected more remotely in time are less affected and have normo-functional B-cells . There are no previous reports as far as we know of LTBI having poor antibody responses following vaccination or infection with other pathogens . This would agree well with our hypothesis that B-cell impairment by Mtb is transient and associated with active bacterial replication such as early following infection and during TB disease or with active ongoing inflammation . However , due to the cross-sectional design of our study we cannot exclude that intrinsic defects in B-cells are responsible for the development of TB disease in patients , nor can we formally prove that the B-cell defect is acquired as a result of Mtb infection . Individuals treated for TB disease had the highest T cell cytokine production as well as the highest frequency of T-cells expressing Treg markers , indicating a high level of T-cell activation in these subjects . The majority of TB treated patients analysed in this study had completed their treatment relatively recently such that memory T-cells should still be abundantly present after the likely massive Mtb antigen release induced by antibiotic treatment . However , only T-cell responses in treated TB patients diminished following B-cell depletion , suggesting that only normally functioning but not atypical-/exhausted B-cells contributed to the augmentation of T-cell activity . In support of this notion , depletion of B-cells which appeared largely exhausted from samples of TB patients or individuals with LTBI indeed did not augment ( nor diminish ) T-cell responses to BCG stimulation . In vivo in mice , B-cell deficiency did not result in hampered T-cell cytokine responses [14 , 15] however , the protracted time scale of the in vivo measurements is completely different from the in vitro analyses performed here . Recently , B-cells were depleted in an acute Mtb infection model in Rhesus Macaques and although the pathology and clinical outcome were not different , local analysis in the tissue , and in particular in the granuloma’s revealed differences in local T-cell responses and cytokine production [18] . Variation between individual granulomas was observed within the same animals , but there was a clear indication that B-cell depletion affected local inflammation . In addition , in vivo tissue resident professional APCs such as dendritic cells and macrophages may compensate for the deficiency of B-cells as APCs in those models . To the best of our knowledge , this is the first report on exhausted B-cells induced by a bacterial infection . Intriguingly and in contrast to our data described here , children chronically exposed to a parasite as the Plasmodium falciparum , had both atypical B-cells and exhausted T-cells [38] . In our study reported here , exhausted T-cells were most predominant in treated TB patients , whereas atypical B-cells were most prominent during active disease , suggesting that their induction is independently regulated . Perhaps systemic antigen exposure is already increased during blood stage malaria disease , with continued cycles of parasite release , as compared to TB , where massive antigen release occurs mostly after initiation of treatment , and therefore differences in kinetics may occur . A possible limitation of the data we present here is the use of cross-sectional groups of TB patients and treated TB patients . A longitudinal follow up of individual patients would be highly valuable to better understand the kinetics of B-cell and T-cell subset populations . Similarly , longitudinal follow up of individuals with recent LTBI will also be valuable to determine the duration of B cell impairment . Moreover , matching of in particular TB patients and controls was suboptimal in terms of sex and ethnicity . In conclusion , we have shown here that B-cells from patients with active TB and with recent LTBI , are decreased in numbers , have a phenotype of atypical B cells and are functionally impaired . TB treatment restores B cells in terms of key phenotypic and functional features—Moreover in the group of TB-treated patients highest levels of T-cell activation , including cytokine production , and expression of Treg markers and inhibitory receptors was observed . Mtb specific antibodies can be detected to a similar extent in active TB and treated TB patients . Moreover , B-cells appear central in antigen specific T-cell activation since B-cell depletion dramatically abrogated the potent T-cell activation in treated individuals . Together these data demonstrate that B-cell function is impaired during active TB disease , and that this has important consequences for T-cell activation , which may affect the overall host immunity towards TB .
This is an experimental laboratory study performed with human peripheral blood samples . This study was designed to describe B-cell numbers , phenotype and function during Mtb infection and TB disease before and after specific treatment . Well-characterized patients were recruited at the National Institute of Infectious Diseases in Rome , Italy ( Table 1 ) . Patients received standard care and blood samples were collected for research purposes . The study was approved by the local ethical committee and all subjects provided informed consent . The number of individuals included is indicated in the tables and figure legends . All clinical samples were anonymized by laboratory codes and tested and analysed blinded to their clinical status in all assays . Once the experimental results were completely obtained the codes were revealed to know the corresponding clinical status . This study was approved by the Ethical Committee of the L . Spallanzani National Institute of Infectious diseases ( INMI ) , approval number 02/2007 and 72/2015 . Informed written consent was required to participate in the study and was obtained before collecting blood samples . Collection and use of histological samples was approved by the ethical committee of INMI , approval number 72/2015 . Informed consent was not required for use of histology samples obtained at autopsy . Active pulmonary TB was sputum culture-confirmed and patients were enrolled within 7 days of starting the specific treatment . Treated TB subjects were patients who had completed a 6-month course treatment for culture-positive pulmonary TB and who were culture-negative after 2 and 6 months of therapy . An additional group of patients was evaluated after therapy completion ( 1–72 months after end of therapy ) . LTBI was defined based on positive response to Quantiferon ( QFT-IT ) in a healthy subjects without radiological signs of active disease [39 , 40] . LTBI subjects were mainly contacts recently exposed ( in the previous 6 months ) to smear-positive active pulmonary TB patients ( 15/22 ) , however infection may have occurred also at earlier time points as no information ( QFT-IT or TST ) is available on any prior time points . Healthy , uninfected controls were QFT-IT negative individuals . All enrolled subjects tested negative for HIV and were not undergoing treatment with immunosuppressive drugs . Demographic and epidemiological information were collected at enrolment ( Table 1 ) . The control group was complemented with 10 anonymous , Dutch , healthy adult blood bank donors ( Sanquin blood bank , Leiden ) that tested negative for in vitro recognition of mycobacterial PPD ( purified protein derivative ) . PPD-reactivity was tested by stimulation of PBMCs with 5 μg/ml PPD ( Statens Serum Institute , Copenhagen , Denmark ) for 6 days and supernatants were tested in an Interferon ( IFN ) γ-ELISA ( U-CyTech , Utrecht , The Netherlands ) . PPD positive responses were defined as IFNγ-production > 150 pg/ml . PBMCs were isolated by Ficoll density centrifugation and were stored and shipped in liquid nitrogen until use . Collection and use of histological samples was approved by the ethical committee of INMI , approval number 72/2015 . Tissues were obtained during the autopsies from 10 subjects with pulmonary tuberculosis ( sputum or broncho-alveolar-lavage positive for Mtb-culture or PCR ) , 10 subjects with pneumonia other than active TB and 5 patients died from causes other than pneumonia , such as liver cirrhosis or heart failure . Demographic , clinical and pathological features of all cases were collected and are summarized in Table 2 . Plasma samples from active and treated TB patients , latently infected individuals and controls were tested for the presence of Mtb specific antibodies . Flat bottom , 96-well Microlon plates ( Greiner ) were coated with 5 μg/ml PPD ( Statens Serum Institute , Copenhagen , Denmark ) , 5 μg/ml Antigen 85B or 5 μg/ml ESAT-6/CFP-10 fusion protein [41] in PBS or 2 hours at 37°C . Plates were blocked with PBS/ 1% BSA/ 1% Tween 20 for one hour before incubation with 6 serial dilutions of plasma ( 1:25–1:800 ) , diluted in PBS/ 1% BSA/ 0 . 05% Tween 20 and incubated at room temperature overnight . IgG antibody binding was detected using HRP labelled polyclonal rabbit anti-human IgG antibodies ( Dako , Glostrup , Denmark ) followed by TMB substrate buffer ( Sigma Aldrich ) , stopping the color reaction using H2SO4 and OD450 reading . The EC50 was determined using a non-linear curve fit and was defined as the midpoint of the linear portion of the dilution curve . For analysis of B-cell subsets PBMC were thawed and stained for 10 minutes at 4°C with the violet fixable live dead stain ( Vivid , LifeTechnologies Europe-Invitrogen , Bleiswijk , Netherlands ) according to the manufacturer’s protocol , immediately followed by a 30 minutes staining at 4°C for CD19 biotin ( clone HIB19; Biolegend , ITK diagnostics , Uithoorn , The Netherlands ) and Streptavidin Brilliant Violet510 ( BD Biosciences , Erembodegem , Belgium ) , CD25 APC-H7 ( clone M-A251; ) , CD27 APC ( clone L128 , ) , CD38 FITC ( clone HIT2; all BD Biosciences ) CD24 PE-Cy7 ( clone ML5; Biolegend ) , CD71 PE-Cy5 ( clone M-A712 , BD Biosciences ) , and CD1d PE ( clone 51 . 1; eBioscience , Vienna , Austria ) for Breg analysis . Plasma cells and memory B-cells were enumerated by staining with CD27 FITC ( clone M-T271 , BD Biosciences ) , FCRL4 PE ( clone 412D12 , Biolegend ) , CD10 PerCP-eFluor710 ( clone eBioSN5c , eBioscience ) , CD19 PE-Cy7 ( clone HIB19 , Biolegend ) , CD21 APC ( clone B-ly4 , BD Biosceinces ) , IgD biotin ( clone AI6-2 , BD Biosciences ) with Streptavidin Qdot525 ( LifeTechnologies Europe-Invitrogen ) or CD10 FITC ( clone HI10a ) , CD20 APC-H7 ( clone 2H7 ) , CD27 APC ( clone L128 ) , IgD HorizonV500 ( clone IA6-2 ) , CD22 PE-Cy7 ( clone HIB22 ) , HLA-DR PE-Cy5 ( clone G46-6 ) , CD21 PE-CF594 ( clone: B-ly4 ) , CD85j PE ( clone GHI/75 ) ( all BD Biosciences ) and FCRL4 PerCP-eFluor710 ( clone 413D12 ) in the presence of purified human FcR binding inhibitor CD16/CD32 ( both eBioscience ) . After staining cells were washed with PBS/BSA 0 . 1% ( Roche , Woerden , The Netherlands ) and fixed with 1% paraformaldehyde ( LUMC pharmacy , the Netherlands ) prior to analysis . B-cell proliferation was further assessed by flow cytometry . After a total of 6 days of culture , B-cells were harvested and the surface was stained with CD20 APC-H7 , CD22 PE-Cy7 , HLA-DR PE-Cy5 , CD21 PE-CF594 , CD85j PE ( all BD Biosciences ) and FCRL4 PerCP-efluor710 in the presence of purified human FcR binding inhibitor ( eBioscience ) . Cells were washed , fixed for 15 minutes using the Fixation buffer A ( ADG , ITK Diagnostics , Uithoorn , The Netherlands ) , washed once more and stained for intracellular IL-6 ( clone: MQ-13A5 , BD Biosciences ) and IL-10 ( clone: JES3-9D7 , Miltenyi Biotec BV , Leiden , The Netherlands ) in Permeablization buffer B ( ADG ) for 30 minutes at RT . Cells were washed for the last time and immediately acquired on a BD LSRFortessa . T-cells were analyzed for T-regulatory markers and for inhibitory receptor markers using the following fluorochrome panel after staining with the violet fixable live dead stain ( Vivid , Life technologies ) ; for surface staining , CD3 Brilliant Violet570 ( clone UCHT1 ) , KLRG1-Pe-Cy7 ( clone 2F1/KLRG1 ) ( both Biolegend ) , CD4 PE-CF594 ( clone S3 . 5; LifeTechnologies Europe-Invitrogen ) , CD8 HorizonV500 ( clone RPA-T8 ) , PD-1 PerCP-Cy5 . 5 ( clone EH12 . 1 ) ( both BD Biosciences ) were used . After 30 minutes cells were washed with PBS/BSA 0 . 1% , fixed with fixation buffer A for 15 minutes at RT , washed and stained for FoxP3-FITC ( clone PCH101; eBioscience ) , CD25 APC-H7 ( clone M-A251; BD Biosciences ) and LAG3-Atto647 ( clone 17B4; Enzo Life Sciences BVBA , Raamsdonksveer , the Netherlands ) , in permeabilization buffer B for an additional 30 minutes at RT . Finally , cells were washed once more and analyzed . All acquisitions were done on a BD FACS Canto or a BD LSRFortessa with FACS Diva software version 6 . 2 . Analysis was performed with FlowJo software version 9 . 7 . 6 . Gating strategies are provided in compliance with the MIATA guidelines in S2 and S3 Figs [42] . PBMCs were incubated with CD19 microbeads according to manufacturer’s protocol ( Milteny Biotec BV , Leiden , The Netherlands ) to isolate B-cells . After positive selection cells were checked for purity and in 76% of the cases the cells were enriched for more than 90% , while in the remaining 24% , the B-cells were between 80–90% purity . The CD19 depleted fraction was further used for the analysis of Bacillus Calmette Guerin ( BCG ) stimulated T-cell responses in the absence of B-cells and compared to total PBMC responses . B-cells were stained with 1μl per 10x106 cells/ml of the CellTrace violet cell proliferation kit ( Invitrogen ) and incubated for 20 minutes at room temperature . Reaction was stopped by adding Fetal Bovine Serum ( FBS ) rich medium according to manufacturer’s protocol . After labeling , the B-cells were plated in 96 wells round bottom plate at 3x105 cells/well and cultured in RPMI supplemented with 100 U/ml penicillin , 100 μg/ml streptomycin , 1mM pyruvaat and 2mM glutamate ( all Gibco Life technologies , Thermo Fisher Scientific Inc , Merelbeke , Belgium ) + 10% FBS ( Greiner Bio-one , Alphen a/d Rijn , The Netherlands ) with medium only or a polyclonal stimulation of αCD40+ anti IgG/M ( αCD40 at 1 μg/ml , Biolegend , ITK Diagnostics , Uithoorn , The Netherlands; 10 μg/ml AffiniPureF ( ab’ ) 2 Fragment Goat Anti-Human IgG + IgM ( H+L ) , Jackson ImmunoResearch Laboratories inc . , Suffolk , UK ) for 5 days . For the last 18 hours Brefeldin A ( 3 μg/ml , Sigma-Aldrich Chemie BV , Zwijndrecht , the Netherlands ) and Monensin ( 1/1000 , Biolegend ) were added . Supernatants of these cultures were harvested at day 2 and day 5 for subsequent determination of immunoglobulin levels . The BCG ( Pasteur strain ) was grown in Middlebrook 7H9 medium supplemented with 10% ADC ( BD Biosciences ) , log phase bacteria were used for infection experiments . Multiplicity of infection was calculated after determination of the number of viable bacilli per ml inoculum by plating serial dilutions of bacteria on Middlebrook 7H10 agar plates supplemented with 10% OADC ( BD Biosciences ) and counting of visible colonies after 3 weeks . Infections of PBMCs were done at a multiplicity of infection ( MOI ) of 3 . PBMCs ( 1x106 cells/well ) or the CD19-depleted PBMC fraction ( 1x106 cells/well ) were stimulated in a 48 wells plate with or without live BCG at a MOI of 3 for 6 days as previously described , resulting in amongst others functional Treg activation [29] . T-cells were cultured in IMDM supplemented with glutamax ( Gibco Life Technologies ) but without antibiotics . Supernatants were collected at day 2 and day 5 for cytokine analysis and cells were harvested at day 5 for flowcytometric analysis . Immunoglobulins were tested in an 1 to 5 dilution using the ProcartaPlex for human Isotyping ( Affymetrix- eBioscience , Vienna , Austria ) according to manufacturer’s instructions . T-cell supernatants were measured with a Bio-plex luminex kit for human IL-10 , IL-13 and IFN-γ ( Bio-rad , Veenendaal , The Netherlands ) . Luminex was performed according to manufacturer’s instructions and analyzed on a Bio-Plex 200 system with Bio-plex software ( Bio-rad ) . Specimens were fixed with 10% neutral phosphate-buffered formalin and paraffin-embedded . All blocks were stained with hematoxylin and eosin and processed in 4μm-thick slices . The slices were stained on Benchmark XT system ( Ventana , Tucson , AZ , USA ) at 37°C for 16 minutes . The following antibodies were used: CD3 rabbit monoclonal antibody ( mAb ) clone 2GV6 ( Roche ) ; CD14 rabbit mAb clone EPR3653 ( Roche ) ; CD20 polyclonal antibody ( pAb ) clone L26 ( Roche ) , Ki67 rabbit mAb clone 30–9 ( Roche ) . No double stainings were performed . Data obtained from individuals infected with Mtb , either LTBI , active TB disease or successfully treated for TB disease were compared to healthy controls using Mann-Whitney testing , with a p-value < 0 . 05 considered significant . However as 3 different groups were compared to the controls , multiple testing adjustments had to be made . Here we used the Kruskal-Wallis test with Dunn’s post testing and considered only those observations significant that had p < 0 . 05 in both the Kruskal-Wallis as well as the post test , these are indicated with a * in addition to the p-value from the Mann-Whitney test . Moreover , the effect of B-cell depletion was assessed within each group in a paired fashion using the Wilcoxon-signed-rank test , considering p < 0 . 05 as significant . All statistical testing was performed in GraphPad Prism version 6 . 02 . | In infections with intracellular pathogens like Mycobacterium tuberculosis ( Mtb ) , B-cells have long been ignored as their primary product , immunoglobulins , are unlikely to recognize intracellular bacteria . However , we have analysed here the frequency , phenotype and function of B-cells in tuberculosis ( TB ) infection and disease . Our data revealed that during active TB disease B-cell numbers are decreased and remaining B-cells are functionally impaired . Surprisingly , also individuals recently infected with Mtb suffered from poorly functional B-cells , but patients cured from the disease recovered with normal B-cell numbers and function . Thus , B-cell dysfunction contributes to impaired immune activation during Mtb infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"immunology",
"tropical",
"diseases",
"cloning",
"bacterial",
"diseases",
"developmental",
"biology",
"molecular",
"development",
"molecular",
"biology",
"techniques",
"infectious",
"disease",
"control",
"antibodies",
"bacteria",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"tuberculosis",
"proteins",
"t",
"cells",
"actinobacteria",
"molecular",
"biology",
"immune",
"system",
"antibody-producing",
"cells",
"biochemistry",
"cell",
"biology",
"b",
"cells",
"mycobacterium",
"tuberculosis",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] | 2016 | Patients with Tuberculosis Have a Dysfunctional Circulating B-Cell Compartment, Which Normalizes following Successful Treatment |
To date , most genetic analyses of phenotypes have focused on analyzing single traits or analyzing each phenotype independently . However , joint epistasis analysis of multiple complementary traits will increase statistical power and improve our understanding of the complicated genetic structure of the complex diseases . Despite their importance in uncovering the genetic structure of complex traits , the statistical methods for identifying epistasis in multiple phenotypes remains fundamentally unexplored . To fill this gap , we formulate a test for interaction between two genes in multiple quantitative trait analysis as a multiple functional regression ( MFRG ) in which the genotype functions ( genetic variant profiles ) are defined as a function of the genomic position of the genetic variants . We use large-scale simulations to calculate Type I error rates for testing interaction between two genes with multiple phenotypes and to compare the power with multivariate pairwise interaction analysis and single trait interaction analysis by a single variate functional regression model . To further evaluate performance , the MFRG for epistasis analysis is applied to five phenotypes of exome sequence data from the NHLBI’s Exome Sequencing Project ( ESP ) to detect pleiotropic epistasis . A total of 267 pairs of genes that formed a genetic interaction network showed significant evidence of epistasis influencing five traits . The results demonstrate that the joint interaction analysis of multiple phenotypes has a much higher power to detect interaction than the interaction analysis of a single trait and may open a new direction to fully uncovering the genetic structure of multiple phenotypes .
In the past several years , we have witnessed remarkable progresses in the development of methodologies for identification of epistasis that detect deviation from summation of genetic additive effects for a quantitative trait [1] . The classical approach to epistasis analysis is a single variant test . The epistasis is typically evaluated by testing interaction between a pair of variants one at a time . The classical methods for epistasis tests are originally designed to detect epistasis for common variants and are difficult applied to rare variants due to multiple testing problems and the low power to detect interaction . To overcome the critical barrier in interaction analysis for rare variants , instead of testing each pair of variants individually , group interaction tests that evaluate cumulative interaction effects of multiple genetic variants in a region or gene have recently been developed . Regression-based methods [2–8] , haplotype-based methods [9–15] , and machine learning-based methods [16–20] are proposed for epistasis analysis . The classical statistical methods for interaction analysis have mainly tested association with single traits , one time analyzing one trait [21] . However , multiple phenotypes are highly correlated . More than 4 . 6% of the SNPs and 16 . 9% of the genes in previous genome-wide association studies ( GWAS ) are reported to be significantly associated with more than one trait [22] . These results demonstrate that genetic pleiotropic effects likely play a crucial role in the molecular basis of correlated phenotypes [23–26] . Joint epistasis analysis of multiple complementary traits will increase statistical power to unravel the interaction structure of multiple phenotypes [27 , 28] . Despite their importance in understanding genetic mechanism underlying the complex diseases , the statistical methods for identifying epistasis in multiple phenotypes have been less developed [1] . The interaction analyses for multiple phenotypes have been limited to common variants in carefully controlled experimental crosses [29 , 30] . Simultaneously analyzing interactions for multiple phenotypes in humans poses enormous challenges for methodologies and computations . Purpose of this paper is to develop a general analytic framework and novel statistical methods for simultaneous epistasis analysis of multiple correlated phenotypes . To unify the approach to epistasis analysis for both common and rare variants , we take a genome region ( or gene ) as a basic unit of interaction analysis and use all the information that can be accessed to collectively test interaction between all possible pairs of SNPs within two genome regions ( or genes ) . Functional data analysis is used to reduce the dimension of next-generation sequencing data . Specifically , genetic variant profiles that will recognize information contained in the physical location of the SNP are used as a major data form . The densely typed genetic variants in a genomic region for each individual are so close that these genetic variant profiles can be treated as observed data taken from curves [8 , 31] . Since standard multivariate statistical analyses often fail with functional data [32] we formulate a test for interaction between two genomic regions in multiple quantitative trait analysis as a multiple functional regression ( MFRG ) model [33] with scalar response . In the MFRG model the genotype functions ( genetic variant profiles ) are defined as a function of the genomic position of the genetic variants rather than a set of discrete genotype values and the quantitative trait is predicted by genotype functions with their interaction terms . By functional principal component analysis , the genotype functions are expanded as a few functional principal components ( FPC ) and the MFRG model is transformed to the classical multivariate regression model ( MRG ) in which FPC scores are taken as variates . Statistics are developed in this publication which can be applied to pairwise interaction tests and gene-based interaction tests for multiple phenotypes . By investigating SNP-SNP interactions or gene-gene interactions that are shared across multiple traits , pleiotropic epistasis can be studied . To evaluate performance for multiple traits epistasis analysis , large scale simulations are used to calculate the Type I error rates of the MFRG for testing interaction between two genomic regions with multiple phenotypes and to compare power with multivariate pair-wise interaction analysis and single trait interaction analysis by functional regression ( FRG ) model . To further evaluate performance , the MFRG for epistasis analysis is applied to five traits: high density lipoprotein ( HDL ) , low density lipoprotein ( LDL ) , total cholesterol , systolic blood pressure ( SBP ) , and diastolic blood pressure ( DBP ) , from exome sequence data from the NHLBI’s Exome Sequencing Project ( ESP ) to detect pleiotropic epistasis .
We assume that both phenotypes and genotype profiles are centered . The genotype profiles xi ( t ) and xi ( s ) are expanded in terms of the orthonormal basis function as: xi ( t ) =∑j=1∞ξijϕj ( t ) and xi ( s ) =∑l=1∞ηilψl ( s ) , ( 3 ) where ϕj ( t ) and ψl ( s ) are sequences of the orthonormal basis functions . The more number of variants in the genes the more accurate the eigenfunction expansion . If the number of variants is less than 3 the eigenfunction expansion of the genotypic profiles is impossible . MFRG can only be used for gene with more than 3 variants . In practice , numerical methods for the integral will be used to calculate the expansion coefficients . Substituting Eq ( 3 ) into Eq ( 2 ) , we obtain ( Appendix ) yik=α0k+∑d=1Dνidτkd+∑j=1∞ξijαkj+∑l=1∞ηilβkl+∑j=1∞∑l=1∞ξijηilγkjl+εik , i=1 , … , n , k=1 , … , K , ( 4 ) The parameters αkj , βkl and γkjl are referred to as genetic additive and additive × additive effect scores for the k-th trait . These scores can also be viewed as the expansion coefficients of the genetic effect functions with respect to orthonormal basis functions . Then , Eq ( 4 ) can be approximated by ( Appendix ) Y=eα0+ντ+ξα+ηβ+Γγ+ε=WB+ε , ( 5 ) where W=[eνξηΓ] and B=[α0ταβγ] . Therefore , we transform the original functional regression interaction model into the classical multivariate regression interaction model by eigenfunction expansions . All methods for multivariate regression interaction analysis can directly be used for solving problem ( 5 ) . The standard least square estimators of B and the variance covariance matrix Σ are , respectively , given by B^= ( WTW ) −1WTY , ( 6 ) Σ^=1n ( Y−WB ) T ( Y−WB ) . ( 7 ) Denote the last JL row of the matrix ( WTW ) −1WT by A . Then , the estimator of the parameter γ is given by γ^=AY . ( 8 ) The vector of the matrix γ can be written as vec ( γ^ ) = ( I⊗A ) vec ( Y ) . ( 9 ) By the assumption of the variance matrix of Y , we obtain the variance matrix of vec ( Y ) : var ( vec ( Y ) ) =Σ⊗I . ( 10 ) Thus , it follows from Eqs ( 9 ) and ( 10 ) that Λ=var ( vec ( γ^ ) ) = ( I⊗A ) ( Σ⊗I ) ( I⊗AT ) =Σ⊗ ( AAT ) . ( 11 ) An essential problem in genetic interaction studies of the quantitative traits is to test the interaction between two genomic regions ( or genes ) . Formally , we investigate the problem of testing the following hypothesis: γk ( t , s ) =0 , ∀t∈[a1 , b1] , s∈[a2 , b2] , k=1 , … , K , which is equivalent to testing the hypothesis: H0:γ=0 . Define the test statistic for testing the interaction between two genomic regions [a1 , b1] and [a2 , b2] with K quantitative traits as TI= ( vec ( γ^ ) ) TΛ−1vec ( γ^ ) . ( 12 ) Then , under the null hypothesis H0: γ = 0 , TI is asymptotically distributed as a central χ ( KJL ) 2 distribution if JL components are taken in the expansion Eq ( 3 ) . Group tests often make implicit homogeneity assumptions where all putatively functional variants within the same genomic region are assumed to have the same direction of effects . However , in practice , the variants with opposite directions of effects will be simultaneously presented in the same genomic region . MFRG can efficiently use information of both risk and protective variants and allow for sign and size heterogeneity of genetic variants . In general , the trait increasing and decreasing variants will be present in different locations in the genomic region . Information of trait increasing and decreasing variants usually will be reflected in different eigenfunctions and hence will be included in different functional principal component scores . The MFRG test statistic is essentially to summarize the square of the functional principal component scores . Therefore , the opposite effects of trait increasing and decreasing variants on the phenotype will not compromise each other in the MFRG test statistics . The MFRG statistics automatically take the opposite effects of the trait increasing and decreasing variants on the phenotype into account and do not require additional computations . MFRG will take the sign and size heterogeneity of the variants into account and be less sensitive to the presence of variants with opposite directions of effect . We can also develop likelihood ratio-based statistics for testing interaction . Setting W=[W1W2] , we can write the model as E[Y]=W1[αβ]+W2γ . Under H0: γ = 0 , we have the model: Y=W1[αβ]+ε . The estimators will be [α^β^]= ( W1TW1 ) −1W1TYandΣ^1=1n ( Y−W1[α^β^] ) T ( Y−W1[α^β^] ) . The likelihood for the full model and reduced model are , respectively , given by L ( α^ , β^ , γ^ , Σ^ ) =e−nK/2 ( 2π ) nK/2|Σ^|n/2and L ( α^ , β^ , Σ^1 ) =e−nK/2 ( 2π ) nK/2|Σ^1|n/2 . The likelihood-ratio-based statistic for testing interaction between two genomic regions with multivariate traits is defined as TIΛ=−nlog ( |Σ^||Σ^1| ) . ( 13 ) Under the null hypothesis H0: γ = 0 , TIΛ is asymptotically distributed as a central χ ( KJL ) 2 distribution if JL components are taken in the expansion Eq ( 3 ) . The genetic models for simulations to calculate Type 1 error rates of the tests are briefly given below . We first assume the model with no marginal effects for all traits: Yi=μ+εi , i=1 , … , n , where Yi = [yi1 , … , yik] , μ = [μ1 , … , μk] , and εi is distributed as [ε1…εk]∼N ( [0…0] , ( 1⋯0 . 5⋮⋱⋮0 . 5⋯1 ) ) . Then , we considered the model with marginal genetic effect ( additive model ) at one gene: yik=μk+∑j=1Jxijαkj+εik , where xij={2 ( 1−Pj ) AjAj1−2PjAjaj−2Pjajaj , αk= ( rk−1 ) f0 , where Pj is a frequency of the allele Aj , rk is a risk parameter of the k-th trait which was randomly selected from 1 . 1 to 1 . 6 . The risk parameter affect the genetic effects and is used to control the contribution effort by genotype to the phenotype . The risk parameter influences the relative magnitude of the genetic effects . f0 is a baseline penetrance and set to 1 and ε are defined as before . Finally , we consider the model with marginal genetic effects ( additive model ) at both genes: yik=μk+∑j=1Jxijαkj+∑l=1Lzilβkl+εik , where xij={2 ( 1−Pj ) AjAj1−2PjAjaj−2Pjajaj , zil={2 ( 1−ql ) BlBl1−2qlBlbl−2qlblbl , αkj=αk= ( rpk−1 ) f0 , βkl=βk= ( rqk−1 ) f0 , Pj and ql are frequencies of the alleles Aj and Bl , respectively , rpk and rqk are risk parameters of the k-th trait for the SNPs in the first and second genes , respectively , and randomly selected from 1 . 1 to 1 . 6 , f0 is a baseline penetrance and set to 1 and ε are defined as before .
To examine the null distribution of test statistics , we performed a series of simulation studies to compare their empirical levels with the nominal ones . We calculated the Type I error rates for rare alleles , and common alleles . To make simulations more close to real whole exome sequencing data , we generated 50 , 000 datasets consisting of 1 , 000 , 000 chromosomes randomly sampled from the NHLBI’s Exome Sequencing Project ( ESP ) with 2 , 016 individuals and 18 , 587 genes . Each dataset included randomly selected a pair of genes from sequenced 18 , 587 genes . We randomly selected 20% of SNPs from each gene as causal variants . The number of sampled individuals from populations of 1 , 000 , 000 chromosomes ranged from 1 , 000 to 5 , 000 . For each dataset , we repeated 5 , 000 simulations . We presented average type I error rates over 50 , 000 randomly selected pairs of genes from whole exome sequencing ESP dataset . Table 1 and S1 and S2 Tables summarized the average Type I error rates of the test statistics for testing the interaction between two genes with no marginal effect and consisting of only rare variants with 5 traits , 2 traits and 10 traits , respectively , over 50 , 000 pairs of genes at the nominal levels α = 0 . 05 , α = 0 . 01 and α = 0 . 001 . Table 2 and S3 and S4 Tables summarized the average Type I error rates of the test statistics for testing the interaction between two genes with marginal effect at one gene consisting of only rare variants with 5 traits , 2 traits and 10 traits , respectively , over 50 , 000 pairs of genes at the nominal levels α = 0 . 05 , α = 0 . 01 and α = 0 . 001 . Table 3 and S5 and S6 Tables summarized the average Type I error rates of the test statistics for testing the interaction between two genes with marginal effect at both genes consisting of only rare variants with 5 traits , 2 traits and 10 traits , respectively , over 50 , 000 pairs of genes at the nominal levels α = 0 . 05 , α = 0 . 01 and α = 0 . 001 . For common variants , we summarized the average Type I error rates of the test statistics for testing the interaction between two genes with marginal effect at both genes consisting of only common variants with 5 traits , 2 and 10 traits , respectively , over 10 pairs of genes at the nominal levels α = 0 . 05 , α = 0 . 01 and α = 0 . 001 , in Table 4 and S7 and S8 Tables , respectively . The statistics for testing interaction between two genomic regions with only common variants have the similar Type 1 error rates in the other two scenarios: with marginal genetic effects at one gene or without marginal genetic effects at two genes . These results clearly showed that the Type I error rates of the MFRG-based test statistics for testing interaction between two genes with multiple traits and common variants with or without marginal effects were not appreciably different from the nominal α levels . For the rare variants when the sample sizes increased to 5 , 000 , the Type 1 error rates were still not appreciably different from the nominal levels . To evaluate the performance of the MFRG models for interaction analysis of multiple traits , we used simulated data to estimate their power to detect interaction between two genes for two , four , five , six and ten quantitative traits . A true multiple quantitative genetic model is given as follows . Consider H pairs of quantitative trait loci ( QTL ) from two genes ( genomic regions ) . Let Qh1 and qh1 be two alleles at the first QTL , and Qh2 and qh2 be two alleles at the second QTL , for the H pair of QTLs . Let uijkl be the genotypes of the u-th individual with ij=Qh1Qh1 , Qh1qh1 , qh1qh1 and kl=Qh2Qh2 , Qh2qh2 , qh2qh2 , and gmuijkl be its genotypic value for the m-th trait . The following multiple regression is used as a genetic model for the m-th quantitative trait: ymu=∑h=1Hgmuijklh+εmu , u=1 , 2 , … , n , m=1 , … , M , where gmuijklh is a genotypic value of the h-th pair of QTLs for the m-th quantitative trait and εmu are distributed as [ε1…εm]∼N ( [0…0] , ( 1⋯0 . 5⋮⋱⋮0 . 5⋯1 ) ) . Four models of interactions are considered: ( 1 ) Dominant OR Dominant , ( 2 ) Dominant AND Dominant , ( 3 ) Recessive OR Recessive and ( 4 ) Threshold model ( S9 Table ) . We assume that the genotypes at two loci affect a complex trait . Intuitively , Dominant OR Dominant model means that presence of risk allele at least one locus will cause the phenotype variation . Dominant AND Dominant model means that only when risk alleles at both loci are present the phenotype variation can be affected . Recessive OR recessive model indicates that when both risk alleles are at least present at one locus the phenotype variation can be observed . Threshold model implies that when two risk alleles at one locus and at least one risk allele at another locus are present , the phenotype variation will be observed . Recessive AND Recessive model is excluded due to low frequency of that condition with rare variants . The risk parameter r varies from 0 to 1 . We generated 2 , 000 , 000 chromosomes by resampling from 2 , 016 individuals of European origin with variants in random two genes selected from the NHLBI’s Exome Sequencing Project ( ESP ) . Two haplotypes were randomly sampled from the population and assigned to an individual . We randomly selected 20% of the variants as causal variants . A total of 2 , 000 individuals for the four interaction models were sampled from the populations . A total of 1 , 000 simulations were repeated for the power calculation . The power of the proposed MFRG model is compared with the single trait functional regression ( SFRG ) model , the multi-trait pair-wise interaction test and the regression on principal components ( PCs ) . For SNPs genotypes in each genomic region principal component analysis ( PCA ) were performed . The number of principal components for each individual which can explain 80% of the total genetic variation in the genomic region will be selected as the variables . Specifically , the principal component score of the i-th individual in the first and second genomic regions are denoted by xi1 , … , xik1 and zi1 , …zik2 , respectively . The regression model for detection of interaction for the m-th trait is then given by ymi=μm+∑j=1k1xijαmj+∑l=1k2zilβml+∑j=1k1∑l=1k2xijzilγmjl+εmi . The power of the MFRG is compared with the traditional point-wise interaction test which takes the following model: ymi=μm+xi1αm1+xi2αm2+xi1xi2γm+εmi , i=1 , … , n , m=1 , … , M . For a pair of genes , we assume that the first gene has k1 SNPs , and the second gene has k2 SNPs , then , the total number of all possible pairs is k = k1 × k2 . For each pair of SNPs , we calculated a statistic for testing pair-wise interaction Tmjpair . Finally , the maximum of Tmjpair: Tmax = max ( T1 , 1pair , T1 , 2pair , … , T1 , kpair , … , TM , 1pair , … , TM , kpair ) is computed . Figs 1 and 2 , S1 Fig and S2 Fig plotted the power curves of the two-trait FRG , single trait FRG , two-trait regression on PCs and two-trait pair-wise interaction tests for a quantitative trait under Dominant OR Dominant , Dominant AND Dominant , Threshold , and Recessive OR Recessive models , respectively . Only two genes include rare variants . These power curves are a function of the risk parameter at the significance level α = 0 . 05 . Permutations in the point-wise interaction tests were used to adjust for multiple testing . In all cases , the two-trait FRG had the highest power to detect epistasis . We observed two remarkable features . First , two-trait test had higher power than the one-trait test . Second , the two-trait FRG had the highest power among all two-trait tests . Figs 3 and 4 , S3 Fig and S4 Fig plotted the power curves of the two-trait FRG , single trait FRG , two-trait regression on PCs and two-trait pair-wise interaction tests for a quantitative trait under Dominant OR Dominant , Dominant AND Dominant , Threshold and Recessive OR Recessive models , respectively . Only two genes include common variants . These power curves are a function of the risk parameter at the significance level α = 0 . 05 . Permutations in the point-wise interaction tests were used to adjust for multiple testing . These figures showed that the power patterns of the epistasis tests for common variants were similar to that for rare variants . Next we investigate the impact of the number of traits on the power . Fig 5 plotted the power curves of two-trait FRG , four-trait FRG , five-trait FRG , six-trait FRG and ten-trait FRG under Dominant OR Dominant interaction model . Fig 5 showed that if the multiple phenotypes are correlated then the power of the MFRG to detect epistasis will increase as the number of phenotypes increases . To investigate the impact of sample size on the power , we plotted Fig 6 and S5–S7 Figs showing the power of three statistics for testing the interaction between two genomic regions ( or genes ) with only rare variants as a function of sample sizes under four interaction models , assuming 20% of the risk rare variants and the risk parameter r = 0 . 05 for Dominant OR Dominant , Dominant AND Dominant , and Recessive OR Recessive , and r = 0 . 5 for Threshold models , respectively . Again , we observed that the power of the two-trait FRG was the highest . To further evaluate the performance , the MFRG for testing epistasis was applied to data from the NHLBI’s ESP Project . Five phenotypes: HDL , LDL , total cholesterol , SBP and DBP were considered with a total of 2 , 016 individuals of European origin from 15 different cohorts in the ESP Project . No evidence of cohort- and/or phenotype-specific effects , or other systematic biases was found [34] . Exomes from related individuals were excluded from further analysis . We took the rank-based inverse normal transformation of the phenotypes [35] as trait values . The total number of genes tested for interactions which included both common and rare variants was 18 , 587 . The remaining annotated human genes which did not contain any SNPs in our dataset were excluded from the analysis . A P-value for declaring significant interaction after applying the Bonferroni correction for multiple tests was 2 . 89×10−10 . Population stratification may inflate the test statistics . To reduce the inflation , the standard strategy is to adjust for population stratification via principal components . All the tests were adjusted for sex , age and population stratification via 5 principal components . To examine the behavior of the MFRG , we plotted the QQ plot of the two-trait FRG test ( Fig 7 ) . The QQ plots showed that the false positive rate of the MFRG for detection of interaction in some degree is controlled . A total of 91 pairs of genes which were derived from 85 genes showed significant evidence of epistasis with P-values < 2 . 7×10−10 which were calculated using the MFRG model and simultaneously analyzing interaction of inverse normally transformed HDL and LDL ( S10 Table ) . The top 30 pairs of significantly interacted genes with HDL and LDL were listed in Table 5 . In Table 5 and S10 Table , P-values for testing interactions between genes by regression on PCA and the minimum of P-values for testing all possible pairs of SNPs between two genes using standard regression model simultaneously analyzed for the HDL and LDL and P-values for testing epistasis by the FRG separately against single trait HDL or LDL were also listed . Several remarkable features from these results were observed . First , we observed that although pairs of genes showed no strong evidence of interactions influencing individual trait HDL or LDL , they indeed demonstrated significant interactions if interactions were simultaneously analyzed for correlated HDL and LDL . Second , the MFRG often had a much smaller P-value to detect interaction than regression on the PCA and the minimum of P-values of pair-wise tests . Third , pairs of SNPs between two genes jointly have significant interaction effects , but individually each pair of SNPs make mild contributions to the interaction effects as shown in Table 6 . There were a total of 60 pairs of SNPs between genes CETP on chromosome 16 and GPR123 on chromosome 10 with P-values < 0 . 0488 . None of the 60 pairs of SNPs showed strong evidence of interaction . However , a number of pairs of SNPs between genes CETP and GPR123 collectively demonstrated significant interaction influencing the traits HDL and LDL . Fourth , 91 pairs of interacting genes formed a network ( Fig 8 ) . The genes C5orf64 that had interactions with 19 genes , CSMD1 that had interactions with 20 genes , were hub genes in the network . 26 genes out of total 85 genes in the network were mainly located in 18 pathways . Each of 12 pathways included at least two interacting genes . However , the majority of interacting genes are located in different pathways . Among 18 pathways , calcium signaling pathway mediates the effect of LDL and plays a role in control of atherosclerosis susceptibility [36] , LDL-cholesterol has multiple roles in regulating focal adhesion dynamics [37] , LDL is involved in free radical induced apoptosis pathway [38] , MAPK and JAK-STAT pathways are involved in dietary flavonoid protection against oxidized LDL [39] , up-regulation of autophagy via AMPK/mTOR signaling pathway alleviates oxidized -LDL induced inflammation [40] , PPARα holds a fundamental role in control of lipid homeostasis [41] and lectin-like ox-LDL receptor 1 mediates PKC-α/ERK/PPAR-γ/MMP pathway [42] , HDL reduces the TGF-β1-induced collagen deposition [43] , the Wnt pathway plays an important role in lipid storage and homeostasis [44] , From the literatures , we found that both common and rare variants in CETP were associated with the HDL [45] , CREBBP regulated LDL receptor transcription [46] , PLTP was associated with HDL and LDL [47] , TMEM57 was associated with serum lipid levels [48] , SH2B3 was associated with LDL cholesterol [49] . It was also reported that CSMD1 was associated with multivariate phenotype defined as low levels of low density lipoprotein cholesterol ( LDL-C < or = 100 mg/dl ) and high levels of triglycerides ( TG > or = 180 mg/dl ) [50] , associated with hypertension [51] . It was also reported that CSMD1 was associated with LDL and total cholesterol [52] . Next we analyzed five traits: HDL , LDL , SBP , DBP and TOTCHOL . Again , for each trait , inverse normal rank transformation was conducted to ensure that the normality assumption of the transformed trait variable was valid . To examine the behavior of the MFRG , we plotted QQ plot of the test ( S8 Fig ) . The QQ plots showed that the false positive rate of the MFRG for detection of interaction is controlled . A total of 267 pairs of genes which were derived from 160 genes showed significant evidence of epistasis influencing five traits with P-values < 1 . 96×10−10 which were calculated using the MFRG model ( S11 Table ) . Of them formed a largest connected subnetwork ( Fig 9 ) . The top 25 pairs of significantly interacted genes with five traits were listed in Table 7 . We observed the same pattern as was observed for the two traits: HDL and LDL . 46 genes out of 160 genes in the networks were mainly located in 42 pathways including 15 signaling pathways . Among them , 14 pathways were in Fig 8 . The interacting genes may be involved in the same biological pathway or in the different biological pathways . We observed 12 pathways , each of which contained at least two genes connected via interaction . However , the majority of interacting genes were not located in the same pathways . Again , we observed that pairs of SNPs between two genes jointly have significant interaction effects , but individually each pair of SNPs might make mild contributions to the interaction effects as shown in S12 Table . There were a total of 6 , 766 pairs of SNPs between genes CSMD1 and FOXO1 . S12 Table listed 101 pairs of SNPs with P-values < 0 . 049 . The majority of the 101 pairs of SNPs showed no strong evidence of interaction . However , they collectively demonstrated significant interaction influencing five traits . Among 42 pathways , in the previous sections we reported that 14 pathways were associated with HDL and LDL . From the literatures , we also know that unsaturated fatty acids stimulated the uptake of the LDL particles [53] , PPAR signaling pathway was correlated with blood pressure [54] , purine metabolism was associated with SBP [55] , Wnt signaling pathway mediated cholesterol transportation [56] , glycerolipid metabolism pathway was correlated with total cholesterol [57] , focal adhesion pathway was involved in lipid modulation [58] , Cell adhesion molecules was correlated with blood pressure [59] . We also observed from the literatures that a number of genes that appeared in the list of interacted genes with five traits had major genetic effects with single trait . Many reports showed that CETP , LIPC and LIPG were associated with HDL and LDL [60–62] and that MTHRR had known main effects for LDL [63] and blood pressure [64] , NR1I3 for lipid metabolism [65] , PLTP for LDL [66] , [67] , FOXO1 for LDL [68] and hypertension [69] , SMAD9 for hypertension [70] , and CSMD1 for SBP [51] .
Most genetic analyses of phenotypes have focused on analyzing single traits or , analyzing each phenotype independently . However , multiple phenotypes are highly correlated . Genetic variants can be associated with more than one trait . Genetic pleiotropic effects likely play a crucial role in the molecular basis of correlated phenotypes . To address these central themes and critical barriers in interaction analysis of multiple phenotypes , we shift the paradigm of interaction analysis from individual interaction analysis to pleiotropic interaction analysis and uncover the global organization of biological systems . MFRG was used to develop a novel statistical framework for joint interaction analysis of multiple correlated phenotypes . By large simulations and real data analysis the merits and limitations of the proposed new paradigm of joint interaction analysis of multiple phenotypes were demonstrated . The new approach fully uses all phenotype correlation information to jointly analyze interaction of multiple phenotypes . By large simulations and real data analysis , we showed that the proposed MFRG for joint interaction analysis of correlated multiple phenotypes substantially increased the power to detect interaction while keeping the Type 1 error rates of the test statistics under control . In real data analysis , we observed that although pairs of genes showed no strong evidence of interactions influencing individual trait , they indeed demonstrated significant interactions if interactions were simultaneously analyzed for correlated multiple traits . Due to lack of power of the widely used statistics for testing interaction between loci and its computational intensity , exploration of genome-wide gene-gene interaction has been limited . Few significant interaction results have been observed . Many geneticists question the universe presence of significant gene-gene interaction . Our analysis showed that although the number of significantly interacted genes for single phenotype was small , the number of significantly interacted genes for multiple phenotypes substantially increased . Our results suggested that joint interaction analysis of multiple phenotypes should be advocated in future genetic studies of complex traits . The interaction analysis for multiple phenotypes has been limited to common variants in carefully controlled experimental crosses and has mainly focused on the pair-wise interaction analysis . Although pair-wise interaction analysis is suitable for common variants , it is difficult to use to test interaction between rare and rare variants , and rare and common variants . There is an increasing need to develop statistics that can be used to test interactions among the entire allelic spectrum of variants for joint interaction analysis of multiple phenotypes . The MFRG utilizes the merits of taking genotype as functions and decomposes position varying genotype function into orthogonal eigenfunctions of genomic position . Only a few eigenfunctions that capture major information on genetic variation across the gene , are used to model the genetic variation . This substantially reduces the dimension in genetic variation of the data . The MFRG can efficiently test the interaction between rare and rare , rare and common , and common and common variants . In both real data analysis of two phenotypes and five phenotypes , the interacted genes formed interaction networks . Hub genes in the interaction networks were also observed . These hub genes usually play an important biological role in causing phenotype variation . An essential issue for interaction analysis of a large number of phenotypes is how to reduce dimension while fully exploiting complementary information in multiple phenotypes . The standard multivariate regression models for joint interaction analysis of multiple phenotypes do not explore the correlation structures of multiple phenotypes and reduce the dimensions of the phenotypes , and hence have limited power to detect pleotropic interaction effects due to large degrees of freedom . Data reduction techniques such as principal component analysis should be explored in the future interaction analysis of multiple phenotypes . The results in this paper are preliminary . The current marginal approaches for interaction analysis cannot distinguish between direct and indirect interactions , which will decrease our power to unravel mechanisms underlying complex traits . To overcome these limitations , causal inference tools should be explored for the joint interaction analysis of multiple phenotypes . The purpose of this paper is to stimulate further discussions regarding great challenges we are facing in the interaction analysis of high dimensional phenotypic and genomic data produced by modern sensors and next-generation sequencing . | The widely used statistical methods test interaction for single phenotype . However , we often observe pleotropic genetic interaction effects . The simultaneous gene-gene ( GxG ) interaction analysis of multiple complementary traits will increase statistical power to detect GxG interactions . Although GxG interactions play an important role in uncovering the genetic structure of complex traits , the statistical methods for detecting GxG interactions in multiple phenotypes remains less developed owing to its potential complexity . Therefore , we extend functional regression model from single variate to multivariate for simultaneous GxG interaction analysis of multiple correlated phenotypes . Large-scale simulations are conducted to evaluate Type I error rates for testing interaction between two genes with multiple phenotypes and to compare power with traditional multivariate pair-wise interaction analysis and single trait interaction analysis by a single variate functional regression model . To further evaluate performance , the MFRG for interaction analysis is applied to five phenotypes of exome sequence data from the NHLBI’s Exome Sequencing Project ( ESP ) to detect pleiotropic GxG interactions . 267 pairs of genes that formed a genetic interaction network showed significant evidence of interactions influencing five traits . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"genetic",
"networks",
"quantitative",
"trait",
"loci",
"variant",
"genotypes",
"quantitative",
"traits",
"genetic",
"mapping",
"epistasis",
"mathematics",
"statistics",
"(mathematics)",
"test",
"statistics",
"genome",
"analysis",
"network",
"analysis",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"mathematical",
"and",
"statistical",
"techniques",
"statistical",
"methods",
"genetic",
"loci",
"phenotypes",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"genomics",
"statistics",
"computational",
"biology"
] | 2016 | Functional Regression Models for Epistasis Analysis of Multiple Quantitative Traits |
P21 activated kinase ( PAK ) , PAK interacting exchange factor ( PIX ) , and G protein coupled receptor kinase interactor ( GIT ) compose a highly conserved signaling module controlling cell migrations , immune system signaling , and the formation of the mammalian nervous system . Traditionally , this signaling module is thought to facilitate the function of RAC and CDC-42 GTPases by allowing for the recruitment of a GTPase effector ( PAK ) , a GTPase activator ( PIX ) , and a scaffolding protein ( GIT ) as a regulated signaling unit to specific subcellular locations . Instead , we report here that this signaling module functions independently of RAC/CDC-42 GTPases in vivo to control the cell shape and migration of the distal tip cells ( DTCs ) during morphogenesis of the Caenorhabditis elegans gonad . In addition , this RAC/CDC-42–independent PAK pathway functions in parallel to a classical GTPase/PAK pathway to control the guidance aspect of DTC migration . Among the C . elegans PAKs , only PAK-1 functions in the GIT/PIX/PAK pathway independently of RAC/CDC42 GTPases , while both PAK-1 and MAX-2 are redundantly utilized in the GTPase/PAK pathway . Both RAC/CDC42–dependent and –independent PAK pathways function with the integrin receptors , suggesting that signaling through integrins can control the morphology , movement , and guidance of DTC through discrete pathways . Collectively , our results define a new signaling capacity for the GIT/PIX/PAK module that is likely to be conserved in vertebrates and demonstrate that PAK family members , which are redundantly utilized as GTPase effectors , can act non-redundantly in pathways independent of these GTPases .
The GIT/PIX/PAK signaling pathway is a highly conserved signaling module which controls cytoskeletal dynamics across metazoans . The functions of this signaling complex are diverse . In humans it controls the migrations of fibroblasts through modulation of adhesion complexes , and participates in T cell receptor signaling in the immune system . The GIT/PIX/PAK complex has also been shown to regulate neuronal plasticity and development in the nervous system [1]–[3] . The importance of this protein complex is further highlighted by the observation that in humans a loss of either PAK3 or αPIX leads to impaired function of the nervous system from nonsyndromic mental retardation [4] , [5] . To further understand how this complex functions in a well-defined in vivo system , we have isolated the C . elegans orthologs of the GIT/PIX/PAK complex and studied their roles in the migrations of the gonad distal tip cells ( DTCs ) . PAKs are downstream effectors of RAC and CDC-42 GTPases [6] . RAC and CDC-42 are RAS superfamily GTPases of the RHO subtype and are known to control cytoskeletal dynamics through their function as molecular switches [7] . In the canonical GTPase/PAK pathway , an activated RAC or CDC-42 GTPase binds to PAK and stimulates the activation of PAK's kinase activity . Despite the importance of the canonical GTPase/PAK pathways it has become increasingly clear that PAKs can also function in non-canonical pathways independent of GTPases [8] . While studies in vertebrates have indicated the likely existence of GTPase-independent PAK activation pathways the mechanistic details , biological relevance and prevalence of these pathways remain poorly understood . GIT and PIX have been shown to regulate cellular processes through PAKs in diverse model systems [2] , [3] , [9] . It is generally thought that GIT/PIX/PAK pathways utilize GTPases , as PIX contains a clear GEF ( guanine exchange factor ) domain for RAC and CDC-42 GTPases and all of these proteins control the same cellular processes . Recently two reports have indicated a possible GTPase-independent GIT/PIX/PAK signaling pathway is likely to exist . These studies found in vitro that PAK can be activated by PIX and GIT in the absence of a GTPase-PAK interaction . In the first of these studies it was shown that a guanine exchange factor ( GEF ) deficient PIX can activate PAK , while the second study demonstrated that the ARF GAP ( ADP-ribosylation factor GTPase activating ) domain of GIT can activate PAK [10] , [11] . These two studies suggested that the GIT/PIX/PAK complex can function independent of GTPases but the possible in vivo function of this pathway remains unclear . We find that in C . elegans the PAKs , RACs , CDC-42 , GIT and PIX are all involved in gonad morphogenesis . During gonad development the DTC functions as a leader cell to direct its elongation [12]–[14] . The movement of the DTC is controlled by guidance molecules [12] , [15] , [16] , as well as other factors that are associated with the formation and regulation of the extracellular matrix ( ECM ) [17]–[22] . However , little is known about the signaling pathways that transduce these environmental cues into directed cell movements . Here we define two distinct signaling pathways that control the guidance of the DTCs during gonad morphogenesis . One is a typical GTPase/PAK pathway that utilizes either PAK redundantly while the other is a GIT/PIX/PAK pathway that also controls the shape and migration of the DTCs . Remarkably we find that the highly conserved GIT/PIX/PAK complex is specific for one of the PAKs and functions in a novel RAC/CDC-42 independent manner during these processes .
While investigating the roles of the C . elegans PAKs we found that the two PAKs , pak-1 and max-2 are redundantly required for proper formation of the gonad . In wild type animals the two DTCs function as leader cells to guide the elongating gonads , which eventually form two bilaterally symmetric U shaped gonad arms ( Figure 1A ) . The elongation of gonad during morphogenesis occurs in three phases ( Figures 1A–C ) . In the first phase , the DTC leads the developing gonad away from a mid-body position along the ventral side of the animal . In the second phase the DTCs turn orthogonally and migrate towards the dorsal side of the animal . In the third and final phase , the two DTCs turn back and then migrate towards each other , reaching the vulva by the young adult stage . Throughout gonad elongation the DTCs exhibit a sharp tapering morphology such that they have a cone-like shape when viewed from the side ( Figures 1B–C ) . In order to understand the role of PAK signaling pathways in gonad morphogenesis , we first examined individual pak mutants . pak-1 mutants were found to display mild defects in DTC morphology and migration ( Figures 1D–E and 2B ) . In pak-1 mutants the DTCs generally lacked the sharp tapering morphology of wild type DTCs and instead had a bloated or distended structure ( morphology defect ) ( Figures 1D–E ) . The pak-1 mutant DTCs also often failed to migrate all the way to the vulva ( migration defect ) ( Figure 2B ) . max-2 mutants did not exhibit any of these defects . To reveal redundancy between these genes we examined PAK double mutants for gonad defects . The pak-1 and max-2 mutants used are putative null alleles [23] . The pak-1 ( ok448 ) allele has a deletion that removes most of the kinase domain , results in a frame shift and introduces an early stop codon . The max-2 allele nv162 has a deletion that removes the start codon , the first 4 exons and does not contain another in frame start codon until midway through the kinase coding sequence . The max-2 ( cy2 ) allele contains a missense mutation resulting in a glycine to glutamate substitution at a conserved residue in the kinase domain . The pak-1;max-2 double mutants exhibit even more severe defects than pak-1 single mutants . In addition to morphology defects , the DTCs in pak-1;max-2 double mutants wandered during their migrations ( guidance defect ) , failed to execute at least one of the turns and did not migrate completely to the vulva ( Figures 1H–M ) . These results demonstrate a role for the PAKs in regulating DTC morphology , migration and guidance during gonad morphogenesis , and suggest that the two PAKs are only partially redundant , such that there is a role for PAK-1 in regulating DTC morphology and migration that MAX-2 does not fulfill . PAKs are the best known RAC GTPase effectors . There are three rac genes in C . elegans: ced-10 , mig-2 and rac-2/3 [24] . The RACs themselves are required for DTC guidance , and they are partially redundant with each other for gonad development [24] , [25] . We therefore investigated whether the PAKs act with the RACs in DTC guidance . We made use of the following rac mutants: for mig-2 we utilized the putative null allele mig-2 ( mu28 ) . As CED-10 is required for embryogenesis , we utilized the ced-10 ( n1993 ) allele which is expected to be a strong loss of function . Because of the presence of the gene duplication in rac-2/3 we utilized RNAi for the rac-2/3 loss of function analysis . As previously reported we observed characteristic extra turns during the last phase of the DTC migrations resulting from a loss of function in any of the racs ( Figure 1N ) . We then examined double mutants of the two paks ( pak-1 and max-2 ) with the racs . Mutations in max-2 did not enhance the DTC guidance defects of any of the racs ( Figure 1N ) , indicating that MAX-2 works with the RACs in DTC guidance . In contrast , pak-1 mutants severely enhanced the DTC guidance defects of any of the rac mutants ( Figure 1N ) , indicating that PAK-1 acts at least partly in parallel to the RAC GTPases . To identify factors that may function with PAK-1 in the RAC-independent pathway , we examined genes that are known to interact with PAKs in other species . In this manner we identified orthologs of vertebrate PIX and GIT genes , which are referred to as pix-1 and git-1 respectively . PIX and GIT proteins are highly conserved among worms , flies , mice and humans ( Figure S1 ) . Utilizing putative promoter regions from the two genes to drive GFP expression in C . elegans we studied their expression patterns and found that both genes were expressed in the DTCs throughout the DTC migrations . ( Figure S1 ) . To begin to address the functions of pix-1 and git-1 in the DTCs we examined deletion mutants for gonad defects . The allele pix-1 ( gk416 ) has a deletion beginning 4 codons after the translational start , which removes the entire SH3 domain and is expected to result in a very early stop codon due to a frame shift . The nature of this deletion indicates that gk416 is a null allele . For git-1 we utilized git-1 ( tm1962 ) which contains a 484 bp genomic deletion ( in frame ) resulting in 133 amino acids of the protein being deleted including the second GIT domain . As this domain is required to bind PIX in fibroblasts [9] , any functional protein generated in the mutant is expected to be unable to bind PIX-1 . The tm1962 deletion likely results in a strong loss of function . Similar to pak-1 mutants , the pix-1 and git-1 mutants exhibited the characteristic defects in DTC migration and DTC morphology ( Figures 2A–D ) , but not the DTC guidance defects seen in the rac single mutants or in the pak-1;max-2 double mutants . We then tested whether pak-1 , pix-1 and git-1 function together in a pathway by examining all possible double mutant combinations . Double mutant combinations of pak-1 , pix-1 and git-1 did not enhance the DTC defects relative to the strongest single mutant ( Figure 2H ) . Nor was there any statistical difference in DTC defects between the triple mutant and any of the single mutants ( Figure 2H ) . Interestingly , the characteristic morphology defects found in the single , double or triple mutants of pak-1 , pix-1 or git-1 could be observed in actively migrating DTCs ( compare Figure 2I with 2J–L ) . Collectively these data indicate that GIT-1/PIX-1/PAK-1 signaling complex is required for the proper migration of the DTCs and the regulation of their cellular morphology and this pathway is not redundant with the classical RAC/PAK pathway . In contrast , MAX-2 functions in parallel to PIX-1 and GIT-1 to mediate DTC guidance . In addition to the morphology and migration defects of the GIT/PIX/PAK pathway , double mutants of max-2 with any of the genes from this pathway ( pak-1 , pix-1 or git-1 ) also showed major guidance defects in all stages of gonad elongation ( Figures 2E–H ) . These results indicate that PAK-1 , PIX-1 and GIT-1 function in a redundant DTC guidance pathway in parallel to MAX-2 . As MAX-2 works with the RAC GTPases this suggests that GIT-1 and PIX-1 may also function independent of the RACs . In support of this , pix-1 and git-1 mutants also profoundly enhance the guidance defects resulting from a loss of function in the rac genes ( Figure 3 ) . The gonadal defects seen in pix-1;rac and git-1;rac double mutants were similar to the pak-1;rac double mutants , which in turn were similar to the double mutants of the git-1/pix-1/pak-1 pathway with max-2 . In summary , our mutant analysis showed that any mutant in the GIT-1/PIX-1/PAK-1 pathway led to migration and morphology defects of the DTCs , while a loss of any of the racs ( which work in a pathway with max-2 ) led to guidance defects of the DTCs . Double mutants between these pathways led to severe defects in DTC guidance . Taken together , these results indicate that there are at least two distinct PAK pathways controlling DTC guidance during gonad morphogenesis: one is a classical RAC/PAK pathway , in which both MAX-2 and PAK-1 are utilized . The other is RAC-independent PAK pathway , in which PAK-1 ( but not MAX-2 ) , PIX-1 and GIT-1 are utilized and this latter pathway is used non-redundantly to regulate DTC migration and morphology . Since the GIT/PIX/PAK pathway functions independent of RAC GTPases , we next sought to explore whether the pathway functions independent of other GTPases . CDC-42 is also a RHO subfamily GTPase that has been shown to activate PAKs , and PIX is predicted to also be a GEF ( Guanine Exchange Factor ) for CDC-42 . If the GIT/PIX/PAK pathway does function independent of CDC-42 , knocking down CDC-42 would enhance the defects resulting from a loss of the GIT/PIX/PAK pathway . As CDC-42 is required for viability , we utilized tissue specific RNAi [26] in the post-embryonically born DTCs , to bypass the embryonic requirement for CDC-42 . Double RNAi of cdc-42 and pix-1 caused much more profound defects than RNAi of either of them alone . However , RNAi of max-2 did not enhance the defects caused by RNAi of cdc-42 ( Figure 4A ) . This data indicates that the GIT/PIX/PAK pathway may function independently of CDC-42 . Interestingly the lack of enhancement with the cdc-42; max-2 double RNAi may suggest that MAX-2 works with CDC-42 during gonad elongation . However these negative results are less than definitive as RNAi causes a partial loss of function and simply may not cause enough of a knock down to generate any possible enhancement of the cdc-42 phenotype . If PAK-1 functions independently of CDC-42 during DTC migrations , PAK-1 may not require conserved amino acids that allow it to bind GTPases . To test this we selectively altered PAK-1 at an amino acid in the GTPase binding domain ( pak-1 ( S76P ) ) that in other systems has been shown to be required for binding to CDC-42 and that is likely to disrupt binding to all GTPases [11] . We found that both wild type and the mutant PAK-1 partially rescued the pak-1 gonad morphology defects ( Figure 4B ) . This suggests that activation by CDC-42 is not necessary for the non redundant PAK-1 function in DTC morphology and migration . As an important control , we tested pak-1 ( S76P ) in the guidance of motor axons , where we showed previously that PAK-1's function is RAC dependent [23] . As expected , injecting pak-1 ( S76P ) failed to rescue the axon guidance defect in pak-1 mutant , while injecting wild-type pak-1 gene did ( Figure 4C ) . This latter result also indicates that the mutated PAK-1 loses its ability to interact with RAC GTPases . Collectively our results demonstrate that the GIT-1/PIX-1/PAK-1 pathway functions at least partly independent of RAC and CDC-42 GTPases . To gain insight into these distinct pathways controlling gonad morphogenesis , we used fluorophore tagged proteins to examine the subcellular localizations of the PAKs , PIX and GIT in migrating DTCs in vivo . These tagged proteins rescued the DTC defects when expressed in the DTCs of the respective mutants ( Figure S2 ) . We found that the tagged PAKs were diffusely present throughout the cytoplasm of the DTC during all stages of its migration ( Figure 5A–B ) suggesting that the PAKs function through a transient local activation mechanism . In contrast , both GIT-1::GFP and PIX-1::GFP localized to punctate structures in the DTC during its migrations ( Figure 5C–D ) . These puncta were observed throughout the cytoplasm of the migrating DTC . In addition , we found that GIT-1::GFP and PIX-1::mRFP co-localize throughout all of the phases of the migrating DTC ( Figure 5E–T ) . The extent of co-localization is nearly complete as there were few , if any , sites in the DTC where the RFP and GFP signals did not overlap ( Figure 5Q–T ) . These results indicate that C . elegans GIT-1 and PIX-1 are likely to interact directly , as has been repeatedly observed for their orthologs in a variety of different systems [1] , [2] , [27] . The punctate pattern is highly reminiscent of the localization of GIT/PIX in other systems where they have been characterized as forming large multimeric complexes that are thought to be scaffolds for intracellular signaling [28] . Collectively our results suggest that the GIT/PIX complex locally activates PAK-1 from a reservoir of cytoplasmically localized inactive PAK-1 . PAKs , PIX , GIT and RACs have all been implicated in integrin-regulated processes in other model systems [9] , [29] . To explore whether integrin signaling in the DTC is mediated by PAK signaling pathways , we first examined the phenotypes of integrin mutants by RNAi . Integrins function as heterodimers that consist of alpha and beta subunits . C . elegans genome contains two alpha ( ina-1 and pat-2 ) and a single beta ( pat-3 ) subunits . The integrins have previously been implicated in controlling DTC migration [15] , [30] , [31] . As all of the integrin genes are required for embryogenesis , we examined their function in DTCs with tissue specific RNAi . We found that a loss of function in any of the integrin genes led to similar defects as those we observed in the double PAK pathway mutants . The integrin mutants have both the severe migration and guidance defects of ( rac/max-2 ) ; ( pak-1/pix-1/git-1 ) double mutants as well as the bloated DTC morphology phenotype observed in mutants of the GIT-1/PIX-1/PAK-1 pathway ( Figures 6A–E ) . This was also observed in the two available ina-1 hypomorphs gm39 and gm144 ( data not shown ) . Unfortunately we were unable to generate double mutants of these hypomorphs with either of the paks , leading us to conclude that these double mutants may be unviable . Nevertheless , these data suggest that the integrins may function with both the GIT/PIX/PAK and the RAC/PAK signaling pathways . To further determine whether the PAK signaling pathways function with the integrins , we made use of a PAT-3 beta-integrin interfering construct ( beta tail ) previously reported to disrupt integrin signaling in the DTCs [30] . If the PAK pathways function with the integrins , a loss of either PAK pathway may not lead to an enhancement of the defects resulting from inhibiting normal integrin signaling . However , if either of the PAK pathways functions independently of the integrins , a loss of that pathway should enhance the defects caused by inhibiting normal integrin signaling . As reported , we found that tissue-specific expression of the beta tail caused low penetrance defects in gonad morphogenesis ( Figure 6F ) . When these transgenic lines were crossed into the triple pak-1;pix-1;git-1 mutants or were examined in a max-2 RNAi background there was no significant enhancement in the defects ( Figure 6F ) . As a control , we also tested whether the beta tail would enhance the defects of a mutant in a pathway that is expected to function independently of integrins . We utilized the UNC-6/UNC-40/UNC-5 pathway which specifically controls the dorsal migrations of the DTCs [12] , [13] . As has previously been reported , we found that loss of the unc-40 gene resulted in defects specifically in the dorso-ventral guidance of the DTCs ( Figure 6F ) . RNAi of unc-40 in the beta tail transgenic background resulted in additive enhancement of the DTC defects . These results collectively suggest that both the RAC/PAK and the GIT/PIX/PAK pathways function with the integrins to control DTC morphology , migration and guidance .
During gonad morphogenesis , the distal tip cell ( DTC ) leads the elongating gonad over a long distance to reach its final destination . Several guidance and motility systems are known to facilitate the elongation of the gonad [32] . For example , a protease system that rearranges the ECM allowing motility ( GON-1 ) and guidance ( MIG-17 ) of the DTC are required for proper gonad elongation . Another is the UNC-6/UNC-40/UNC-5 system which specifically directs the dorsal ( phase 2 ) turning of the gonad . Finally there is the integrin system , which controls multiple aspects of gonad elongation by coordinating the interactions between the ECM and the DTC [15] , [30] , [31] . We have extensively studied the signaling pathways inside the DTC that are regulated by PAKs during gonad morphogenesis , and have identified two distinct PAK signaling pathways that differentially control the morphology , migration and guidance of the DTC . Our analysis also suggests that these PAK pathways are regulated through integrin signaling during gonad elongation . The two PAK signaling pathways are a classical RAC dependent PAK pathway and a RAC/CDC-42 independent GIT/PIX/PAK pathway . Both pathways function in the guidance of the migrating DTC , but only the latter is required for maintaining the DTC morphology during DTC migrations ( Figure S3 ) . What are the roles of PAK-1 and MAX-2 in these two separate pathways ? Although our genetic analysis indicates that PAK-1 contributes significantly to the GIT/PIX/PAK signaling pathway , PAK-1 also likely functions in the RAC/CDC-42 dependent pathway . This conclusion comes from the observation that max-2 single mutants do not yield DTC guidance defects yet double pak-1;max-2 mutants have profound DTC guidance defects . Therefore a loss of max-2 is being compensated for by the presence of pak-1 . However , we also find that double mutants of max-2;pix-1 or max-2;git-1 are profoundly defective in guidance even though there is still a functional PAK-1 present . These results suggest that PAK-1 by itself cannot completely compensate for MAX-2 in DTCs . One possible explanation is that pak-1 is only partially redundant with max-2 , perhaps due to differential kinase specificity of MAX-2 and PAK-1 while acting as RAC effectors . An alternate interpretation is that the loss of a functional PAK-1/PIX-1/GIT-1 pathway sensitizes the system such that the entire RAC pathway must now remain intact . The latter is supported by our observation that the loss of any component of the PAK/PIX/GIT pathway causes major DTC guidance and migration defects when combined with the loss of any of the racs ( Figures 1N and 3 ) . That git-1and pix-1function together with pak-1 in a genetic pathway in C . elegans strongly supports the notion that these genes have a conserved function across phyla . In addition to our results these proteins have been implicated as working together to regulate cellular processes in diverse model systems . Using genetic analysis in C . elegans we demonstrate that this highly conserved GIT/PIX/PAK pathway can function independent of RAC and CDC-42 GTPases . Interestingly , only PAK-1 , but not MAX-2 , is required , indicating that PAKs are not redundant for this pathway , demonstrating PAK specificity in a RAC/CDC-42 independent pathway . We also attempted to address whether all GTPases are not required in the GIT/PIX/PAK pathway . We generated a mutated PAK-1 that specifically disrupts its P21 binding domain and does not bind to any GTPase , and have found that this mutated PAK-1 can still partially rescue the DTC phenotype in pak-1 mutants . Our results suggest that perhaps the GIT/PIX/PAK pathway is independent of all GTPases . In addition , our genetic and cell localization studies suggest a model where the GIT/PIX complex is selectively activating PAK-1 through a direct interaction . This conclusion is supported by previous studies in fibroblasts that GIT can activate PAK in the absence of GTPase binding [11] . Furthermore it was recently shown that in T cells a GIT/PIX/PAK pathway functions in parallel to a pathway utilizing VAV ( a RAC GEF ) along with RAC and PAK [33] . Together , these results suggest that the GTPase-independent GIT/PIX/PAK signaling pathway is a conserved signaling pathway utilized for multiple cellular processes . In addition to migration defects , the GIT/PIX/PAK pathway mutants exhibit abnormal DTC morphology . Both the migration and morphology phenotypes are consistent with a defect in adhesion to the ECM substrate or the failure to execute coordinated changes in the cytoskeleton . Failure to elongate the proper distance may indicate that the DTCs have difficulty in removing/recycling their contacts with the basal lamina , which could result in the DTCs stalling prior to their targeted final destination . The bloated cell morphology may also result from an adhesion defect . The mutant DTCs may not properly adhere to their substrate and therefore adopt a less organized morphology . Similar DTC phenotypes are also observed in integrin mutants . Regulation of integrin signaling has previously been attributed to the GIT/PIX/PAK pathway in migrating fibroblasts where they are involved dismantling the integrin associated adhesion complexes . Interestingly , orthologs of PAK-1 , PIX-1 , and GIT-1 are all known to be involved in turnover of focal adhesions [9] , [34] , and GIT has also been reported to cycle between several different locations including the focal adhesions and cytoplasmic structures [35] . Taken together , it is likely that the GIT/PIX/PAK pathway functions to control either the sorting or the stability of integrin based organization of the cytoskeleton of the migrating DTC . Our genetic analysis indicates that the two distinct PAK signaling pathways are functioning with the integrins during gonad morphogenesis . First , the overall integrin mutant phenotypes are similar to the combination of mutants from the GIT/PIX/PAK pathway and the RAC/PAK pathway . Second , an interfering construct that is reported to perturb integrin signaling and does cause a gonad phenotype does not significantly enhance the defects of mutants from either of the PAK signaling pathways . Collectively these data support the model that the PAK pathways are all functioning with the integrins . Unfortunately due to the lack of a viable null mutant in any of the integrin subunits our results are less than definitive and there are caveats to our conclusions . First , phenotypic similarity just suggests that they control the same process and does not necessitate that they function together to control that process . Second , the interfering construct causes only weak defects . Because of this we tested whether the construct could enhance an unrelated pathway ( UNC-6/UNC-5/UNC-40 ) and we found that it did enhance this pathway . This clear enhancement of an unrelated pathway strengthens the significance of the non-enhancement with the PAK pathways result and indicates that the interfering construct is likely to disrupt aspects of the integrin signaling pathways that are involved with the PAK signaling pathways . The simplest explanation of our results is therefore that the PAK pathways act with integrin signaling . It is well known that the RACs are highly redundant for many processes . In C . elegans the RACs are only partly redundant . The specific DTC guidance defects in single rac mutants ( an inappropriate reversal of direction in the final phase of migration ) indicate that RAC GTPases are required in a non redundant manner at a specific stage in DTC guidance . It was previously reported that the RACs act with each other to inhibit this extra turn [24] . Such a lack of redundancy in the RAC GTPases may result from RAC specificity at the level of the RACs activator's ( the GEFs ) or at the level of the RAC effectors . Our results here do not address the redundancy of the RAC GTPases , but they do indicate that any such effector specificity is not occurring through the PAKs ( PAK-1 and MAX-2 ) . Instead our results indicate that PAKs are always redundant as RAC GTPase effectors . That is to say either PAK can be activated by any of the RACs . This model predicts that in the case where the RACs are non-redundant either PAK can act with any RAC therefore the PAKs will still be redundant with each other . Similarly if the RACs act together to mediate a pathway the PAKs can both act at either and both steps of the pathway and will still be redundant with each other . Our conclusion that the two PAKs are completely redundant as RAC effectors comes from multiple lines of evidence . Previously we found that the both PAKs function completely with the RACs to mediate P cell migrations . That is they are completely redundant for this process . However in commissural motor neuron axon guidance max-2 has a phenotype alone while pak-1 does not , yet the double is extremely severe ( they were partly redundant ) [23] . Here we find that the converse relationship is true; pak-1 has a phenotype alone and the double is very severe . Collectively examining these situations we found that if the paks are completely redundant then the individual pak mutants do not enhance the individual rac mutants . If the paks are partly redundant then the PAK with the phenotype would enhance any of the racs while the other would not enhance any of them . Our model to account for this describes that the PAKs are redundant as RAC effectors but additional PAK activators exist that do not require RAC GTPases and they activate with specificity towards the PAKs . In P cell migrations there is no such activator , in axon guidance the activator is specific for MAX-2 and during gonad morphology the activator is likely the GIT/PIX complex and it is specific for PAK-1 . It is easy to speculate how such a phenomenon could arise evolutionarily . First redundancy at the level of the highly utilized RAC effector pathway would be favorable; after all the RACs themselves are highly redundant and are so in most organisms . This would favor a gene duplication of the PAKs . New roles could then evolve for the PAKs that do not come at a cost of the RAC effector pathway . This would add to the signaling capacity of a cell yet allow it to retain the improved capacity for RAC signaling arising from the gene duplication . Finally , it is worth noting that the movement of the DTC is distinctly different from the migration of many other migrating cells . Cell migrations are typically characterized by protrusion of filopodia and lamelopodia followed by invasion of the cytosol into these structures , steadily dragging the cell forward . In DTCs we do not observe front protrusion of membranous structures . Instead the migrating DTCs maintain an arrowhead shape during migration ( Figure 2 ) , suggesting that they are not moving through a normal fibroblast type mechanism . The DTCs while migrating are also capping a rapidly growing gonad and seem to be pushed from behind by the elongating gonad . Thus the movement of the DTCs is likely to be controlled by the directional secretion of the proteases [17] , [21] as well as the regulation of its contacts with the ECM . Our studies indicate that integrin signaling through a novel GIT/ PIX/PAK pathway is important for maintaining the structural integrity and regulating the ECM contacts . Further studies will be necessary to elucidate how these signaling pathways inside the DTC coordinate all these and other factors to properly direct its movement during gonad morphogenesis .
Worm cultures were maintained with standard methods [36] . All newly characterized mutants were backcrossed at least five times to wild type prior to analysis . Mutant genotypes were confirmed by PCR or direct sequencing of PCR products or by confirmation of a known phenotype . For RNAi experiments , dsRNA was microinjected into the gonad of young adult animals [37] . The following RNAi clones , Ahringer Library Clones [38] unless otherwise specified , were utilized in this study: max-2 ( II 8F19 ) , pak-1 ( C09B8 . 7 ( open biosystems ) ) , pix-1 ( made from the YK clone YK447g6 ) , ina-1 ( III 4N10 ) , pat-2 ( III 4P15 ) , pat-3 ( III 1P02 ) and rac-2/3 ( IV 7L24 ) . LG II: max-2 ( cy2 ) , max-2 ( nv162 ) ; LG IV: ced-10 ( n1993 ) , eri-1 ( mg366 ) ; LG V: rde-1 ( ne215 ) ; LG X: oxIs12[Punc-47::GFP , lin-15 ( + ) ] , pak-1 ( ok448 ) , pix-1 ( gk416 ) , git-1 ( tm1962 ) , mig-2 ( mu28 ) . To score distal tip cell ( DTC ) defects , we analyzed young adult hermaphrodites with completely formed vulvas that had yet to pass an oocyte through the spermatheca . For each animal , the anterior and posterior gonads were scored separately . A gonad was deemed to have a DTC defect if the DTC failed to make proper turns ( guidance defect ) , if the DTC failed to reach the vulva ( migration defect ) , or if the DTC had a bloated structure ( morphology defect ) . Specifically , a DTC was deemed to have a guidance defect if it lacked the characteristic U Shape . A DTC was scored as having a migration defect if the DTC was greater than 24 micrometers away from reaching the midline of the vulva . A DTC was deemed to have a morphology defect if the cells diameter ( as judged by the diameter of the distal most region of the gonad ) was greater than 24 micrometers . The 24 micrometer distance in migration and morphology was chosen as we found that greater than 99% of wild-type animals' DTCs ( n = 80 ) were within this range . For graphical representations these phenotypes were combined and displayed together as the percent of animals with abnormal gonads . The DD and VD commissural motor axon guidance defects were scored as previously described [23] . The allele pix-1 ( gk416 ) which was generated by the Vancouver branch of the C . elegans Gene Knockout Consortium has a deletion beginning 4 codons after the translational start , which removes the entire SH3 domain and is expected to result in a very early stop codon due to a frame shift . The allele can be followed by the primers 416 . f1 gagatacaccccgcaaaaga , 416 . f2 gggaaggaacacatgaagga ( internal to deletion ) and 416 . r1 gccgatccacgttgtaaatc . For git-1 we have utilized the tm1962 allele generated by Shohei Mitani . git-1 ( tm1962 ) contains a 484 bp genomic deletion ( in frame ) resulting in 133 amino acids of the protein being deleted including the second GIT domain . As this domain is required to bind PIX in fibroblasts [9] , any functional protein generated in the mutant is expected to be unable to bind PIX-1 . The allele can be followed by the primers 1962 . f1 ttctccgttgttttcccaag , 1962 . f2 gcaccagtatccgaaccacccaa ( internal to deletion ) and 1962 . r1 tagccaatggagatggcatc . For the tissue specific RNAi experiments we expressed an rde-1 ( cDNA ) in an rde-1 ( ne219 ) mutant [26] resulting in a transgenic line ( HJ229 ) that only has functional RNAi where rde-1 is expressed . To drive the expression of rde-1 we utilized the lag-2 promoter ( 5′ primer ctagacagtcagcggcccataag ) up to but not including the start codon and fused this to a rde-1::unc-54 3′UTR PCR fragment generated from the pKK1253 plasmid ( gift from Hiroshi Qadota ) . Cloning of DNA and generation of transgenes were accomplished by standard techniques . In particular we made extensive use of PCR based gene fusion and subsequent cloning of PCR products into TOPO vectors ( Invitrogen ) . The Ppak-1::max-2::venus construct was constructed by fusing the 5′ region of pHJ102 [23] to the 3′ region of the partial cDNA clone Y38F1A . 10::venus ( A gift from Queelim Ch'ng ) . The resulting construct contained a full length max-2 cDNA under its own promoter fused to YFP ( venus ) . We then fused the max-2 ( cDNA ) ::venus region to a pak-1 promoter [23] to generate Ppak-1::max-2::YFP . To generate Ppak-1::pak-1::mRFP we generated a Ppak-1::pak-1 ( cDNA ) minigene and fused it to the mRFP::UNC-54 ( 3′UTR ) from Punc-25::mRFP [39] ( A gift from Ken-Ichi Ogura ) . For the PIX-1 translational reporters , we utilized the partial cDNA yk447g6 and fused it to the 5′ pix-1 genomic region ending at the second exon ( 5′ primer: gccatggtagtaagagcattccg ) . This Ppix-1::pix-1 ( cDNA ) minigene was then fused to mRFP or GFP as described in the preceding and following text . To generate Pgit-1::git-1::GFP we utilized the yk1688c03 ( Yuji Kohara ) full length cDNA and fused it to its 5′ genomic region ( 5′ primer gggtgaacggtcacttgactaga ) generating a Pgit-1::git-1 ( cDNA ) minigene . This was then fused to the GFP::UNC-54 ( 3′ UTR ) from pPD95 . 75 ( Fire Vector Kit ) yielding Pgit-1::git-1::GFP . Lag-2 promoter regions used for DTC specific expression consisted of 2 , 790 bp of DNA 5′ to the ORF through the start codon ( 5′ primer acgtcttgtaaccccctcccacc ) . For microscopy animals were mounted on 2% agarose pads with 5 mM sodium azide . Animals were scored by examination with microscopy at 400× on a Zeiss Axioplan II . Confocal images were captured with a Zeiss ( Thornwood , NY ) LSM 510 META laser-scanning confocal microscope . Images were analyzed using Zeiss META software version 3 . 2 SPZ . | Cell migration is essential for the development and maintenance of metazoan tissue . A migrating cell must navigate through complex environments and properly interpret the signals present in its path . This cellular movement is accomplished through transduction of the signals into directed reorganization of the cellular structure . Among the most important molecules that orchestrate signals from the exterior of the cells into cellular movement are the small GTPases , which function in intracellular signal transduction cascades . We have studied the interactions between GTPases , their effectors , and the environmental signals during cellular migrations in C . elegans . We have found that while some GTPases do control the guidance of these migrating cells , a certain highly conserved complex of proteins thought to be involved in mediating GTPase signaling during cellular migrations in fact functions independently of these GTPases to specifically control the structure and movement of the migrating cells . These results have revealed an unexpected role of a well-known and highly conserved signaling complex , which is particularly important since members of this complex are associated with human mental retardation . Our results may imply that the disease phenotype is likely more complex than previously thought and may in fact occur from disruption of this novel pathway . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology/morphogenesis",
"and",
"cell",
"biology",
"cell",
"biology/cell",
"signaling",
"cell",
"biology/morphogenesis",
"and",
"cell",
"biology",
"cell",
"biology/developmental",
"molecular",
"mechanisms",
"developmental",
"biology/molecular",
"development",
"developmental",
"biology/organogenesis",
"developmental",
"biology/developmental",
"molecular",
"mechanisms"
] | 2008 | A RAC/CDC-42–Independent GIT/PIX/PAK Signaling Pathway Mediates Cell Migration in C. elegans |
Several recent studies have shown a genetic influence on gene expression variation , including variation between the two chromosomes within an individual and variation between individuals at the population level . We hypothesized that genetic inheritance may also affect variation in chromatin states . To test this hypothesis , we analyzed chromatin states in 12 lymphoblastoid cells derived from two Centre d'Etude du Polymorphisme Humain families using an allele-specific chromatin immunoprecipitation ( ChIP-on-chip ) assay with Affymetrix 10K SNP chip . We performed the allele-specific ChIP-on-chip assays for the 12 lymphoblastoid cells using antibodies targeting at RNA polymerase II and five post-translation modified forms of the histone H3 protein . The use of multiple cell lines from the Centre d'Etude du Polymorphisme Humain families allowed us to evaluate variation of chromatin states across pedigrees . These studies demonstrated that chromatin state clustered by family . Our results support the idea that genetic inheritance can determine the epigenetic state of the chromatin as shown previously in model organisms . To our knowledge , this is the first demonstration in humans that genetics may be an important factor that influences global chromatin state mediated by histone modification , the hallmark of the epigenetic phenomena .
Polymorphisms and quantitative differences in gene expression provide the genetic basis for human variation . Studies in humans and other organisms suggest that variation at the transcript level accounts for the majority of the phenotypic variation among species and across individuals within species [1–4] . Recent studies have demonstrated that inherited factors influence gene expression variation between both copies of a gene within an individual [5] as well as between individuals [1 , 6–8] . In a large-scale analysis of allele-specific gene expression using Affymetrix HuSNP chip [9] , we found that allelic variation in gene expression is common , affecting about half of the genes in human genome . This conclusion was supported from the studies of digital gene expression in UniGene database [10 , 11] and allele-specific gene expression using a custom-designed single nucleotide polymorphism ( SNP ) chip [12] . Analysis of allelic variation in gene expression can facilitate identification of regulatory SNPs when the regulatory SNPs are in linkage disequilibrium with an exonic SNP used in the analyses of allele-specific gene expression [13 , 14] . Eukaryotic genomes are organized into chromatin , formed by DNA and protein complex . The basic unit of chromatin is the nucleosome structure containing 146 bp DNA that wraps around a histone octamer . At the chromosome level , gene expression is regulated by distinct chromatin structures . This epigenetic information is often encoded in post-translational modifications of histone proteins such as acetylation , methylation , and phosphorylation [15] . Histone modifications can be maintained through mitotic cell divisions . This stable transmission of epigenetic state through mitosis provides the basis for cellular differentiation and organism development . Although there are a few examples of inheritance of epigenetic information across generations in model organisms [16 , 17] , no investigation of the global effect of genetic inheritance on chromatin state in humans has been reported . In light of genetic influence on allelic gene expression variation in pedigree [5] , we set out to analyze if genetic inheritance also affects chromatin variation in humans , as measured by variations in histone modifications using an allele-specific chromatin immunoprecipitation ( haploChIP ) assay [13] . The use of allele-specific variation in chromatin state in a heterozygous individual is a powerful approach to study genetic influence since other sources of variations in the cellular environment are likely affecting both alleles more or less equally . Our study demonstrates that specific chromatin states as a quantitative trait show familial aggregation .
To evaluate the allele-specificity of our chromatin immunoprecipitation ( ChIP ) assay , we first examined protein binding at three imprinted genes ( LIT1 , H19 , and SNRPN ) and two X-linked genes ( HPRT1 and PGK1 ) loci . We used 12 lymphoblastoid cells derived from 12 individuals from two Centre d'Etude du Polymorphisme Humain ( CEPH ) families . Each cell line was characterized by six antibodies targeting at chromatin proteins . The description of cell lines and experiments can be found in Tables S3 , S6 , S7 , and S8 . Three antibodies target at active chromatin proteins , which are RNA polymerase II ( Pol II ) , histone H3 lysine 9/14 acetylation ( H3Ac ) , and lysine 4 dimethylation ( H3K4 ) . The remaining three antibodies target at inactive chromatin proteins , which are histone H3 lysine 9 dimethylation ( H3K9 ) , lysine 27 dimethylation ( H3K27di ) and trimethylation ( H3K27tri ) . The control DNAs were from whole cell extract , which were prepared as the ChIP experiments , except for the omission of antibodies . We refer to this control DNA as input . We analyzed DNAs that were co-immunoprecipitated by the antibodies using oligo ligation assay ( OLA ) . The results for the differential methylation region in LIT1 promoter are shown in Figure 1A . The paternal allele was specifically pulled down by antibodies targeting at active chromatin ( Pol , Ac , and K4 in Figure 1A; the paternal allele is C for GM10858 , GM11872 , and GM11875 and T for GM10859 , GM10861 , GM10870 , and GM11982 ) . This is consistent with the previous study , which demonstrated that the LIT1 gene was imprinted and was expressed from paternal chromosome only [18] . The maternal allele was preferentially pulled down by antibodies targeting at inactive chromatin ( K9 , K27di , K27tri in Figure 1A; the maternal allele is T for GM10858 , GM11872 , and GM11875 and C for GM10859 , GM10861 , GM10870 , and GM11982 ) . As a control , the input showed nearly equal intensities of both alleles . Promoter regions of H19 , SNRPN , HPRT1 , and PGK1 also displayed expected allele-specificity in our ChIP assays ( Figure S6 ) . After we have established allele-specificity for our ChIP assay using the imprinted genes and X-linked genes , we proceeded to analyze genome-wide allele-specific chromatin states by ChIP-on-chip method with a SNP chip . Since Affymetrix 10K SNP chip was designed for genotyping purpose , we had to modify the protocol in order to use the 10K SNP chip for doing ChIP-on-chip studies . The modified protocol is illustrated schematically in Figure 1B . We first repaired DNA fragments that were co-immunoprecipitated by antibodies or from the nonenriched control DNAs ( input ) by flushing the ends with a nuclease and adding adaptors to the DNA ends ( Figure 1B ) . The DNA fragments were amplified and hybridized separately to the 10K SNP chips . We used 12 lymphoblastoid cell lines derived from 12 individuals , six of them from each of the two CEPH families ( 1347 and 1362 , two parents and four children ) . Each cell line was analyzed with the six antibodies ( Pol II , H3Ac , H3K4 , H3K9 , H3K27di , and H3K27tri ) and two controls ( input and genomic DNA using unmodified protocol ) , which gave 96 ChIP-on-chip experiments . The data from the 96 ChIP-on-chip experiments can be represented in a data matrix , with 96 rows ( experiments ) and 10 , 000 columns ( SNPs ) . Each SNP had two measurements , one for chromatin binding from the A allele and the other from the B allele . We were interested in two derived values . The first one was the total intensity , which was the sum of chromatin-binding intensities from A allele plus B allele . The total intensity was similar to those obtained in conventional ChIP-on-chip experiments . The second one was the relative intensity , which was the ratio of A allele chromatin-binding intensity divided by the total intensity . The relative intensity was uniquely produced in this study due to the use of the SNP chip in ChIP-on-chip experiment . The input serves as an important control for two purposes . First , both input and the ChIP-on-chip experiment used our modified protocol . Comparison of genotype call between genomic DNA and input allowed us to evaluate the allelic specificity of our protocol for this experimental system . We found that the concordance of genotype call between genomic DNA and input was usually around 99% ( Table S1 ) . Thus , the result validated our protocol . Second , it allows us to define biological activity specifically due to chromatin beyond a baseline . The baseline can be assessed by input . Because the complexity in this high dimensional ChIP-on-chip data , we need to reduce the complexity in order to effectively understand the variance structure . We used principal component analysis ( PCA ) for this purpose . In our data , the first two principal components typically account for between 10%–50% of the total variance . Therefore , we can now focus our analyses in two dimensions instead of the original 10 , 000 dimensions . The result from PCA analysis using total intensity ( A + B ) for the ChIP-on-chip data is shown in Figure 2A . We plotted the 96 samples using the scores from the first principal component ( PC1 ) and the second principal component ( PC2 ) . As shown in Figure 2A , the samples were clustered by the antibodies using the total chromatin-binding intensities ( A + B ) . For example , the samples from antibodies targeting at active chromatin ( Pol II , H3Ac , and H3K4 ) are on the left . Samples from two ( H3K27di and H3K27tri ) of the three antibodies targeting at inactive chromatin are on the right . Samples from the third antibody targeting inactive chromatin ( H3K9 ) and the controls are in the middle . This is expected because chromatin states are determined by histone modifications and Pol II activity . Samples from the two families ( red and blue , Figure 2A ) , are all intermixed . Therefore , we concluded that the major determinant of the total variance in the ChIP-on-chip experiments was due to variations in chromatin states as revealed by the antibodies targeting at different modification forms of histone H3 proteins and Pol II when using the total intensity . However , we got a totally different picture when PCA was performed with the relative intensity ( A/A + B ) ( Figure 2B ) . Now the samples from the family 1 and family 2 ( red and blue , Figure 2B ) were separated from each other into two clusters . The separation was the largest for the antibodies targeting at active chromatin , which were represented by the open symbols at the bottom of Figure 2B . So the global chromatin states as measured by the relative intensity from the A allele differ for the individuals in family 1 versus the individuals in family 2 . This observation led us to conclude that genetic inheritance can influence chromatin modifications . To validate this important finding , we carried out the same ChIP-on-chip experiments and analyses for two additional families ( 1331 and 1413 ) . We analyzed twelve lymphoblastoid cell lines , six from each of the two CEPH families ( 1347 and 1362 ) , with the two antibodies ( H3Ac and H3K4 ) and the two controls ( input and genomic DNA ) . Once again , we saw clustering of the samples by controls/antibody ( H3Ac and H3K4 ) when PCA was performed using the total intensity ( A + B ) ( Figure 3 , left panels , three pair-wise comparisons among CEPH families 1347 , 1331 , and 1413 ) . More importantly , samples from the two different families were separated into two clusters when PCA was analyzed using the relative intensity ( A/A + B ) ( Figure 3 , right panels , three pair-wise comparisons among CEPH families 1347 , 1331 , and 1413 ) . To better understand the genetic influences on chromatin variation , we constructed pairs of genetically related individuals ( siblings or parent-child ) , as well as pairs of genetically unrelated individuals from heterozygous individuals in the four CEPH families . We then computed the difference between the two individuals in each pair . To identify those SNPs that had similar chromatin state within genetically related individuals , we compared the variance in genetically related pairs versus the variance in genetically unrelated pairs . We identified seven SNPs ( F-test , p < 0 . 05 ) . Variation in chromatin state for the seven SNPs was smaller in the genetically related pairs than the variation in the genetically unrelated pairs ( Figure 4 ) , indicating similar chromatin state between the related individuals . These differences were specifically observed in the ChIP experiment ( in H3Ac but absent in input ) . Yan et al . previously demonstrated that allelic gene expression variation segregated as a Mendelian trait [5] . To evaluate if allelic chromatin variation also follows Mendelian inheritance , we performed inheritance analysis for the seven genes analyzed in Figure 4 . All seven genes showed segregation patterns that were consistent with Mendelian inheritance ( Figure 5 and Figure S8 ) . For examples , ABB haplotype in GM10859 ( mother in CEPH family 1347 ) in the case of rs938335 had low H3Ac binding activity ( below two standard deviations from the mean intensity of B allele ) , whereas BAA haplotype in GM10859 had high H3Ac binding activity ( above two standard deviations from the mean of A allele ) . The two heterozygous children are GM11871 and GM11875 , both of whom received BAA from the mother . The allelic fraction values ( A/A + B ) are 0 . 61 and 0 . 73 , respectively , which are higher than 0 . 5 . But the allelic values are not as extreme as the one in GM10859 . This is because the paternal allele AAB has normal level of H3Ac binding activity . Similarly , BAA and ABB haplotypes in CEPH family 1362 have low H3Ac binding activity . Therefore , GM11982 and GM11983 had low allelic fraction values , 0 . 41 and 0 . 36 , respectively . However , GM11984 received both alleles that had low H3Ac binding activities . Consequently , the allelic fraction was 0 . 48 , very close to 0 . 5 . Conversely , GM11987 received both alleles that had normal H3Ac binding activities , thus the allelic fraction value was also close to 0 . 5 . Note that this is different from conventional Mendelian inheritance analysis in that it uses the allelic fraction as a phenotypic trait , and this depends on relative quantities between the two alleles . Nevertheless , our results agree very well with inheritance of the chromatin state , in turn providing direct support for genetic influence on the chromatin state . However , we must qualify our results by noting that , in contrast to the relatively large number of informative individuals studied in Yan et al . [5] ( eight and ten informative individuals per family for two different genes ) ( Figure 1 ) , the maximum number of individuals informative for any SNP tested in any family in our study is five . This limits the statistical power of our inheritance analysis , despite a highly suggestive result . Taken together , these results suggest that inherited genetic components could determine the epigenetic state of the chromatin . To our knowledge , this is the first demonstration in humans that genetic inheritance may be an important factor directing the global chromatin state mediated by histone modification , the hallmark of the epigenetic phenomena .
Our aim was to determine if genetic inheritance can influence chromatin state globally in humans . Our studies support the notion that inherited genetic components can determine the epigenetic state of the chromatin . Our strategy was to use samples from different pedigrees to assess the genetic effect . The use of the SNP chip to measure allele-specific chromatin-binding intensity in a heterozygous individual and the use of relative binding intensity between the two alleles enables us to detect difference in chromatin state in individuals between different families since other sources of variations in the cellular environment are likely affecting both alleles more or less equally . The use of PCA made it possible to focus analyses on a few components , which have the capacity to combine weak signals from multiple genetic loci . Otherwise , the weak signal may not be detectable when analyzed individually . We used a combination of 12 lymphoblastoid cell lines and six antibodies plus two controls in the experiment . This is two-factor experiment design . Genetic factor has two levels , one for each family; whereas chromatin factor has three levels , one for active chromatin , one for inactive chromatin , and one for control . This study design allows us to assess genetic inheritance effect as well as chromatin states targeted by the six antibodies on the total variance across the 96 experiment data . We are interested in the variance across the 96 samples . In our variance component model , we decomposed the total variance into three components , genetics , chromatin , and residual variance . Because the complexity in this high dimensional ChIP-on-chip data , we need to reduce the complexity to effectively understand the variance structure . We used PCA for this purpose . What PCA does is to transform the data matrix by rotating the coordinate system . After transformation , we have a new set of variables , denoted by principal components . Each principal component is a linear combination of the original variables . PCA has two useful mathematic properties . First , all principal components are orthogonal to each other , so the total variance is simply the sum of variances from each principal component . Second , principal components are ranked so that PC1 accounts for the largest variance in the data , followed by PC2 . In our studies , the first two principal components usually account for about 10%–50% of the total variance . Therefore , we were able to focus the analyses in two dimensions instead of the original 10 , 000 dimensions . PCA using total ChIP signal as ( A + B ) ( Figures 2A and 3 , panels on the left ) indicated that the total variance in the samples was comprised mostly by antibodies targeting at various chromatin proteins , which also demonstrated the specificity of the ChIP assay . In contrast , PCA using the relative signal ( A/A + B ) indicated that the total variance in the samples comprised primarily the difference between two families and secondarily antibodies targeting at various chromatin proteins ( Figures 2B and 3 , panels on the right ) . The separation between different families in controls served as the baseline , which captured the background level of difference due to genotypes . The separation between different families is much too large for the antibodies targeting at active chromatin , indicating specific chromatin state differences between different families . This result suggests that genetic inheritance can influence the global chromatin state . The relative intensity measurement ( A/A + B ) has better sensitivity in detecting the genetic effect than the total intensity ( A + B ) , since other sources of variations in the cellular environment that could affect the total intensity are likely affecting both alleles more or less equally , thus not masking the genetic effect on the relative intensity in chromatin state . In the case of PCA of families 1 and 2 ( Figure 2B ) using the relative intensity , the largest variance , captured by PC1 , was due to the difference between family 1 and family 2 . In the case of PCA analysis of families 3 and 4 ( Figure 3 , top right ) , the largest variance ( always captured by PC1 because of the algorithm ) was due to the difference between control and antibodies targeting active chromatin states ( H3Ac and H3K4 ) . PC2 was the vector that contained the second largest variance in these data , corresponding to the difference between family 3 and family 4 . The conclusion of genetic influence on chromatin state is supported by the clustering of the families when samples are projected in the 2-D space defined by PC1 and PC2 . The conclusion is valid regardless of whether the separation is on PC1 or PC2 , which is determined by the variance-covariance structure of the data . In other words , principal components are data driven . PCA is an unsupervised method . Furthermore , our allelic segregation analysis agrees very well with Mendelian inheritance of the chromatin state ( Figures 5 and S8 ) , thus providing direct support for genetic influence on the chromatin state . It is interesting to note that familial aggregation of allelic-specific DNA methylation variation at imprinted gene loci has been previously reported [19] as well as Mendelian inheritance of DNA methylation [20] . A total of three recent studies also indicated germline inheritance of methylation epimutation in MSH2 and MLH1 in families with hereditary nonpolyposis colorectal cancer [21–23] . DNA methylation and histone acetylation showed nearly identical patterns in young monozygotic twins but marked differences in old monozygotic twins [24] . All these observations support the notion of the influence of genetic inheritance on epigenetic processes . A genetic effect on chromatin state is well known in model organisms . Examples include position-effect variegation in Drosophila melanogaster [25] . A related observation is transgenerational epigenetic inheritance . For example , agouti viable yellow mice display inheritance of yellow fur as a result of incomplete erasure of the methylation signal associated with a retrotransposon insertion [16] , and kinked-tail mice transmit phenotype through multiple generations due to the loss of the silent epigenetic state at the Axin gene [26] as well as a heritable white-tail phenotype associated with Kit-specific microRNAs [17] . Meiotic transmission of epigenetic states has also been described in several studies in plant [27–29] . Although total gene expression as expression quantitative trait loci was regulated by genetic loci [6–8] , the detection of the expression quantitative trait loci usually required a much larger sample size than the 12 samples used here . Our ability to detect familial aggregation by allele-specific chromatin state , but not in total chromatin state , resulted from the increased specificity of probing chromatin state with the relative intensity ( A/A + B ) . The use of PCA might further enhance our ability to detect the genetic effects on chromatin state , since PCA had the capacity to detect a robust signal captured in the principal components even though signals from individual SNPs might be weak . The allelic differences in chromatin provided an explanation for the observed allelic variation in gene expression [5] .
ChIP was carried out using a ChIP assay kit ( Upstate , ( http://www . upstate . com/img/coa/17-295-33519A . pdf ) . Lymphoblastoid cells of 24 individuals from CEPH/Utah pedigrees ( family identification 1347 , 1362 , 1331 , and 1413 ) were used in this study . ChIP was carried out using a ChIP assay kit ( Upstate ) with antibodies against histone H3 acetylated at K9 and K14 ( Upstate , 06–599 ) , dimethylated at K4 ( Upstate , 07–030 ) , dimethylated at K9 ( Upstate , 07–441 ) , dimethylated at K27 ( Upstate , 07–452 ) , trimethylated at K27 ( Upstate , 07–449 ) , and Pol II ( Abcam , ab5408 , http://www . abcam . com ) . All cell lines are described in Table S3 . Briefly , 2 × 107 cells were grown in RPMI medium 1 , 640 supplemented with 2 mM L-glutamine and 15% FBS . The cells were fixed by adding formaldehyde solution into the culture medium to a final concentration of 1% . After centrifugation the cell pellets were rinsed twice with an ice-cold PBS solution and then suspended in a lysis buffer ( all buffers used in the ChIP experiment are described in http://www . upstate . com/img/coa/17-295-33519A . pdf ) . Sonication was performed on ice using Cole Parmer economical ultrasonic processor at power 9 for 12 cycles of sonication , each cycle for 10 s followed by a 30-s break on ice . The cell pellets were centrifuged at 10 , 000 RCF ( ×g ) for 10 min , and the resulting lysates in the supernatant were stored at −80 °C until use . The chromatin lysates were diluted by 10-fold in a ChIP dilution buffer . They were precleared by Salmon sperm protein A agarose and incubated with each of the six antibodies individually overnight at 4 °C . The chromatin complexes were sequentially washed in low salt , high salt , LiCl salt , and TE buffers . The protein/DNA complex was eluted in an SDS elution buffer ( 1% SDS , 0 . 1 M NaHCO3 ) . The crosslink between protein and DNA was reversed . The protein/DNA complex was treated with Proteinase K . DNAs were purified using Qiagen mini-elute reaction clean-up kit ( http://www1 . qiagen . com ) . PCR was carried out using primer pairs described in Table S2 . Antarctic Phosphatase ( New England Biolabs , http://www . neb . com ) and Exonuclease I ( New England Biolabs ) was used to remove unincorporated primers and dNTPs . Allele-specific OLA was carried out in a 5-μl reaction containing 1× Ampligase buffer ( Epicentre Biotechnologies , http://www . epibio . com ) , 100 nM each ligation primers , 0 . 5 U Ampligase , and 1 μl of phospho/exo treated PCR product ( ∼10 ng ) for 30 cycles , with each cycle at 95 °C for 30 s , 50 °C for 30 s , and 65 °C for 2 min . All primers are described in Table S2 . Ligation products were resolved by ABI3730XL genetic analyzer and analyzed using GeneMapper 3 . 5 software ( Applied Biosystems , http://www . appliedbiosystems . com ) . We treated 500-ng input DNA or 50-ng immunoprecipitated DNAs in the ChIP experiment with mung bean nuclease to flush the ends . The DNA was phosphorylated and ligated to an Xba-linker ( Table S2 ) . Following an Xba I digestion , DNA was purified by Qiagen mini-elute reaction clean-up kit and was ligated to an Xba-adaptor . DNA was then amplified using an Xba-primer . This amplification step did not introduce biased representation of the initial ChIP DNA ( Figure S7 ) . It also retained the allelic specificity as demonstrated by the experiment described in Figures 1A and S6 . Next , 10-μg PCR products from the input or 5-μg PCR products from the ChIP experiments were digested and labeled as described in the 10K SNP chip manual . We carried out the hybridization , washing , and scanning as described in the manual . All statistical analyses were developed using R and Splus packages . The missing values in PM or MM probes were replaced by the mean MM across all SNPs . For each SNP , we computed the ratio PM/MM and then applied the Robust Multi-array Average ( RMA ) method [30] . Probe intensity was computed by the function of max ( mean ( log2[PM/MM] , 0 ) for allele A and allele B . The intensity at the probe set level was the average of the ten pairs of the probes from each allele of the SNP . The signal for each allele of an SNP was evaluated by t-test for the measurement of ( PM − MM ) with H0 hypothesis of ( PM − MM ) = 0 for the ten probes for a given SNP . We chose a p-value of 0 . 01 as a threshold for the presence of a signal . We used PCA to visualize similarity and variability among the 96 samples containing the ChIP data done on 12 individuals , each characterized with six antibodies plus the controls of input and DNA , using either the total binding intensity ( A + B ) or the relative binding intensity ( A/A + B ) . PCA transforms the data matrix by rotating the coordinate system . After transformation , we had a new set of variables , denoted by principal components . Each principal component was a linear combination of the original variables with different weights ( loadings ) . The loadings reflected the degree of contribution of each SNP to the principal component . PCA has two useful mathematic properties . First , all principal components are orthogonal to each other so the total variance is the sum of variances from each principal component . Second , principal components are ranked in such a way so that PC1 accounts for the largest variance in the data followed by PC2 . In our data , the first two principal components typically accounted for about 10%–50% of the total variance . Therefore , we focused our analyses in two dimensions instead of the original 10 , 000 dimensions . The utilities of the PCA in this study are 2-fold . First , PCA provides dimension reduction , allowing visualization of data structure in 2-D . Second , it provides a quantitative assessment of data structure and interactions among variables . In this study , the data structure refers to the clustering of samples by family or antibody type . The separation of samples by different principal components reflects the degree of difference due to either family or antibody . The separation in PC1 is always the largest , by definition , due to the PCA algorithm . The relative contribution of the components can be assessed by eigen-values , which are provided in Tables S4 and S5 . The expected value of relative binding intensity ( A/A + B ) is 0 . 5 . Deviation from 0 . 5 for an SNP among heterozygous individuals for genomic DNA and input suggests erroneous behavior of the SNP . We removed SNPs whose deviation from 0 . 5 exceeded two standard deviations . A total of 2 , 365 SNPs were removed by this criterion . We used 0 . 5 for homozygous individuals in the PCA for the relative binding intensity ( A/A + B ) . All samples were projected in the space defined by the first and second principal components .
The National Center for Biotechnology Information ( NCBI ) Entrez Gene ( http://www . ncbi . nlm . nih . gov/entrez/query . fcgi ? db=gene ) accession numbers for the genes discussed in this paper are ASTN2 , 23245; C6orf190 , 387357; CD19 , 930; CD3G , 917; GAPDH , 2597; MYOD1 , 4654; NES , 10763; PKHD1 , 5314; RPLP1 , 6176; SYT9 , 143425; TCBA1 , 154215; TIAM1 , 7074; and TMEM16D , 121601 . | Human health and disease are determined by an interaction between genetic background and environmental exposures . Both normal development and disease are mediated by epigenetic regulation of gene expression . The epigenetic regulation causes heritable changes in gene expression , which is not associated with DNA sequence changes . Instead , it is mediated by chemical modification of DNA such as DNA methylation or by protein modifications such as histone acetylation and methylation . Although much has been known about epigenetic inheritance during development , little is known about the influence of the genetic background on epigenetic processes such as histone modifications . In this report the authors studied five histone modifications on a genome-wide level in cells from different families . Global epigenetic states , as measured by these histone modifications , showed a similar pattern for cells derived from the same family . This study demonstrates that genetic inheritance may be an important factor influencing global chromatin states mediated by histone modifications in humans . These observations illustrate the importance of integrating genetic and epigenetic information into studies of human health and complex diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"in",
"vitro",
"homo",
"(human)",
"genetics",
"and",
"genomics"
] | 2007 | Allele-Specific Chromatin Immunoprecipitation Studies Show Genetic Influence on Chromatin State in Human Genome |
Finding functional DNA binding sites of transcription factors ( TFs ) throughout the genome is a crucial step in understanding transcriptional regulation . Unfortunately , these binding sites are typically short and degenerate , posing a significant statistical challenge: many more matches to known TF motifs occur in the genome than are actually functional . However , information about chromatin structure may help to identify the functional sites . In particular , it has been shown that active regulatory regions are usually depleted of nucleosomes , thereby enabling TFs to bind DNA in those regions . Here , we describe a novel motif discovery algorithm that employs an informative prior over DNA sequence positions based on a discriminative view of nucleosome occupancy . When a Gibbs sampling algorithm is applied to yeast sequence-sets identified by ChIP-chip , the correct motif is found in 52% more cases with our informative prior than with the commonly used uniform prior . This is the first demonstration that nucleosome occupancy information can be used to improve motif discovery . The improvement is dramatic , even though we are using only a statistical model to predict nucleosome occupancy; we expect our results to improve further as high-resolution genome-wide experimental nucleosome occupancy data becomes increasingly available .
Finding functional DNA binding sites of transcription factors ( TFs ) throughout the genome is a necessary step in understanding transcriptional regulation . However , despite an explosion of TF binding data from high-throughput technologies like ChIP-chip ( [1 , 2] , and many more ) , DIP-chip [3] , PBM [4] , and gene expression arrays ( [5 , 6] , and many more ) , finding functional occurrences of binding sites of TFs remains a difficult problem because the binding sites of most TFs are short , degenerate sequences that occur frequently in the genome by chance . In particular , matches to known TF motifs in the genome often do not appear to be bound by the respective TFs in vivo . One popular explanation for this is that when the DNA is in the form of chromatin , not all parts of the DNA are equally accessible to TFs . In this state , DNA is wrapped around histone octamers , forming nucleosomes . The positioning of these nucleosomes along the DNA is believed to provide a mechanism for differential access to TFs at potential binding sites . Indeed , it has been shown that functional binding sites of TFs at regulatory regions are typically depleted of nucleosomes in vivo [7–12] . If we knew the precise positions of nucleosomes throughout the genome under various conditions , we could increase the specificity of motif finders by restricting the search for functional binding sites to nucleosome-free areas . Here , we describe a method for incorporating nucleosome positioning information into motif discovery algorithms by constructing informative priors biased toward less-occupied promoter positions . Our method should improve motif discovery most when it has access to high-resolution nucleosome occupancy data gathered under various in vivo conditions . Unfortunately , this data is not currently available for any organism at a whole-genome scale , let alone under a variety of conditions . Nevertheless , because our method is probabilistic , even noisy evidence regarding nucleosome positioning can be effectively exploited . For example , Segal et al . [12] recently published a computational model—based on high-quality experimental nucleosome binding data—that predicts the probability of each nucleotide position in the yeast genome being bound by a nucleosome; these predictions are intrinsic to the DNA sequence and thus independent of condition , but were purported to explain around half of nucleosome positions observed in vivo . In addition , Lee et al . [9] have used ChIP-chip to profile the average nucleosome occupancy of each yeast intergenic region . We show that informative positional priors , whether learned from computational occupancy predictions or low-resolution average occupancy data , significantly outperform not only the commonly used uniform positional prior , but also state-of-the-art motif discovery programs .
We formulate a probabilistic motif discovery framework for identifying TF motifs in sets of DNA sequences , such as those arising from ChIP-chip experiments . The goal is to find a TF motif of length W in a set X of sequences that are presumed to be bound by the TF . We proceed in three steps . First , we compute a score for each W-mer present at each position of each sequence in X that reflects the probability the TF binds the W-mer at that position . Second , from these scores we compute an informative “positional prior , ” a non-uniform probability distribution over the positions of each sequence in X . Third , we incorporated this positional prior into a search algorithm that simultaneously learns the position of a binding site in each sequence in X , along with the parameters of the motif recognized by the TF . Although our method can be used with any motif model , we use a position-specific scoring matrix , or PSSM [13] . Regarding the first step , we examine two different choices of score: S℘N℘ and S℘D℘N℘ ( Figure 1 ) . The score S℘N℘ for a particular position is computed from the nucleosome occupancy of the W-mer beginning at that position . In contrast , the score S℘D℘N℘ for a particular position is computed from a discriminative perspective , incorporating information about the nucleosome occupancy of all occurrences of the W-mer in all intergenic regions , including those in the set of unbound sequences Y . This builds on the observation made by Segal et al . [12] that nucleosome occupancy is lower at sites that are bound in vivo than sites that are not bound in vivo . In the second step of our method , from these two choices of score S℘N℘ and S℘D℘N℘ , we build two positional priors N℘ and D℘N℘ , respectively ( for further details , see Materials and Methods ) . In the third and final step , we incorporate these two priors into a Gibbs sampling–based search method called PRIORITY [14] , and we call the two variations PRIORITY-N℘ and PRIORITY-D℘N℘ , respectively . To quantify the extent to which the two new informative priors N℘ and D℘N℘ improve motif discovery , we compare their performance with the performance of a uniform prior U℘ . We similarly incorporate this prior into PRIORITY , and call this variation PRIORITY-U℘ . We apply all three algorithms to sequence-sets arising from ChIP-chip experiments published by Harbison et al . [2] . In assessing accuracy , we consider only the 80 TFs for which a consensus binding motif is known . These 80 TFs were profiled under various environmental conditions , resulting in a total of 156 sequence-sets where we can reasonably assess the accuracy of a motif discovery algorithm . We consider an algorithm to be successful when applied to a sequence-set if the top-scoring motif matches the literature consensus for the corresponding TF , where a match is defined as a distance of less than 0 . 25 ( using a slight variant of the inter-motif distance measure described by Harbison et al . ; see Protocol S1 ) . Figure 2 summarizes the results of the three algorithms PRIORITY-U℘ , PRIORITY-N℘ , and PRIORITY-D℘N℘ using this criterion on the 156 sequence-sets . Overall , PRIORITY-D℘N℘ finds the correct motif in 70 sequence-sets , resulting in an improvement of 52% over the baseline PRIORITY-U℘ which finds 46 . The last four columns in Figure 2 reveal that there is no case where PRIORITY-D℘N℘ fails but PRIORITY-U℘ or PRIORITY-N℘ succeeds . In other words , the D℘N℘ prior was never harmful to motif discovery . The N℘ prior finds the true motif 51 times , not nearly as often as D℘N℘ , and only marginally more often than U℘ . We now discuss these results in greater detail . We expect the simple nucleosome prior N℘ to perform well when functional binding sites of the profiled TF are generally less occupied by nucleosomes than other locations within the same DNA sequence . One instance where this is known to occur is in sequences bound by Leu3 , since the experimental data of Liu et al . [15] show that loci bound by Leu3 in vivo are typically depleted of nucleosomes . As expected , PRIORITY-N℘ finds the true motif of Leu3 in both of the environments where it was profiled by Harbison et al . When Leu3 is profiled in SM , PRIORITY-U℘ also succeeds , but when profiled in YPD , PRIORITY-U℘ fails . We take a closer look at this case to understand better why prior N℘ is more effective in identifying the true motif of Leu3 . To do so , we calculate the average S℘N℘ score for each 10-mer present in the Leu3_YPD sequence-set ( Figure 3A ) . Leu3 is known to recognize the 10-mer CCGGNNCCGG , with a slight preference for CCGGTACCGG [15 , 16] , and indeed we find that fewer than 10% of 10-mers score higher than CCGGTACCGG , revealing that the prior N℘ is assigning a higher prior probability to positions containing the true motif . Although PRIORITY-N℘ is more successful than PRIORITY-U℘ overall ( 51 successes versus 46 ) , the second column in Figure 2 reveals that in five sequence-sets , PRIORITY-U℘ performs better than PRIORITY-N℘ . The score S℘N℘ used to compute the prior N℘ reflects the accessibility of the W-mer at a particular position . While it is true that regions bound by the profiled TF should be accessible , it does not follow that every accessible region is bound by the profiled TF . Some accessible regions could be binding sites of other TFs or other functional DNA elements . Indeed , in four of the five cases where PRIORITY-U℘ does better , PRIORITY-N℘ finds a motif rich in A's and T's; it has been previously shown that many yeast promoters contain poly ( dA-dT ) sequences that stimulate transcription [17] . Furthermore , due to their intrinsic DNA structure , poly ( dA-dT ) sequences are often free of nucleosomes , and they are believed to increase TF accessibility by delocalizing nucleosomes in vivo [17–19] . Since PRIORITY-N℘ is expected to find highly accessible DNA sequences that occur often in a given set of bound promoters , it is not surprising that it sometimes finds poly ( dA-dT ) sequences . However , we notice that such sequences occur often and are accessible not only in the bound set X , but also in the rest of the genome , so they are not specific to the profiled TF . The computation of the score S℘D℘N℘ used to compute the prior D℘N℘ addresses the issue of nucleosome-free regions that are not specific to the profiled TF . A ChIP-chip experiment gives rise to sequences that are bound by the profiled TF as well as those that are not bound . Using both these sets of sequences , each W-mer in the bound set can be scored according to how many times it occurs in each set , as well as how accessible it is in each set . This discriminates between sites that are highly accessible only in the bound set and sites that are highly accessible throughout the genome . The former are more likely to be true binding sites of the profiled TF . Figure 4 shows a range of examples where S℘D℘N℘ is able to correctly upweight the prior probability of the location of the true binding site . When we perform the same word-analysis for S℘D℘N℘ in Leu3_YPD as we did for S℘N℘ , we see that S℘D℘N℘ is even better at predicting the true binding site than S℘N℘ ( Figure 3B ) . In fact , no 10-mer has an S℘D℘N℘ score higher than CCGGTACCGG , the known consensus Leu3 binding site . In 14 sequence-sets , motif discovery benefits from nucleosome occupancy information only when this information is used in a discriminative manner ( column 4 in Figure 2 ) . We perform an analysis for S℘D℘N℘ in these sequence-sets similar to the one we did earlier for S℘N℘ in Leu3_YPD . For simplicity , we restrict our attention to the nine sequence-sets which have a known literature consensus of length less than ten bases ( see Figure S1 ) . In seven of the nine cases , fewer than 5% of the S℘D℘N℘ scores are better than that of the true motif; the average over all nine being only 8% . The corresponding average for S℘N℘ is 39%; in three of the nine cases , more than 50% of the scores are better than that of the true motif ( even with a uniform prior , the number should be only 50% in expectation , implying that in these cases , S℘N℘ is worse than uniform ) . Thus , it is not surprising that when PRIORITY-U℘ fails in these cases , PRIORITY-N℘ also fails . Note that the prior N℘ over a particular intergenic sequence does not change regardless of which TF binds it . However , since S℘D℘N℘ is computed using both bound and unbound sequences , the prior D℘N℘ can be different over the same sequence depending on the TF that binds it . Figure 5 shows the different S℘D℘N℘ scores computed over the intergenic sequence iYMR280C which belongs to four sequence-sets: Reb1_H2O2Lo , Reb1_YPD , Ume6_H2O2Hi , and Ume6_YPD . Figure 5 demonstrates the specificity toward binding sites of only the profiled TF when the nucleosome prior is computed from a discriminative perspective . We compiled results from six state-of-the-art motif discovery programs as reported by Harbison et al . on the same 156 sequence-sets: AlignACE [20] finds 16 , MEME [21] finds 35 , MDscan [22] finds 54 , MEME_c [2] finds 49 , a method by Kellis et al . [23] finds 50 , and CONVERGE [2] finds 56 correct motifs . Each of these methods makes use of different sources of information for motif discovery . AlignACE and MEME use different search techniques ( Gibbs sampling and Expectation Maximization [24] ) , but use no additional information and thus are directly comparable to PRIORITY-U℘ . MDscan uses p-values resulting from the ChIP-chip experiments , while the last three programs make use of sequence conservation across various species of yeast . PRIORITY-D℘N℘ , with 70 correct motifs , outperforms all these methods . Table S1 shows the performance of each program in detail . PRIORITY-D℘N℘ is able to capture true protein–DNA interactions even in the case of TFs that form multiple complexes , such as Ste12 . It has been shown experimentally that Ste12 is part of two distinct complexes , Ste12/Dig1/Dig2 and Tec1/Ste12/Dig1 , which control two distinct transcriptional programs: filamentation and mating [25] . Chou et al . [25] show that the promoters of most filamentation genes are bound by the Tec1/Ste12/Dig1 complex , with Tec1 binding DNA directly ( Figure 6A ) . The promoters of most mating genes , however , are bound by either the Ste12/Dig1/Dig2 or the Tec1/Ste12/Dig1 complex , with Ste12 binding DNA directly in both cases ( Figure 6B ) . Dig1 is not currently known to have a DNA binding site , and a literature search did not reveal any evidence of Dig1 binding DNA directly . In the experiments of Harbison et al . [2] , Dig1 , Ste12 , and Tec1 were all profiled after treatment with alpha factor for 30 min ( Alpha ) and after treatment with butanol for 14 h ( BUT14 ) . In all six sequence-sets corresponding to the three TFs in Alpha and BUT14 , both the Tec1 binding site ( CATTCy ) and the Ste12 binding site ( ATGAAAC ) occur often and are statistically significantly enriched . However , taking into account the experimental results of Chou et al . , and the fact that butanol treatment induces the expression of filamentation genes , one would expect that in BUT14 , the Tec1 binding site is the real site of interaction between DNA and the transcriptional complex Tec1/Ste12/Dig1 ( Figure 6A ) . Indeed , when we run our algorithm PRIORITY-D℘N℘ on the sequence-sets Ste12_BUT14 , Tec1_BUT14 , and Dig1_BUT14 , the learned motif in all three cases is the Tec1 motif ( CATTCy ) , as shown in Figure 6A . On the other hand , treatment with the alpha factor pheromone induces the expression of mating genes , and therefore in Alpha one would expect both Dig1 and Tec1 to bind DNA indirectly through Ste12 ( Figure 6B ) . Indeed , the Ste12 motif ( ATGAAAC ) was reported by PRIORITY-D℘N℘ for all three sequence-sets , Ste12_Alpha , Tec1_Alpha , and Dig1_Alpha . In both Ste12_BUT14 and Tec1_Alpha sequence-sets , PRIORITY-U℘ fails to find a motif matching either the Ste12 or the Tec1 motif . Interestingly , the average predicted nucleosome occupancy of Ste12 and Tec1 binding sites in Ste12_BUT14 is 0 . 91 and 0 . 84 , respectively , and in Tec1_Alpha is 0 . 81 and 0 . 90 , respectively . In other words , Tec1 binding sites are less occupied by nucleosomes in Ste12_BUT14 , while Ste12 binding sites are less occupied in Tec1_Alpha . This fact is exploited successfully by PRIORITY-D℘N℘ . For every input sequence-set , PRIORITY-D℘N℘ returns the top-scoring motif along with its score ( see Protocol S1 for the computation of the score ) . To assess whether a motif score is significant , we run PRIORITY-D℘N℘ on 50 randomly generated sequence-sets of the same cardinality . The observed scores from these random sequence-sets of a particular cardinality are well-fit by a normal distribution . Thus , each motif learned by PRIORITY-D℘N℘ on a particular ChIP-chip sequence-set can be assigned an empirical p-value calculated from this distribution . Figure S2 shows the motifs learned from the 156 sequence-sets of TFs with literature consensus DNA binding sites , along with their p-values . We can plot precision-recall and receiver operating characteristic curves based on the p-values of these known motifs ( Figure S3 ) . For a given p-value cutoff , we notice that in many false positive instances , PRIORITY-D℘N℘ finds a high-scoring motif that resembles TGTGTGTG or CACACACA . Poly ( GT/CA ) tracts are known to be common in yeast [26] , so for the remainder of this part of the analysis we disregard sequence-sets for which PRIORITY-D℘N℘ learns a motif of this form . For the others , we can use the precision-recall curve to estimate the false discovery rate ( FDR ) of our novel predictions . A consensus DNA binding motif was not known for 67 of the TFs profiled by Harbison et al . at the time the ChIP-chip experiments were performed . These 67 TFs were profiled under various environmental conditions , yielding a total of 82 sequence-sets . We run PRIORITY-D℘N℘ on these sequence-sets and obtain the top-scoring motif , along with its score . As before , we compute the p-values of each of the learned motifs ( Figure S4 ) . At a p-value of 5 . 0 × 10−6 , we estimate the FDR to be less than 15% . Of the 82 new motifs , 14 have a p-value lower than 5 . 0 × 10−6 when we exclude motifs resembling TGTGTGTG; our FDR estimate would suggest that 12 of these are likely to be correct . Two motifs are for Dig1_Alpha and Dig1_BUT90 . As expected , the motif learned from Dig1_Alpha resembles the Ste12 motif , while the motif learned from Dig1_BUT90 resembles the Tec1 motif ( see Figure 6 ) . Another significant motif is that of Rfx1_YPD and the binding site of Rfx1 now listed in TRANSFAC 11 . 1 matches the learned motif . We construct a condition-dependent , nucleosome-guided map of TF binding sites derived from these 14 motifs , along with the 72 matching the literature consensus ( including the Tec1 motif learned in Ste12_BUT90 and the Ste12 motif learned in Tec1_Alpha ) . The 86 sequence-sets correspond to 55 TFs profiled in one or more of ten environmental conditions . In their ChIP-chip experiments , Harbison et al . report a total of 2 , 387 promoter sequences to be bound by one of these TFs . Our map contains a total of 2 , 347 high-confidence TF binding sites within these sequences . Lee et al . [9] report results from ChIP-chip experiments where the densities of histones H3 and H4 are profiled over the whole genome . This in vivo nucleosome occupancy data is at a resolution of approximately one kilobase , so we cannot use it to obtain distinct scores over individual nucleotide positions . However , we can still use it to weight entire intergenic regions in a discriminative manner . We first use a logit transformation to map the reported intensity over each intergenic region into a probability ( see Materials and Methods ) . We then assume that each position within a sequence has an occupancy probability equal to the occupancy probability of the whole sequence , and compute a new version of the S℘D℘N℘ score , which we call S℘D℘N℘′ . Figure 3C shows the distribution of the S℘D℘N℘′ scores of all 10-mers present in Leu3_YPD . As in the case of the S℘D℘N℘ score , S℘D℘N℘′ assigns the 10-mer CCGGTACCGG the highest rank , which is encouraging . Indeed , the corresponding prior , which we call D℘N℘′ , performs admirably overall as well: PRIORITY-D℘N℘′ learns a total of 66 motifs correctly . A more detailed look shows that it does worse than PRIORITY-D℘N℘ in seven sequence-sets , but better in three . Since this nucleosome occupancy data is obtained in YPD , one might expect the benefits to be primarily in sequence-sets obtained from TFs profiled in YPD . However , of the three sequence-sets where D℘N℘′ does better , two are not in YPD . Perhaps the nucleosome landscape does not change much across various environmental conditions for these TFs; this has been shown to be true in the case of certain TFs , like the heat shock protein Hsf1 [11] . Or perhaps these represent sequence-sets where the computational model on which D℘N℘ is based is not as accurate as the low-resolution in vivo data . What happens when nucleosome occupancy data is not available ? In this case , a special version of the D℘N℘ prior can be computed in which the occupancy is assumed to be uniform over all sequences ( note that this is different from D℘N℘′ where the occupancy is assumed to be uniform over the positions within each individual sequence , but may change across sequences ) . The information in this simple discriminative prior derives not from any nucleosome data whatsoever , but only from the sequence content of the bound and the unbound sets . The Gibbs sampler incorporating this prior correctly identifies 60 true motifs , demonstrating the utility of a discriminative perspective . Although not as effective as PRIORITY-D℘N℘ or PRIORITY-D℘N℘′ , the improvement of 30% of this prior over U℘ is nevertheless significant . Detailed results obtained using this prior are available in Table S1 .
Although it has been known for a while that nucleosomes modulate the binding activity of TFs by providing differential access to DNA binding sites [7–12] , we believe we are the first to use nucleosome occupancy information to more accurately predict de novo binding sites of TFs . To be clear , we do not assume that nucleosomes bind DNA first and that TFs bind whatever remains accessible ( nor the other way around ) . Rather , we imagine that nucleosomes and TFs are together in competition for positions on the genome and their binding configurations are sampled from a thermodynamic statistical ensemble . All other things being equal , places where nucleosomes bind strongly may be places where TFs are less likely to successfully compete , and , conversely , places where TFs bind strongly may be places where nucleosomes are less likely to successfully compete . In this manner , a high probability of nucleosome occupancy suggests that a TF binding site is less likely . We show that while nucleosome occupancy used as a simple positional prior only marginally improves the performance of a motif discovery algorithm , when it is used to compute a discriminative prior—taking into account accessibility over the whole genome—the accuracy of motif discovery improves dramatically . In situations where no nucleosome occupancy information is available , the prior D℘N℘ simplifies to a new kind of informative prior that can exploit discriminative information from the bound and unbound sequences in a purely generative setting . The prior performs admirably , finding 30% more true motifs than the uniform prior . The use of unbound sequences has previously been shown to improve both enumerative and probabilistic motif discovery approaches . Enumerative discriminative approaches compute the significance of the enrichment of every W-mer in the bound versus the unbound set using hypergeometric [27] or binomial distributions [28 , 29] . These methods are fast , but they usually work better when the TF binding sites have limited sequence variability [30] . Probabilistic approaches [31–35] attempt to learn the parameters of a discriminative motif that appears often in the bound set but less often in the unbound set . Since these discriminative sequence models try to distinguish between bound and unbound sets , they must traverse an enormous search space and become hampered by many local optima . In addition , at every step of the search algorithm , they have to evaluate the parameters of the model on each sequence in both sets . In contrast , while our prior D℘N℘ is calculated in a discriminative manner , the motif discovery problem itself remains formulated in a generative setting . Consequently , PRIORITY-D℘N℘ only needs to sample over the bound set , causing the overall time and space complexities of the search to be much less than those of other discriminative approaches ( even for the largest sequence-set Cbf1_SM with 194 sequences , PRIORITY-D℘N℘ takes fewer than four minutes on a desktop machine with a 2 . 4 GHz Intel Core2 CPU ) . Our discriminative approach can be viewed as a combination of both enumerative and probabilistic learning: the prior is primarily computed using “word counts” over bound and unbound sets , while the actual motif discovery is carried out using Gibbs sampling to optimize a posterior distribution . Our final motif retains the discriminative information through the prior contribution to the posterior objective function . Also , our discriminative approach is general enough to handle not only nucleosome occupancy information , but other kinds of biological data such as conservation , local DNA structure , etc . Throughout the paper , we have used PSSMs to model motifs . Although the PSSM is a popular choice for a motif model , recent biological [36] and computational [37 , 38] findings indicate that more expressive ( and hence , more complex ) models might be more appropriate . Since our method assigns a prior on the locations within each sequence and not on any specific form of the motif model , it is not tied to the PSSM model , but can be used with any motif model . In addition , although we have focused on ChIP-chip data here , both our priors N℘ and D℘N℘ can be computed for data resulting from other large-scale experimental methodologies such as gene expression , PBM , and DIP-chip microarrays . In closing , we stress that incorporating informative priors over sequence positions is of great benefit to motif discovery algorithms . Low signal-to-noise ratio , especially in higher organisms , makes it difficult to successfully use algorithms based only on statistical overrepresentation . Narlikar et al . [14] have shown that using informative priors based on structural classes of TFs improves motif discovery , and this paper shows that other kinds of informative priors improve motif discovery as well . Although PRIORITY-U℘ performs better than AlignACE and MEME , it falls short of the other four programs described earlier which use additional information like p-values or sequence conservation , illustrating the general utility of additional information in motif discovery . Additionally , although PRIORITY-D℘N℘ does better overall than these conservation-based methods , certain motifs are found by one or more of these methods but not by PRIORITY-D℘N℘ ( Table S1 ) . This suggests that combining conservation and nucleosome occupancy might further improve the performance of motif finders .
We compiled ChIP-chip data published by Harbison et al . [2] , who profiled the intergenic binding locations of 203 yeast TFs under various environmental conditions: always YPD ( rich medium ) and sometimes one or more of Acid ( acidic medium ) , Alpha ( alpha factor pheromone treatment ) , BUT14 ( butanol treatment for 14 h ) , BUT90 ( butanol treatment for 90 min ) , GAL ( galactose medium ) , H202Hi ( highly hyperoxic ) , H202Lo ( mildly hyperoxic ) , HEAT ( elevated temperature ) , Pi− ( phosphate deprived medium ) , RAFF ( raffinose medium ) , RAPA ( nutrient deprived ) , SM ( amino acid starvation ) , or THI− ( vitamin deprived ) over 6 , 140 intergenic regions . For each TF , we define its sequence-set X for a particular condition to be those intergenic sequences reported to be bound with p-value < 0 . 001 in that condition . We denote the set of all other sequences , those that are bound by that TF with a higher p-value , as the unbound set Y . Each sequence-set X is represented as TF_ condition . We restrict our attention to sequence-sets of size at least 10 , which yields 238 sequence-sets , encompassing 147 TFs . Of these sequence-sets , 156 correspond to the 80 TFs with a consensus binding motif in the literature ( as summarized by Harbison et al . at the time their paper was published , or as earlier reported by Dorrington and Cooper [39] or Jia et al . [40] ) , and these 156 are used throughout the paper to compare the performance of various motif-finding algorithms . The remaining 82 sequence-sets , corresponding to 67 TFs with unknown binding motifs , are used to make novel motif predictions . Assume the profiled TF is reported to bind a sequence-set X containing n DNA sequences X1 to Xn . Although in reality each bound sequence might have multiple binding sites , we model only one binding site in each sequence for simplicity . Because the experimental data might be erroneous , we also model the possibility that some sequences have no binding site . This is analogous to the zero or one occurrence per sequence ( ZOOPS ) model in MEME [21] . Let Z be a vector of length n denoting the starting location of the binding site in each sequence: Zi = j if there is a binding site starting at location j in Xi and we adopt the convention that Zi = 0 if there is no binding site in Xi . We assume that the TF motif can be modeled as a PSSM of length W parameterized by φ while the rest of the sequence follows some background model parameterized by φ0 . We present results here for W set to 8 . We wish to find φ and Z that maximize the joint posterior distribution of all the unknowns given the data . Assuming two independent priors P ( φ ) and P ( Z ) over φ and Z , respectively , our objective function is: We use Gibbs sampling to sample repeatedly from the posterior over φ and Z so that we are likely to visit those values of φ and Z with the highest posterior probability ( see Protocol S1 ) . We run the Gibbs sampler , which we call PRIORITY [14] , for a predetermined number of iterations after apparent convergence to the joint posterior and output the highest-scoring PSSM at the end . Although PRIORITY generates a posterior sample which is useful for other analyses in the style of MCMC , here we use only the single best motif ϕ to evaluate the algorithm and compare it with other popular methods . The source code of PRIORITY and the data used in the paper can be downloaded from http://www . cs . duke . edu/~amink . The prior on the positions P ( Z ) in Equation 1 is assumed to be uniform in conventional motif discovery algorithms . We call such a prior U℘ . Here , we discuss two informative positional priors based on nucleosome occupancy information . We assume we have this information as O℘ ( S , j ) : the probability of the jth position in sequence S being occupied by a nucleosome . Simple nucleosome prior N℘ . We use O℘ ( S , j ) to compute a simple nucleosome score S℘N℘ ( Xi , j ) for each W-mer starting at position j in the bound sequence Xi: We use this score to compute a positional prior N℘ which can be used in motif discovery . Note that the values S℘N℘ ( Xi , j ) themselves do not define a probability distribution over j . S℘N℘ ( Xi , j ) is only the probability that the W-mer at location j in Xi is a binding site of the profiled TF . As mentioned earlier , we model each sequence Xi as containing at most one such binding site . If Xi has no such binding site , none of the positions of Xi can be the starting location of such a binding site , so it must be that: where Li is the length of sequence Xi . On the other hand , if Xi has one such binding site at position j , not only must a binding site start at location j but also no such binding site should start at any of the other locations in Xi . Formally , we write: We then normalize P ( Zi ) using the same proportionality constant in Equations 3 and 4 , so that under the assumptions of our model we have: Discriminative nucleosome prior D℘N℘ . In addition to DNA sequences X , which are bound by the profiled TF , genome-wide ChIP-chip experiments also produce DNA sequences not bound by the TF . Assume we get m such sequences Y1 to Ym . We compute a discriminative nucleosome score S℘D℘N℘ ( Xi , j ) by taking into account the occupancies O℘ over both sets X and Y . For each W-mer in X , we ask the following question: “Of all the accessible occurrences of this W-mer , what fraction occur in the bound set ? ” The motivation behind this is to ensure a high score for W-mers that are accessible only in the bound set but not for W-mers that are accessible in general throughout the genome . To answer this question , we subject each accessible W-mer to a Bernoulli trial . Since we only know the probability that a certain location is accessible , we count the number of accessible W-mers in expectation , weighing each occurrence of the W-mer according to how accessible it is . Using the S℘N℘ scores derived from O℘ over both sets X and Y , we calculate S℘D℘N℘ ( Xi , j ) as: where is the W-mer starting at location j in sequence Xi . As in the case of S℘N℘ ( Xi , j ) , S℘D℘N℘ ( Xi , j ) is only the probability that the W-mer is a binding site of the profiled TF . To convert these values into a positional prior , we substitute S℘D℘N℘ for S℘N℘ in Equations 3 and 4 . After normalizing the resulting P ( Zi ) as in Equation 5 , we get the positional prior D℘N℘ . Predictions from computational model . We applied the computational model learned by Segal et al . [12] over the whole yeast genome ( March 2006 version ) . We used the resulting nucleosome occupancy predictions directly as O℘ ( S , j ) for each position j in an intergenic sequence S . Low-resolution in vivo data . We used the whole-genome ChIP-chip results for Myc-tagged H4 and H3 published by Lee et al . [9] . We used the median H4 intensity ratios ( the authors obtained nearly identical results for H3 and H4 ) which range from −1 . 757 ( least occupied ) to 1 . 112 ( most occupied ) and converted them to probabilities using a logit transformation to get occupancy O℘: where I ( S ) is the log ratio of intensities ( H4-Myc ChIP versus input genomic DNA ) , and λ is the logit parameter . We tried three different values of λ ( 1 , 4 , and 10 ) and noted results did not change significantly . Here , we report the best results , obtained with λ = 10 . We call the variant of S℘D℘N℘ computed with the low-resolution data S℘D℘N℘′ , and the prior derived from it D℘N℘′ . Note that the S℘N℘ derived from this data is the same over all positions within a sequence , and thus not very informative . We therefore present results of only the D℘N℘′ prior here . | Identifying transcription factor ( TF ) binding sites across the genome is an important problem in molecular biology . Large-scale discovery of TF binding sites is usually carried out by searching for short DNA patterns that appear often within promoter regions of genes that are known to be co-bound by a TF . In such problems , promoters have traditionally been treated as strings of nucleotide bases in which TF binding sites are assumed to be equally likely to occur at any position . In vivo , however , TFs localize to DNA binding sites as part of a complicated thermodynamic process of cooperativity and competition , both with one another and , importantly , with DNA packaging proteins called nucleosomes . In particular , TFs are more likely to bind DNA at sites that are not occupied by nucleosomes . In this paper , we show that it is possible to incorporate knowledge of the nucleosome landscape across the genome to aid binding site discovery; indeed , our algorithm incorporating nucleosome occupancy information is significantly more accurate than conventional methods . We use our algorithm to generate a condition-dependent , nucleosome-guided map of binding sites for 55 TFs in yeast . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"saccharomyces",
"computational",
"biology"
] | 2007 | A Nucleosome-Guided Map of Transcription Factor Binding Sites in Yeast |
Tsutsugamushi disease is an infectious disease transmitted to humans through the bite of the Orientia tsutsugamushi-infected chigger mite; however , host-pathogen interactions and the precise mechanisms of damage in O . tsutsugamushi infections have not been fully elucidated . Here , we analyzed the global metabolic effects of O . tsutsugamushi infection on the host using 1H-NMR and UPLC-Q-TOF mass spectroscopy coupled with multivariate statistical analysis . In addition , the effect of O . tsutsugamushi infection on metabolite concentrations over time was analyzed by two-way ANOVAs . Orthogonal partial least squares-discriminant analysis ( OPLS-DA ) showed distinct metabolic patterns between control and O . tsutsugamushi-infected mice in liver , spleen , and serum samples . O . tsutsugamushi infection caused decreased energy production and deficiencies in both remethylation sources and glutathione . In addition , O . tsutsugamushi infection accelerated uncommon energy production pathways ( i . e . , excess fatty acid and protein oxidation ) in host body . Infection resulted in an enlarged spleen with distinct phospholipid and amino acid characteristics . This study suggests that metabolite profiling of multiple organ tissues and serum could provide insight into global metabolic changes and mechanisms of pathology in O . tsutsugamushi-infected hosts .
Orientia tsutsugamushi , a mite-borne disease , is the causative agent of scrub typhus ( tsutsugamushi disease ) , the most prevalent febrile illness in the Asia-Pacific region [1] , [2] . This disease is characterized by fever , exanthematous rash , eschar , pneumonitis , meningitis , myalgia , and diffuse lymphadenopathy , symptoms similar to those of other acute febrile illnesses such as murine typhus , dengue fever , and viral hemorrhagic fevers . Delayed or inappropriate treatment can lead to severe multi-organ failure [3] , [4] . The pathological characteristics of O . tsutsugamushi-infection have been described in the literature [5]–[10] . A previous study [8] reported the development of an acute febrile illness within 8–10 days of bites from the larva of trombiculid mites ( which carry O . tsutsugamushi in their salivary glands ) , with bacteremia present 1–3 days before the onset of fever . When mice were infected intraperitoneally , O . tsutsugamushi was observed in smears of peritoneal exudates , liver , spleen , kidney and lung , and splenomegaly and peritonitis were evident [5] , [9] , [10] . The extent of pathological change in host organs has been well documented by several studies , but the precise mechanism of damage caused by O . tsutsugamushi infection remains unclear . Additionally , the host-pathogen interaction has not been clearly defined . Thus , novel approaches are required to explore the pathogenesis of O . tsutsugamushi . Metabolomics is the study of systemic biochemical profiles by analysis of biofluids and tissues . Recent metabolomic investigations have sought to elucidate the nature of host-parasite interactions . Several studies have examined the effects of infection by Mycobacterium tuberculosis , Plasmodium falciparum and Salmonella using various analytic platforms [11]–[14] , revealing the systemic effects of parasite diversion of nutrients on host metabolism . In particular , the application of several analytical platforms to complement the strengths and weaknesses of different types of equipment has been emphasized . For example , 1H NMR demonstrates high reproducibility and provides detailed information regarding the quantity and identity of metabolites , but shows low sensitivity and is not suitable for lipid profiling . Conversely , ultra-performance liquid chromatography-mass spectrometry ( UPLC-MS ) is highly sensitive and has a high capacity for analysis of diverse chemical characteristics , including lipid species [15] , [16] . Thus , many metabolomic studies utilized NMR and LC-MS in a complementary fashion [17] , [18] . In the present study , we used 1H-NMR and UPLC-Q-TOF mass spectroscopy-based metabolic profiling to characterize the responses of BALB/c mice to O . tsutsugamushi Karp infection . The metabolic changes in various organs and serum for sham control and O . tsutsugamushi-infected mice were also investigated to explore host-pathogen interactions in an O . tsutsugamushi-infected host system .
L929 , a mouse fibroblast cell line , was obtained from the American Type Culture Collection ( Rockville , MD . ) . L929 was cultured in Dulbecco's modified Eagle's medium ( DMEM; Gibco BRL ) supplemented with 5% fetal bovine serum ( Gibco BRL ) , 5 mM L-glutamine , penicillin ( 100 U/ml ) , and streptomycin ( 100 µg/ml ) in a humidified atmosphere containing 5% CO2 . The Karp strain of O . tsutsugamushi was propagated in monolayers of L929 cells , as described previously with slight modifications [19] . When more than 90% of the cells were infected , as determined by an indirect immunofluorescence antibody assay ( IFA ) technique [2] , cells were collected , homogenized using a glass Dounce homogenizer ( Wheaton , Inc . ) , and centrifuged at 500×g for 5 min . The supernatants were stored in liquid nitrogen until use . The titers of inocula were determined as follows: the bacterial stock was serially diluted and inoculated onto L929 cell layers in a 24-well tissue culture plate containing 12-mm diameter glass cover slip . After the cells were infected with O . tsutsugamushi for 4 h in a humidified 5% CO2 atmosphere at 34°C , the culture medium was removed . The cells were washed with phosphate-buffered saline ( PBS ) , fixed in 100% acetone for 10 min at −20°C , and stained by IFA . The number of infected-cell-counting units ( ICU ) of O . tsutsugamushi was determined by fluorescence microscopy [20] . Six-week-old female BALB/c inbred mice ( ORIENT BIO ) , weighing 17 to 18 g , were used throughout the study . The test mice were inoculated intraperitoneally ( i . p . ) with 3 . 5×104 ICU of O . tsutsugamushi Karp in 200 µl of PBS . After inoculation , mice were randomly separated into four groups of 12 and monitored for seven days ( one control mouse dropped out during the experiment ) . The mice were observed at least twice daily for mortality , morbidity , and body weight . To identify changes in metabolite levels in the liver , spleen and blood , mice were sacrificed at 4 and 7 days post-infection . After measuring the length and weight of the spleen and liver , respectively , they were stored in liquid nitrogen until use . Blood samples were collected into a plain blood correction tube which does not contain any anticoagulant . Then , immediately after separation from blood by centrifugation , serum samples were stored in liquid nitrogen until use . The mice were housed in animal biosafety level three facilities , where they received water and food ad libitum . Approval for animal experiments was obtained from the institutional animal welfare committee ( KCDC-023-11 ) . Liver and spleen tissues were rapidly frozen in liquid nitrogen and stored at −80°C prior to NMR analysis . For metabolite extraction , each tissue sample ( 100 mg of spleen or 200 mg of liver ) was placed into a 1 . 5-mL tube containing 2 . 8-mm zirconium oxide beads and homogenized twice at 5000 rpm with 350 µL of methanol ( d4 ) and 150 µL of 0 . 2 M ( pH 7 . 0±0 . 1 ) sodium phosphate buffer for 20 s using a Precellys 24 tissue grinder ( Bertin Technologies , Ampère Montigny-le-Bretonneux , France ) . After homogenization , 210 µL of methanol-d4 , 90 µL of 0 . 2 M ( pH 7 . 0±0 . 1 ) sodium phosphate buffer and 400 µL of chloroform were added to the tube . The mixture was then vortexed vigorously for 1 min and allowed to separate for 15 min . Samples were centrifuged at 13000 rpm for 10 min at 4°C . The upper layer was transferred as 630-µL aliquots to new 1 . 5 mL Eppendorf tubes and mixed with 70 µL of 2 . 5 mM TSP ( trimethylsilyl propionate ) dissolved in D2O . The mixture was then centrifuged at 13000 rpm for 5 min . Supernatants ( 600 µL ) were transferred to 5-mm NMR tubes . Serum samples were collected and stored at −80°C until NMR analysis . Prior to NMR , frozen serum samples were thawed at room temperature and vortexed . Samples ( 100 µL serum with 200 µL saline ) were transferred into 5-mm NMR outer tubes and 260 µL of 0 . 5 mM DSS [3- ( trimethylsilyl ) -1-propanesulfonic acid sodium salt] dissolved in D2O were transferred into the inner tube [21] . Three serum samples were not used for NMR analysis due to insufficient sample volume . 1H NMR spectra at 298 K were acquired using a VNMRS-600 MHz NMR spectrometer ( Agilent Technologies Inc . , Santa Clara , CA ) with a triple-resonance HCN salt-tolerant cold probe . For spleen and liver tissue samples , a one-dimensional ( 1D ) NOESYPRESAT pulse sequence was applied to suppress the residual water signal . The spectrum was collected with 64 transients into 33 , 784 data points using a spectral width of 8445 . 9 Hz , relaxation delay of 2 . 0 s , an acquisition time of 4 . 0 s , and a mixing time of 100 ms . For serum samples , the water-suppressed Carr–Purcell–Meiboom–Gill ( CPMG ) spin-echo pulse sequence ( RD-90°- [τ-180°-τ] n-ACQ ) was used to attenuate broad signals from proteins and lipoproteins . The 1H-NMR spectrum was collected with 128 transients into 32 K data points using a spectral width of 12019 . 2 Hz with a relaxation delay of 2 . 0 s and an acquisition time of 2 . 662 s . The individual CPMG spin echo ( in total 0 . 8 ms ) was repeated 80 times , resulting in a total spin-spin relaxation delay of 64 ms . All NMR spectra were phased , baseline corrected and divided into 0 . 005 ppm bins using Chenomx NMR suite 7 . 1 ( Chenomx , Edmonton , AB , Canada ) software . The binning data were normalized with the total area of each spectrum by excluding the water resonance ( 4 . 19–5 . 21 , 4 . 88–4 . 67 and 4 . 88–6 . 67 ppm for serum , liver and spleen , respectively ) . Binning data files were imported into MATLAB ( R2008a , Mathworks , Inc . , 2008 ) , and all spectra were aligned using the correlation-optimized warping method , which shifted , stretched , and shrunk the segmented spectra to maximize the correlation coefficient between the segments of a spectrum and target spectrum ( user-defined segmentation ) , resulted in the good alignment in single peaks as well as broad misalignment areas with minimizing the loss of resolution [22] . The resulting data sets were imported into SIMCA-P version 12 . 0 ( Umetrics , Umeå , Sweden ) for multivariate analysis , and Pareto-scaled [23] . Orthogonal partial least squares- discriminant analysis ( OPLS-DA ) was conducted , and the quality of each model was determined based on the goodness of fit parameter ( R2 ) , and goodness of prediction parameter ( Q2 ) [24] . In addition , reliability of each OPLS-DA model was validated by CV-ANOVA test [25] . NMR spectral data analysis was accomplished using targeted profiling with Chenomx NMR Suite 7 . 1 , and concentrations were determined using the 600 MHz library from Chenomx NMR Suite 7 . 1 , which compares the integral of a known reference signal ( DSS , TSP ) with signals derived from a library of compounds containing chemical shifts and peak multiplicities . The ambiguous peaks caused by overlap or slight shifts were confirmed by spiking samples with standard compounds and utilized two-dimensional ( 2D ) total correlation spectroscopy ( TOCSY ) and correlation spectroscopy ( COSY ) experiments for liver and spleen samples ( S1 Fig . ) . Spleen lipid metabolites were extracted with a chloroform: methanol mixture ( 2∶1 , v/v ) , as described by Folch et al . [26] . Lipid extracts were diluted with an isopropanol: acetonitrile: water mixture ( 2∶1∶1 , v/v/v ) and 10 µl was used for LC/MS analysis . LC-MS analysis was performed on a triple TOF™ 5600 MS/MS system ( AB SCIEX , Concord , Canada ) combined with an Ultra-performance Liquid Chromatography system ( Waters , Milford , MA ) . Separations were accomplished on an Acquity UPLC HSS C18 column ( 2 . 1×100 mm ) with 1 . 8-µm particles ( Waters , Milford , MA ) . A binary gradient system consisting of acetonitrile and water ( 4∶6 , v/v ) with 10 mM ammonium acetate was used as eluent A . Eluent B consisted of acetonitrile and isopropanol ( 1∶9 , v/v ) with 10 mM ammonium acetate . The gradient profile was 40% B at 0–3 . 5 min , 70% B at 8 . 5 min , 75% B at 12 min , 99% B at 12–13 min , and 40% B at 13 . 1–15 min . The flow rate was kept at 0 . 35 mL/min for 15 min . The mass spectrometer was operated in electrospray ionization ( ESI ) positive and negative modes , and the mass range was set at m/z 100–1500 . The following parameter settings were used: ion spray voltage of 5500 V , temperature of 500°C , curtain gas of 30 psi , declustering potential of 90 V , and collision energy of 10 V . In addition , information-dependent acquisition ( IDA ) was used to trigger acquisition of MS/MS spectra for ions matching the IDA criteria . MS/MS experiments were performed with collision energies of 40 V and collision energy spreads of 15 V . Mass accuracy was maintained by the automated calibrant delivery system ( AB Sciex , Concord , Canada ) interfaced to the second inlet of the DuoSpray source . Experiments were performed in duplicate , and percent relative standard deviation ( %RSD ) for duplicate analysis of homogenized samples was used to measure precision . All MS data , including retention times , m/z , and ion intensities , were extracted using the Markerview software ( AB SCIEX , Concord , Canada ) incorporated in the instrument , and the resulting MS data were assembled into a matrix . Ion intensity was normalized with the internal standard ( 1-heptadecanoyl-2-myristoleoyl-sn-glycero-3-phosphocholine ) , and the mean of duplicate results , showing <10% RSD , was utilized in statistical analysis . Metabolites were identified using the lipidmap ( www . lipidmaps . org ) and human metabolome ( www . hmdb . ca ) databases , and confirmed using standard compounds based on both retention times and mass spectra . Activities of hydroxymethylglutaryl CoA reductase ( HMG-CoAR ) and 3-hydroxybutyrate dehydrogenase were measured using ELISA kits for mouse HMG–CoAR and mouse 3-hydroxybutyrate dehydrogenase ( BDH2 ) ( Cusabio Biotech Co . Ltd , Wuhan , China ) , respectively , according to the manufacturer's instructions , and enzyme activities were measured in duplicate . A two-way ANOVA combined with Bonferroni posttests was performed using GraphPad PRISM version 5 . 0 ( GraphPad Software , Inc . , La Jolla , CA ) to analyze the influence of time and infection on metabolite concentrations , body weight , and spleen length . All enzyme activities including AST , ALT , HMG–CoA reductase and 3-hydroxybutyrate dehydrogenase were subjected to Mann-Whitney U test using SPSS 15 . 0 for Windows ( SPSS , Chicago , IL ) . Statistical significance was set at P<0 . 05 .
By day 3 after inoculation , several clinical signs of infection such as ruffled hair , slowed movement , and body weight changes were evident in O . tsutsugamushi-infected mice . Compared with control mice , the body weights of O . tsutsugamushi-infected mice decreased by 3 . 7% and 13 . 2% on days 4 and 7 after infection , respectively ( Fig . 1a ) , and the spleens of infected mice were significantly longer than those of control mice on day 7 ( Fig . 1b ) . Levels of ALT , an indicator of liver dysfunction , also increased after infection ( Fig . 1c and d ) . Representative 600-MHz 1H-NMR spectra of liver , spleen , and serum samples obtained from control and O . tsutsugamushi-infected mice are shown in S2-S4 Figs . We used these spectra to identify endogenous metabolites , and 19 , 17 and 15 metabolites in liver , spleen and serum , respectively , were identified by 2D TOCSY spectra and/or spiking experiments ( Tables 1–3 ) . To identify differences in the metabolite levels between control and O . tsutsugamushi-infected mice , orthogonal partial least squares- discriminant analysis ( OPLS-DA ) was applied to the NMR spectra of the liver , spleen , and serum samples . OPLS-DA score plots ( Fig . 2 ) showed clear differences in the liver , spleen , and serum samples of control and infected mice , indicating significant changes in the metabolism of O . tsutsugamushi-infected mice . R2Y and Q2Y values of the OPLS-DA models for the liver , spleen , and serum samples were 0 . 885 and 0 . 624 , 0 . 900 and 0 . 486 , and 0 . 630 and 0 . 331 , respectively . In addition , all OPLS-DA models had p values lower than 0 . 05 ( p<0 . 001 [liver] , p<0 . 001 [spleen] , and p = 0 . 02 [serum] ) in CV-ANOVA test . O . tsutsugamushi-infected groups also showed significant differences in metabolite levels between days 4 and 7 , but a change in control mice was not apparent , indicating that progression of O . tsutsugamushi infection had a marked effect on the metabolism of mice compared to the time-dependent metabolic changes that occur during natural aging . Two-way ANOVAs were applied to investigate the effect of O . tsutsugamushi infection and time on metabolite concentrations in liver tissues . Among the 19 metabolites detected in liver tissues , levels of 15 metabolites ( isoleucine , leucine , valine , 3-hydroxybutyrate , lactate , alanine , glutamate , glutathione , glucose , choline , O-phosphocholine , betaine , glycine , glycerol and nicotinurate ) differed significantly between control and O . tsutsugamushi-infected mice . Furthermore , betaine , glutathione , and O-phosphocholine showed an interaction ( infection × time ) effect . We found that isoleucine , leucine , valine , lactate , alanine , glutamate , glutathione , glucose , choline , O-phosphocholine , betaine , glycine , glycerol , and nicotinurate levels decreased , while 3-hydroxybutyrate and O-phosphocholine levels increased . The significant decrease in the majority of metabolite levels indicates that O . tsutsugamushi infection affected the liver ( Table 1 ) , and all significantly altered metabolites except nicotinurate also showed high values of weight in OPLS-DA model ( w>|0 . 04| , S5 Fig . ) . All metabolites ( choline , betaine and glycine ) associated with remethylation cycles providing substrates for transmethylation were significantly decreased , and the antioxidant metabolite ( glutathione ) was also decreased on days 4 and 7 post-infection ( Fig . 3 ) . In the spleen , 10 metabolites ( isoleucine , leucine , valine , isopropanol , 3-hydroxybutyrate , lactate , choline , O-phosphocholine , O-phosphoethanolamine and betaine ) were affected by infection . Betaine , choline , isopropanol , O-phosphocholine , O-phosphoethanolamine , taurine and threonine showed interaction ( infection × time ) effects . We found that isoleucine , leucine , valine , isopropanol , 3-hydroxybutyrate , lactate , choline and O-phosphoethanolamine levels increased , while those of O-phosphocholine and betaine decreased after infection ( Table 2 ) , and all significantly altered metabolites also showed high values of weight in OPLS-DA model ( w>|0 . 04| , S6 Fig . ) . In sera , 7 metabolites ( 3-hydroxybutyrate , lactate , pyruvate , citrate , creatine , glucose and betaine ) were affected by infection . 3 metabolites ( pyruvate , citrate and creatine ) showed interaction ( infection × time ) effects . We found that levels of lactate , pyruvate , citrate , glucose and betaine decreased significantly post-infection , while those of 3-hydroxybutyrate and creatine increased ( Table 3 ) , and all significantly altered metabolites also showed high values of weight in multivariate analysis ( w>|0 . 04| , S7 Fig . ) . Interestingly , among the metabolites whose levels were decreased in the liver and/or serum of O . tsutsugamushi-infected mice , several metabolites ( glucose , lactate , pyruvate , glycerol , and citrate ) are associated with glycolysis and the TCA cycle . In addition , levels of ketone bodies and 3-hydroxybutyrate were significantly elevated in the liver and serum of infected mice ( Fig . 4a and b ) . This indicates that energy source utilization was significantly affected by O . tsutsugamushi infection . To identify the metabolic pathways affected by infection , we measured the activities of HMG–CoA reductase and 3-hydroxybutyrate dehydrogenase ( Fig . 4c ) . In O . tsutsugamushi-infected mice , the activities of both enzymes were significantly increased , indication beta oxidation; i . e . , breakdown of fatty acids to produce energy . Nicotinurate was also significantly decreased in the liver of O . tsutsugamushi-infected mice ( Fig . 4d ) . In NMR data for the spleen tissues , we observed significant alterations of choline , O-phosphocholine , and O-phosphoethanolamin , which are involved in phosphatidylcholine ( PC ) and phosphatidylethanolamine ( PE ) metabolism . Thus , we further analyzed PC , PE and lipid metabolites using UPLC-MS . In O . tsutsugamushi-infected mice , long-chain fatty acid levels , PC and PE levels , PC/PE ratios , and polyunsaturated fatty acid levels in the spleen ( 18∶2 ) were all significantly increased when compared to control mice ( Fig . 5 , S1 and S2 Tables ) .
To explore host–O . tsutsugamushi interactions , we first applied NMR-based metabolic profiling to the liver , spleen and serum of O . tsutsugamushi-infected mice . At 5 days post-infection , we observed significantly reduced body weight , splenomegaly , and elevated levels of AST and ALT in O . tsutsugamushi-infected mice . PLS-DA analysis showed that O . tsutsugamushi infection particularly affected liver metabolism . Significantly elevated ALT levels post-infection indicated hepatocyte destruction , leading to alteration of energy production by the host liver . Decreased levels of glycolysis and TCA cycle intermediate metabolites—such as glucose , lactate , pyruvate , glycerol and citrate—were also found in the serum and/or liver of O . tsutsugamushi-infected mice . Previous studies have suggested that O . tsutsugamushi organisms obtain the majority of their energy from sources other than sugar catabolism , due to a lack of sugar-phosphate transporters or hexokinase [27] , [28] . Consistent with this assumption , we found decreased levels of branched-chain-amino-acids ( BCAA; isoleucine , leucine and valine ) in the serum and liver of O . tsutsugamushi-infected mice . These changes were likely due to the higher energy efficiency of BCAA in conditions of liver dysfunction [29] . Min et al . stated that in O . tsutsugamushi , the TCA cycle begins with alpha-ketoglutarate , which is converted from glutamate likely obtained from the host cell . In this study , we found decreased glutamate levels in the livers of O . tsutsugamushi-infected mice . We observed excess beta-oxidation of fatty acids as evidenced by elevated levels of 3-hydroxybutyrate , HMG–CoA reductase , and 3-hydroxybutyrate dehydrogenase in O . tsutsugamushi-infected hosts . These results indicate that O . tsutsugamushi infection affected the energy metabolism of host mice , suppressing oxidation of carbohydrates by reducing the availability of substrates for glycolysis and the TCA cycle , and increasing fatty acid oxidation for energy generation . Unlike other Rickettsia , O . tsutsugamushi has no beta-oxidation system for fatty acid energy generation [30] . Therefore , the ketone body , 3-hydroxybutyrate , is obtained from the host during O . tsutsugamushi infection , unlike other infectious diseases [31] , and serves as a carbohydrate substitute during energy production . Nicotinuric acid is the major product of nicotinic acid metabolism , and serves as a simple quantitative index of hepatic biotransformation of nicotinic acid [32] . Reduced levels of nicotinuric acid are indicative of inhibited synthesis of nicotinate metabolites , such as NADH , NAD , NAD+ , and NADP , which play essential roles in energy metabolism and DNA repair [33] . In this study , we also observed decreased levels of all metabolites associated with folate-mediated one-carbon metabolism ( choline , betaine and glycine ) in the livers of O . tsutsugamushi-infected mice . Choline and betaine are particularly necessary for methionine synthesis in the homocysteine remethylation process; methionine subsequently provides methyl groups in transmethylation reactions [34]–[37] . Therefore , transmethylation in the liver is affected by O . tsutsugamushi infection , although O . tsutsugamushi contains only a subset of the genes necessary for folate metabolism [27] . Methylation defects found in HIV-infected patients are thought to be linked to neurological damage and immune cell depletion [38]–[40] . Some reports have found that O . tsutsugamushi infection involves the central nervous system and may cause Guillain–Barré syndrome , and meningitis [41]–[43] , which may be attributed to the lack of methylation in O . tsutsugamushi-infected hosts reported that O . tsutsugamushi affected the host cell antioxidant system and free radical levels [44] . In agreement with this result , we found decreased levels of glutathione ( which plays an important role in defense against oxidative stress ) in the liver of O . tsutsugamushi-infected mice . O . tsutsugamushi infection resulted in splenomegaly and alteration of splenic components . A previous TEM study of O . tsutsugamushi-infected spleens found large numbers of phagosomes and phagolysosomes in the intracytoplasm of macrophages , swollen and decrepit red blood cells , and the presence of additional membranes enveloping O . tsutsugamushi organisms . Thus , the high levels of PC and PE observed in our study may be due to phagocytic destruction of erythrocytes [45] , [46] . We observed both increases and decreases in the levels of O-phosphocholine and O-phosphoethanolamine , PE and PC , long-chain fatty acids , and polyunsaturated fatty acids . These changes may be related to the breakdown of O . tsutsugamushi-enveloping membranes and/or spleen cell membranes in addition to breakdown of erythrocytes . In O . tsutsugamushi-infected mice , splenic enlargement may be related to increased BCAA levels and lymphocyte proliferation [47] , [48] . However , further studies are needed to definitively determine the mechanisms of lipid and BCAA changes in the O . tsutsugamushi-infected spleen . Global metabolite profiling of various organ tissues and serum provided insight into the effects of O . tsutsugamushi infection on host metabolism . O . tsutsugamushi-infected mice exhibited altered metabolism of the liver , as in energy source utilization , and the spleen . In conclusion , a global systems biology approach provides more information on host-pathogen interactions , and will likely facilitate efficient diagnosis and treatment of O . tsutsugamushi . | Scrub typhus is an acute febrile illness caused by attacks of Orientia tsutsugamushi-carrying mites , and is the most prevalent febrile illness in the Asia-Pacific region . If not properly treated with antibiotics , patients often develop severe vasculitis that affects multiple organs , and the mortality rate can reach 30% . To explore the pathogenic mechanisms underlying the host-pathogen interaction , we characterized metabolic changes in various organs and the serum of O . tsutsugamushi-infected hosts . After O . tsutsugamushi infection , the host experienced decreased energy production , as well as a severe deficiency in re-methylation sources and glutathione , which impaired purine synthesis , DNA and protein methylation . In addition , abnormal pathways for phosphatidylcholine ( PC ) biosynthesis and phosphoethanolamine methylation were utilized in the enlarged spleen of O . tsutsugamushi-infected hosts . These results suggested that metabolic profiling could provide insight into global metabolic changes in O . tsutsugamushi-infected hosts , and increase our understanding of the pathogenic mechanisms of O . tsutsugamushi , as well as providing novel therapeutic targets for scrub typhus . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"biochemistry",
"metabolomics",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"host-pathogen",
"interactions",
"biology",
"and",
"life",
"sciences",
"pathogenesis",
"metabolism"
] | 2015 | Metabolic Responses to Orientia tsutsugamushi Infection in a Mouse Model |
The spatial organization of chromosomes within interphase nuclei is important for gene expression and epigenetic inheritance . Although the extent of physical interaction between chromosomes and their degree of compaction varies during development and between different cell-types , it is unclear how regulation of chromosome interactions and compaction relate to spatial organization of genomes . Drosophila is an excellent model system for studying chromosomal interactions including homolog pairing . Recent work has shown that condensin II governs both interphase chromosome compaction and homolog pairing and condensin II activity is controlled by the turnover of its regulatory subunit Cap-H2 . Specifically , Cap-H2 is a target of the SCFSlimb E3 ubiquitin-ligase which down-regulates Cap-H2 in order to maintain homologous chromosome pairing , chromosome length and proper nuclear organization . Here , we identify Casein Kinase I alpha ( CK1α ) as an additional negative-regulator of Cap-H2 . CK1α-depletion stabilizes Cap-H2 protein and results in an accumulation of Cap-H2 on chromosomes . Similar to Slimb mutation , CK1α depletion in cultured cells , larval salivary gland , and nurse cells results in several condensin II-dependent phenotypes including dispersal of centromeres , interphase chromosome compaction , and chromosome unpairing . Moreover , CK1α loss-of-function mutations dominantly suppress condensin II mutant phenotypes in vivo . Thus , CK1α facilitates Cap-H2 destruction and modulates nuclear organization by attenuating chromatin localized Cap-H2 protein .
Interphase genome organization in eukaryotic cells is non-random [1 , 2 , 3] . Indeed , organization of the genome is crucial because it influences nuclear shape and processes such as DNA repair and replication , as well as gene expression [4 , 5 , 6] . While chromosomes are highly organized within the nucleus , they must also remain extremely dynamic . Chromosome dynamics facilitate events that occur not only during cell division , but also during interphase , when cells respond to developmental and environmental cues that require changes in gene expression . Interphase events include trans-interactions such as homolog pairing , chromosome remodeling and compaction , and DNA looping . Although numerous studies using Fluorescent In-Situ Hybridization ( FISH ) , live cell imaging , and chromosome conformation capture techniques have revealed the three-dimensional ( 3D ) organization of genomes , much remains to be discovered regarding the factors that govern the overall conformation of interphase chromosomes . An equally important task is to identify the molecular mechanisms that regulate and maintain specific 3D genome organizational states . Condensin complexes are highly conserved from bacteria to humans [7 , 8 , 9] and have been identified as key drivers of genome organization [10] . Eukaryotes have two condensin complexes , condensin I and II , which share the core SMC2 and SMC4 ( Structural Maintenance of Chromosomes ) subunits but differ in their non-SMC Chromosome Associated Protein ( CAP ) subunits . Condensins have long been known to play vital roles in shaping mitotic chromosomes . While condensin I promotes lateral chromosome compaction , condensin II promotes axial compaction; both of which are necessary for faithful mitotic condensation and chromosome segregation [11] . Condensins also display different localization patterns: condensin I only associates with mitotic chromosomes , whereas condensin II is present in the nucleus , where it is bound to chromatin throughout the cell cycle [12 , 13 , 14 , 15] In Drosophila cells , condensin II performs a variety of functions during interphase , including chromosome compaction , unpairing of homologous chromosomes , and driving the formation and maintenance of chromosome territories . The condensin II subunit , Cap-H2 , has been shown to be the rate limiting subunit , as overexpression of Cap-H2 results in the increase of condensin II chromosome activity [16 , 17] [18] [19 , 20] [21] . Furthermore , all of these Drosophila Cap-H2 functions were shown to either require or be dependent on one or more other condensin subunits , such as SMC4 , SMC2 and/or Cap-D3 . This strongly suggests that these Cap-H2 functions are likely to be performed in the context of an active condensin II complex . Drosophila condensin II also plays an essential function in anaphase-I of male meiosis , where it is thought to be required for resolving entanglements between homologs as well as heterologs [17] . Strikingly , loss or depletion of condensin II function leads to lengthening of interphase chromosomes , suggesting that chromosome axial compaction must be actively maintained even after exit from mitosis [21 , 22] . Additionally , Drosophila Cap-D3 regulates the expression of immunity genes while repressing somatic cell transposon activation , although it is unclear if these effects are linked to condensin mediated compaction activity [23 , 24] . In vertebrates , condensin II is required for axial compaction of chromosomes and sister chromatin resolution in S-phase [15 , 25] . Loss of the Cap-D3 condensin II subunit also leads to loss of compaction in S-phase [26] . That interphase chromosome compaction levels are important for gene expression has been recently highlighted in mammalian cells . Condensin II has been found to regulate the STAT5 transcription factor by highly condensing interphase chromatin during T-cell differentiation; failure to repress condensin II activity prevents STAT5 access to its binding sites , resulting in cells that are unable to differentiate normally [27] . Similarly , the bromodomain protein Brd4 , recruits condensin II to chromatin and local condensation serves to attenuate signaling from the damaged DNA sites [28] . Thus physical compaction of interphase chromatin can be an effective mechanism for limiting transcription factor binding as well as modulating important signaling . While it is evident that the role of condensin II in interphase genome organization , DNA repair , and gene expression is important in both vertebrate and Drosophila cells , how condensin II executes these functions is not well understood . In fact , it is unknown whether processes such as anti-pairing , gene regulation , and transposon repression are secondary effects to its potential primary role in shaping chromatin architecture . Moreover , the molecular mechanism regulating interphase condensin II remains unclear . Therefore , elucidating these regulatory mechanisms will lead to important insights into how interphase genome organization is modulated . Previously , it was demonstrated that overexpression of the condensin II regulatory subunit , Cap-H2 , was sufficient to drive both interphase compaction and the unpairing of homologous chromosomes in vivo and in cultured cells [16 , 20 , 21 , 22] . In contrast , mutations that inactivate other condensin II genes , by decreasing their dosage or depleting their expression by RNAi , suppress all Cap-H2 overexpression phenotypes . These observations strongly suggest that 1 ) Cap-H2-induced homolog anti-pairing and compaction is reliant on a functional condensin II complex , and 2 ) that Cap-H2 is the rate-limiting component in the activation of the catalytic SMC2/4 subunits , which are present during interphase and are able to bind chromatin . Thus , Cap-H2 availability controls condensin II activity , and consequently , Cap-H2 protein levels and chromatin localization represent key steps for SMC2/4 regulation in interphase . This idea is consistent with the observation that Cap-H2 loading onto chromatin is partially dependent on the chromodomain protein Mrg15 [21] . Moreover , Cap-H2 protein levels are controlled by the SCFSlimb ubiquitin-ligase , maintaining low levels of Cap-H2 in vivo and in cultured Drosophila cells [20] . Interestingly , Slimb recognizes its target proteins through a phosphodegron motif [29] , suggesting that one or more kinases must phosphorylate Cap-H2 before Slimb can target it for destruction . A Slimb-binding site consensus sequence ( DSGXXS ) exists in the extreme C-terminus of Cap-H2 and deletion of this region renders Cap-H2 non-degradable [20] . As expected for a Slimb substrate , Cap-H2 protein mobility on SDS-PAGE was sensitive to phosphatase treatment , suggesting that Cap-H2 is phosphorylated [20] . Given that Cap-H2 protein levels may be regulated by its phosphorylation state , we set out to identify kinases that target Cap-H2 for Slimb recognition and that lead to its degradation . We show that in Drosophila cultured S2 cells , Casein Kinase I alpha ( CK1α ) depletion results in the hypercondensation of interphase chromatin in a condensin II-dependent manner . We also found that CK1α and condensin II genetically interact in vivo , and that CK1α depletion leads to Cap-H2 protein enrichment on polytene and cultured cell chromosomes . Similar to Slimb depletion [20] , CK1α depletion also results in stabilization of Cap-H2 protein in cultured cells . Our findings further elucidate the mechanism by which Cap-H2 , and thus condensin II , is regulated and contribute significantly to our understanding of how interphase genome organization , homolog pairing , and chromosome compaction is modulated .
Previously , we discovered that the Cap-H2 subunit of condensin II is a SCFSlimb ubiquitination-target in Drosophila cells [20] . In a whole genome RNAi screen , Slimb was also identified as a homolog pairing-promoting factor , and it was shown to affect pairing in a Cap-H2 dependent manner[18] . In cultured S2 and Kc cells , depletion of SCFSlimb components Slimb , Cul-1 and SkpA prevents Cap-H2 degradation and leads to condensin II hyperactivation during interphase and the remodeling of each chromosome into a compact globular structure ( Fig . 1A-C ) . Based on their overall appearance , we refer to these hypercondensed chromosomes as “chromatin-gumballs” ( Fig . 1A ) . Overexpression of a GFP tagged wild type Cap-H2 also induces this phenotype [20] . Since phosphorylation of the Slimb-binding domain within its substrates is required for Slimb binding [29] , we reasoned that depletion of a kinase involved in this pathway would also stabilize Cap-H2 and phenocopy the effect on chromatin remodeling observed after Slimb depletion . We first analyzed S2 cells that were depleted of either: Glycogen Synthase Kinase 3 Beta ( GSK3β ) , Protein Kinase A ( PKA ) , or Casein Kinase I alpha ( CK1α ) . We chose these three kinases because they typically phosphorylate Slimb targets to initiate their degradation [30] [31 , 32] [33 , 34] . Strikingly , depletion of only CK1α resulted in a dramatic increase in the formation of chromatin-gumballs in interphase cells that was comparable to Slimb depletion ( Fig . 1F ) . CK1α depletion was verified indirectly by assessing Armadillo ( Drosophila β-catenin ) protein levels , as the use of CK1α antibodies to Drosophila CK1α [35] and commercially available anti-human CK1α were both unsuccessful . CK1α and Slimb function to negatively regulate Armadillo , therefore we probed Armadillo protein levels to confirm that CK1α and Slimb were efficiently being depleted in our RNAi treatments , as previously shown [32 , 34] . In addition to the S2 cells , identical gumball phenotypes were observed in CK1α-depleted Kc cells , as previously observed in Slimb depleted cells ( Fig . 1B–D ) . CK1α depletion in S2 cells induced chromatin-gumball formation ( weak and strong ) to levels significantly higher ( p < 7 . 5x10−8 ) than control-treated cells ( CK1α RNAi: 89 ±4%; control RNAi: 1 ±0 . 17%; sum percentage of weak and strong gumballs ) ( Fig . 1F ) . To rule out the possibility that the observed gumball phenotypes in the CK1α depleted cells were a result of apoptosis , we assessed cell viability in our Kc cell RNAi treatments . CK1α did not increase cell death over that of control treated cells ( control RNAi: 18 ±2 . 4%; CK1α RNAi: 16 . 7 ±3 . 9% ) ( S1A and B Fig . ) . In addition to RNAi , we also inactivated CK1α in cultured S2R+ cells with the cell permeable CK1 chemical inhibitor D4476 ( Fig . 1G-H ) [36 , 37] . We observed that D4476 significantly increased ( p < 0 . 01 ) the proportion of chromatin-gumball formation ( weak and strong ) compared to DMSO-treated control cells ( DMSO: 7 . 9 ±1 . 5% , D4476: 62 ±6%; sum percentage of weak and strong gumballs ) ( Fig . 1I ) . Thus , similar to SCFSlimb , CK1α depletion by RNAi or pharmacological inhibition with the D4476 leads to global chromatin remodeling . We next tested whether the CK1α RNAi-induced gumball phenotype is dependent on condensin II activity . To test this , we co-depleted both CK1α and the condensin II non-SMC subunits Cap-D3 or Cap-H2 in both Kc and S2 cells ( Fig . 1E-F ) . Co-depletion of CK1α with either subunits resulted in a significant suppression ( CK1α + Cap-D3 RNAi: p < 0 . 0005 , CK1α + Cap-H2 RNAi: p < 3 . 9x10−6; comparison between percentage of normal cells ) of the chromatin-gumball phenotype . Interestingly , CK1α co-depletion with Cap-D3 did not suppress the gumball phenotype as well as co-depletion with Cap-H2 ( CK1α + Cap-D3 RNAi: 62 ±2 . 3% normal nuclei , CK1α + Cap-H2 RNAi: 98 ±0 . 35% normal nuclei ) ( Fig . 1F ) . These observations suggest that in the absence of CK1α , condensin II-mediates hyper-compaction of interphase chromosomes . One function of condensin II in cultured cells and in vivo is to drive dispersal of centromeres , as visualized by immunostaining of Centromeric Identifier ( CID ) protein or by FISH to pericentric heterochromatin[20 , 22] . It has been proposed that centromere dispersal is a direct consequence of chromosome compaction [18 , 22] . The results from our RNAi experiments suggest that CK1α may be functioning to repress Cap-H2 , as co-depletion of CK1α with Cap-H2 strongly suppresses the gumball phenotype ( Fig . 1E-F ) . Stabilization of Cap-H2 protein levels , if functional , is expected to drive dispersal of centromeric regions and result in a greater number of CID foci in each nucleus . To test this hypothesis , cultured Drosophila cells ( Kc and S2 ) were depleted of CK1α via RNAi treatment and immunostained using an antibody specific to CID . The number of CID spots per nucleus was counted , with an increase in CID spots per nucleus indicating an increase in centromere dispersal . CID spots in control treated cells appear clustered , whereas CK1α depletion results in CID signal dispersal and a significant increase ( p < 3x10−6 ) in the number of CID spots ( for Kc cells , CK1α RNAi: 4 . 7 ±0 . 17 and Control RNAi: 3 . 6 ±0 . 13 spots per nucleus ) ( Fig . 2A , E , H-I ) . Furthermore , this increase in CID dispersal was suppressed when either condensin subunits SMC2 or Cap-H2 were co-depleted with CK1α ( CK1α + SMC2 RNAi: 3 . 9 ±0 . 16 and CK1α + Cap-H2 RNAi: 3 . 76 ±0 . 15 spots per nucleus ) ( Fig . 2F-I ) . In addition , co-depletion of the Condensin I specific subunit Barren ( Drosophila Cap-H ) with CK1α did not suppress the increase in CID dispersal ( CK1α + Barren RNAi: 4 . 8 ±0 . 18 spots per nucleus ) ( S2C Fig . ) . Similar to Slimb , CK1α acts as an inhibitor of condensin II mediated centromere dispersal ( Fig . 2D-E , H ) . This was also observed in S2 cells ( S2A Fig . and B ) . To exclude the possibility that the increases in CID dispersal may be explained by an increase in cell ploidy , DNA content in RNAi treated cells was analyzed by flow cytometry . Flow cytometry on S2 cells demonstrates that CK1α depletion slightly increases the proportion of cells in G1 ( CK1α RNAi: 51 . 5% and Control RNAi: 42 . 4% ) ( S3C Fig . ) , therefore , the increase in number of CID foci in CK1α RNAi cells is not due to increases in centromere numbers resulting from polyploidy . These results indicate that CK1α is normally acting to inhibit condensin II dependent centromere dispersal . In addition to promoting the dispersal of centromeric regions , Cap-H2 has been shown to be important for maintenance of interphase chromosome axial length [21 , 22] . If CK1α is a negative regulator of Cap-H2 , then CK1α depletion should lead to an increase in chromosome compaction and a decrease in axial length . To measure chromosome compaction , we performed 3D DNA FISH in RNAi treated cultured cells using three probes specific to euchromatic loci on the X chromosome ( Fig . 3 ) . FISH probes were designed approximately 2Mb apart . We found that CK1α depletion resulted in a significant decrease in pairwise 3D distances between FISH probes compared to control treated cells ( X1-X2 = p < 0 . 0004 , X1-X3 = p < 0 . 001 ) ( Fig . 3A , D , G ) . In control treated cells , the distance between X1 and X2 probes was 0 . 96 ±0 . 04μm and the distance between X1 and X3 probes was 1 . 08 ±0 . 05μm . CK1α depletion caused these distances to decrease about 20% to 0 . 76 ±0 . 05μm between X1 and X2 probes and 0 . 85 ±0 . 04μm between X1 and X3 probes . This increase in chromosome compaction resulting from depletion of CK1α suggests that CK1α normally antagonizes chromosome compaction . Interestingly , CK1α co-depletion with condensin subunits SMC2 or Cap-H2 increased the axial length of chromosomes , relative to control treated cells ( CK1α + SMC2 RNAi: X1-X2 = 1 . 5 ±0 . 1μm and X1-X3 = 1 . 4 ±0 . 07μm , CK1α + Cap-H2 RNAi: X1-X2: 1 . 4 ±0 . 1μm and X1-X3 = 1 . 7 ±0 . 1μm ) ( Fig . 3E-G ) . We noted that the axial chromosome length seen with co-depletion of CK1α with SMC2 or Cap-H2 was significantly increased compared to depletion of Cap-H2 or SMC2 alone ( p < 0 . 05 for X-chromosome probes X1-X2 and X1-X3 , Fig . 3G ) . It is unclear why co-depletion of CK1α and codensin II subunits would lead to the observed axial lengthening that is greater than in control cells . We considered the possibility that CK1α depletion led to a mitotic arrest phenotype where chromosome compaction status could be due to mitotic condensation . Control cells and CK1α depleted cells were stained with anti-lamin and the mitotic marker anti-phospho-histone H3 ( phospho-H3 ) to assess whether the RNAi treated cells were undergoing mitosis . We found that the number of phospho-H3 positive cells in CK1α depleted cells was not significantly increased compared to control cells . In fact , CK1α depleted cells showed a significant decrease in phospho-H3 positive cells ( CK1α RNAi = 0 . 71 ±0 . 12% , Control = 2 . 15 ±0 . 11%; p < 3 . 36x10−7 ) ( S3A Fig . and B ) . Furthermore , nuclear envelope staining by anti-lamin and DNA visualization by DAPI staining both indicated that the vast majority ( >95% ) of cells examined were in interphase . Lastly , flow cytometry measuring DNA content of RNAi treated S2 cells also demonstrated that there was only moderate differences in cell cycle profile of CK1α depleted and CK1α , Cap-H2 co-depleted cells compared to control RNAi treated cells ( S3C Fig . ) . Thus , these data demonstrating that CK1α functions to inhibit chromosome compaction in cultured cells provide strong evidence that CK1α limits interphase condensin II activity , and that these compaction differences cannot be explained by dramatic shifts in cell cycle distribution . Cap-H2 and other condensin II subunits have been shown to function as factors that antagonize homologous chromosome pairing during interphase [16 , 18 , 19] . In contrast , Slimb has been identified as a pairing-promoting factor that negatively regulates the Cap-H2 anti-pairing activity [18 , 20] . Therefore , we predicted that CK1α also is a pairing promoting factor not identified in previous chromosome pairing screens . To test this , we performed FISH in RNAi treated Kc cells using FISH probes designed to label two different euchromatic loci on the X chromosome . Homolog pairing can be assessed by counting the number of fluorescent spots per nucleus for each of the two probes . A single spot for each FISH probe signifies chromosomes are paired , whereas two spots signify unpairing of homologs . Normally , homologs are paired in most Drosophila cells , and in Kc cells at levels of 85% or greater [18 , 38] . CK1α depletion results in an increase in unpairing of chromosomes compared to control cells , evident by the significant increase ( p < 1 . 3x10−8 ) in number of FISH spots counted per nucleus ( control RNAi: X1 = 1 . 3 ±0 . 04 and X2 = 1 . 2 ±0 . 04 , CK1α RNAi: X1 = 2 . 5 ±0 . 16 and X2 = 2 ±0 . 11 spots per nucleus ) ( Fig . 3A , D , H and S2D Fig . ) . To test whether this increase in unpairing was condensin dependent , cells were treated with RNAi to both CK1α and Cap-H2 or CK1α and SMC2 ( Fig . 3E-F ) . Depletion of either Cap-H2 or SMC2 suppressed the increased unpairing caused by CK1α RNAi ( CK1α + Cap-H2 RNAi: X1 = 1 . 8 ±0 . 1 and X2 = 1 . 3 ±0 . 08 , CK1α + SMC2 RNAi: X1 = 1 . 6 ±0 . 08 and X2 = 1 . 4 ±0 . 08 spots per nucleus ) ( Fig . 3H ) . We also depleted the condensin I specific subunit , Barren ( Cap-H ) , along with CK1α and found that co-depletion did not significantly suppress the increased unpairing seen with CK1α RNAi ( CK1α + Barren RNAi: X1 = 2 . 3 ±0 . 1 and X2 = 2 ±0 . 1 spots per nucleus ) ( S2D Fig . and E ) . This indicates that adding a double-stranded RNA targeting CK1α and a second gene does not lead to suppression of anti-pairing activity simply by diluting the CK1α RNAi-depletion . Instead , this strongly suggests that the pairing promoting function of CK1α is likely due to its ability to repress condensin II anti-pairing activity . In addition to the X1 and X2 probes specific to euchromatic loci , FISH probes specific to heterochromatic sequences on the right arms of chromosomes 2 and 3 ( 2R and 3R ) were used to assess the pairing status of heterochromatin . Similar to the observations at the euchromatic loci , FISH using heterochromatin probes exhibited the same behavior , showing that CK1α promotes pairing at euchromatin and heterochromatin . Depletion of CK1α significantly increased ( p < 0 . 005 ) the number of spots per nucleus for each of the two heterochromatic FISH probes used , compared to control treated cells ( CK1α RNAi: 3R = 2 . 25 ±0 . 14 and 2R = 2 . 43 ±0 . 11 , control RNAi: 3R = 1 . 68 ±0 . 14 and 2R = 1 . 75 ±0 . 13 spots per nucleus ) ( Fig . 3I , K , M and S2F Fig . ) . This increase in heterochromatin unpairing was similar to that observed when an inducible GFP tagged version of full-length Cap-H2 was expressed in cells ( Cap-H2 O . E . : 3R = 2 . 18 ±0 . 1 , 2R = 2 . 23 ±0 . 11 spots per nucleus ) ( Fig . 3J , M and S2F Fig . ) . The increase in unpairing of both euchromatic and heterochromatic loci after CK1α depletion is consistent with what is observed when Cap-H2 activity is increased and provide further evidence that CK1α functions as a pairing promoting factor by antagonizing condensin II anti-pairing function . Based on the increased level of chromosome unpairing observed in CK1α depleted cultured cells , we wanted to test whether CK1α is repressing condensin II mediated chromosome organization in vivo . In order to do this , we turned to the salivary glands of Drosophila third instar larvae . Salivary gland nuclei contain polytene chromosomes that are highly paired . We have previously shown that overexpression of Cap-H2 is sufficient for driving unpairing of homologs and sister chromatids that are normally paired into polytenes chromosomes[16] , and Slimb-RNAi depletion resulted in the disruption of polytene chromosomes[20] . If CK1α is negatively regulating Cap-H2 in vivo , then RNAi depletion of CK1α is expected to also drive polytene disruption , as has been shown for Slimb RNAi in the salivary gland . To test this , we performed FISH on whole mount salivary glands from third instar larvae , using the number of FISH spots as a direct measure of chromatid and homolog pairing ( increased FISH spots indicates increased unpairing ) . We used FISH probes specific to euchromatic regions on chromosome X and the left arm of chromosome 2 to assay pairing at two loci . We found that salivary glands depleted of CK1α displayed a significant increase ( p < 7 . 2x10−7 ) in the number of FISH spots compared to the controls ( 43B>Gal4 driver without CK1α-RNAi ) with both FISH probes ( Control: X = 1 . 3 ±0 . 09 and 2L = 1 . 1 ±0 . 04 , CK1α RNAi: X = 17 . 3 ±1 . 83 and 2L = 20 . 9 ±2 . 94 spots per nucleus ) ( Fig . 4 ) . The increase in the number of FISH spots indicates that polytene chromosomes are unpaired in the CK1α RNAi expressing larvae . These data demonstrate that RNAi depletion of CK1α in vivo leads to increased unpairing of polytene chromosome , consistent with CK1α being a negative regulator of condensin II anti-pairing activity . Note that these FISH probes cannot distinguish between unpairing of chromatids and unpairing of homologs , and therefore we infer that disruption of paired homologous sequences includes chromatin fibers from both sisters and homologs . These data suggest that CK1α and Slimb may function together to antagonize condensin II activity . Therefore , it is possible that CK1α , Slimb double mutants phenocopy Cap-H2 overexpression phenotypes . Because both CK1α and Slimb genes are essential for viability making homozygous mutants of either gene is not possible . However , given that Slimb heterozygotes can rescue condensin II partial loss of function phenotypes[20] , we speculated that CK1α/+; Slimb/+ double heterozygous mutants may be sufficient to produce Cap-H2 gain-of-function phenotypes . To investigate whether CK1α and Slimb genetically interact to produce Cap-H2 gain-of-function phenotypes we created flies that were heterozygous for both CK1α and Slimb and measured chromosome pairing levels . For these experiments , we used two different Slimb heterozygous mutant alleles: SlimbUU11/+ ( null mutant ) or Slimb3A1/+ ( loss of function ) , in a CK1α heterozygous mutant ( CK1α8B12/+ , null mutant ) background . We did not observe a significant increase in number of FISH spots ( increased unpairing ) in salivary glands of larvae heterozygous for both CK1α and Slimb , when compared to wild-type and single heterozygous mutants ( for 2L probe , Oregon-R: 1 . 21 ±0 . 08 , CK1α8B12/+: 1 . 15 ±0 . 06 , SlimbUU11/+: 1 . 2 ±0 . 07 , Slimb3A1/+: 1 . 07 ±0 . 05 , CK1α8B12/+; SlimbUU11/+: 1 . 19 ±0 . 16 , CK1α8B12/+; Slimb3A1/+: 1 . 09 ±0 . 05 spots per salivary gland nucleus , S4 Fig . ) . Additionally , an increase in FISH spots was not observed in CK1α or Slimb single heterozygous mutant larval polytenes , when compared to a wild-type control ( Oregon-R ) . Thus , reducing either Slimb or CK1α dosage by half or reducing dosage of both by half is not sufficient to produce condensin II gain-of-function phenotypes in salivary glands . This may be because in salivary glands , having one normal copy of CK1α and/or Slimb is sufficient to maintain normal chromosome pairing status . To test whether CK1α and Cap-H2 genetically interact in other tissues , we examined polytene chromosome pairing in the Drosophila ovary . Ovarian nurse cells are highly polyploid , and these polyploid chromosomes switch from a highly paired state ( polytene ) to an unpaired state during stage 5/6 of development [39] . We have previously shown that condensin II and , more specifically , Cap-H2 , is required for this developmental switch in pairing state [16] . Flies homozygous for Cap-H2 mutations fail to unpair their chromosomes at this stage 5-to-6 transition . Combining a Cap-H2 heterozygous mutation ( Cap-H2z3–0019/+ ) with an SMC4 heterozygous mutation ( SMC4k00819/+ ) also results in an intermediate chromosome unpairing defect . We tested whether the introduction of a mutation in CK1α into this double heterozygous ( SMC4k00819/+; Cap-H2z3–0019/+ ) background would modify the condensin II nurse cell unpairing defect . The CK1α8B12 allele is a missense mutation in the CK1α gene that transforms a glycine into an aspartic acid within the kinase domain and is thought to be a null allele [40] . We hypothesized that loss of about half the normal gene dose of normal kinase activity in the CK1α heterozygous mutant may lead to sufficient stabilization of endogenous Cap-H2 protein . Stabilization of Cap-H2 is predicted to suppress the polytene unpairing defect observed in the SMC4k00819/+; Cap-H2z3–0019/+ double heterozygous mutants . To test this , we used FISH probes to X chromosome and 2nd chromosome sequences on whole-mount ovaries and examined stage 10 nurse cells , as previously described [20 , 21] . We chose to examine stage 10 nurse cells specifically because any defects in unpairing should be evident at this later developmental stage . Control flies wild-type for all three genes displayed 25 . 9 ±2 . 83 spots for the 2L probe and 27 . 3 ±2 . 59 spots for the X probe per nurse cell nucleus ( Fig . 5A , G ) . Nurse cells from flies harboring a heterozygous CK1α8B12 mutation alone did not significantly increase the number of FISH spots ( 2L = 21 . 2 ±1 . 5 and X = 23 . 3 ±1 . 9 spots per nurse cell nucleus ) ( Fig . 5B , G ) , likely due to the chromosomes already being unpaired to their maximal degree at this developmental stage . In flies with double heterozygous mutations in condensin II ( SMC4k00819/+; Cap-H2z3–0019/+ ) , the stage 10 nurse cell nuclei displayed 1 . 9 ±0 . 2 and 5 . 1 ±0 . 6 spots per nurse cell nucleus for the 2L and X probe , respectively ( Fig . 5D , G ) . This significant decrease ( p < 4x10−8 ) in number of FISH spots represents a failure of these chromosomes to unpair , consistent with our previous findings [16] . However , introducing a CK1α8B12/+ heterozygous mutant allele into the double heterozygous condensin II mutant background ( SMC4k00819/+; Cap-H2z3–0019/+ ) increased the number of FISH spots back to wild type levels ( 2L = 23 . 5 ±1 . 4 and X = 25 ±1 . 0 spots per nurse cell nucleus; p-value compared to condensin II double mutant: p < 1 . 4x10−12; p-value compared to control: p = 0 . 44 ) ( Fig . 5E , G ) . This triple heterozygous mutant ( CK1α8B12/+; SMC4k00819/+; Cap-H2z3–0019/+ ) was able to completely suppress the unpairing deficiency seen in the condensin II mutants , suggesting that reducing CK1α dosage to half is sufficient to suppress the stage 10 nurse cell condensin II-dependent polytene unpairing defects . This genetic interaction is consistent with CK1α functioning as a negative regulator of condensin II activity . In addition to the CK1α8B12 allele , we also tested two other CK1α heterozygous mutants . The CK1αEP1555 allele is a P-transposable element insertion into the promoter of CK1α [41] . The CK1αG0492 allele [42] is a CK1α mutation derived from a P-element lacW insertion into the CK1α gene . When crossed into the condensin II double heterozygous mutant background ( SMC4k00819/+; Cap-H2z3–0019/+ ) , both the CK1αEP1555 and the CK1αG0492 mutants significantly suppressed the nurse cell chromosome unpairing defect ( CK1αEP1555/+; SMC4k00819/+; Cap-H2z3–0019/+: 2L = 11 . 3 ±1 . 0 and X = 13 . 8 ±0 . 8 spots per nurse cell nucleus; p < 2 . 3x10−10 , CK1αG0492/+; SMC4k00819/+; Cap-H2z3–0019/+: 2L = 11 . 4 ±0 . 6; p < 1 . 2x10−14 ) ( Fig . 5F-G and S5D–E Fig . ) . However , these additional mutants did not suppress the unpairing defect as well as the CK1α8B12/+ mutant . The CK1αEP1555 and CK1αG0492 alleles were only able to rescue the condensin II unpairing defect to levels approximately 50% of the control flies . It is possible that these two additional mutants , CK1αEP1555/+ and the CK1αG0492/+ , are not complete protein nulls , like the CK1α8B12 mutation . Nevertheless , three different and independently derived alleles of CK1α suppress the condensin II unpairing defect , providing strong genetic support for CK1α functioning as a negative regulator of Cap-H2 in vivo . These observations are also consistent with CK1α RNAi depletion in cultured cells resulting in hyper-activation of Cap-H2 chromosome anti-pairing function ( Fig . 3H , M , N ) . A possible mechanism for CK1α regulation of condensin II activity may be by maintaining low levels of chromatin bound Cap-H2 . To test whether chromatin bound Cap-H2 is elevated in response to reduction of CK1α levels , we performed immunostaining on polytene chromosomes from salivary glands of third instar larvae . Maintaining a low level of Cap-H2 in this tissue is necessary for the maintenance of the highly paired state of polytene chromosomes , as overexpression of Cap-H2 can drive these chromosomes to become highly unpaired[16] . Salivary glands were dissected from wild type control and CK1α heterozygous mutants , squashed onto microscope slides , and immunostained for Cap-H2 to assess levels of endogenous chromatin bound protein . Two different CK1α mutants , CK1α8B12 and CK1αEP1555 were used ( as in Fig . 5 ) , each containing an independently derived disruption of the CK1α gene . Salivary glands from wild-type ( Oregon R ) and CK1α mutants were squashed onto the same microscope slide such that all chromosome spreads simultaneously received the same mixture of washes and antibody reagents , as previously described [43] . This approach allowed us to use identical imaging parameters to make comparisons between different genotypes from the same slide . In wild type ( Oregon-R ) salivary glands , Cap-H2 is present on the chromatin at low levels ( Fig . 6A-B ) and requires a higher exposure to detect the protein signal . However , salivary glands from mutants with either of the two heterozygous CK1α mutations show a visibly higher level of Cap-H2 on the chromatin , which is apparent even at the lower exposure , demonstrating that chromatin bound Cap-H2 protein levels are elevated ( Fig . 6A-B ) . Quantitation of fluorescence intensity confirms that Cap-H2 protein levels on the salivary gland chromosomes is significantly higher in CK1α heterozygous mutants ( CK1α8B12/+: p < 1 . 8x10−7 , CK1αEP1555/+: p < 0 . 05 when compared to wild-type controls ) ( Fig . 6D ) . These results suggest that CK1α is required to maintain low levels of chromatin bound Cap-H2 , as half dosage CK1α mutation is sufficient to increase chromatin bound Cap-H2 . In addition to the CK1α heterozygous mutants , we performed immunostaining for Cap-H2 in flies expressing CK1α RNAi under a UAS-regulated promoter crossed with a 43B>Gal4 transgene , a salivary gland specific driver [44] . Similar to the results seen with the CK1α heterozygous mutants , CK1α RNAi resulted in a visible increase in chromatin bound Cap-H2 ( Fig . 6C ) . The significant increase ( p < 4 . 4x10−5 ) in Cap-H2 fluorescence intensity confirms this observation ( Fig . 6D ) . In addition to the increased level of Cap-H2 staining on the DNA , we also noticed chromosome aberrations , where the polytene banding pattern was clearly disrupted . These abnormal chromosome structures are reminiscent of the unpaired chromosomes seen in whole mount salivary gland nuclei that are overexpressing Cap-H2 ( Fig . 4 ) [16 , 22] . The abnormal chromosomes also display a higher intensity of Cap-H2 staining ( Fig . 6D ) and further support the idea that an increase in Cap-H2 levels via CK1α depletion is driving the chromosomes to become highly unpaired . Depletion of CK1α resulted in increased condensin II activity and chromatin bound Cap-H2 levels . We next wanted to test if CK1α is functioning to degrade Cap-H2 protein . In order to do this , we assayed Cap-H2 levels in cultured cells treated with CK1α RNAi . In order to functionally validate depletion of CK1α by RNAi , the level of the Slimb and CK1α substrate Armadillo was determined by immunochemistry [32 , 34] . Slimb and CK1α normally function to repress Armadillo protein levels . RNAi depletion of CK1α resulted in stabilization of Armadillo to levels higher than that of Slimb depletion , confirming efficient depletion of CK1α ( Fig . 7A ) . Whole cell extracts from cultured S2 cells stably expressing an inducible Cap-H2-EGFP depleted of CK1α by RNAi showed stabilization of EGFP tagged Cap-H2 as compared to control treated cells ( Fig . 7A ) . In addition , Kc cells treated with the CK1 inhibitor D4476 also resulted in stabilization of a transiently transfected and induced Cap-H2-GFP protein ( Fig . 7B ) . These results further support the idea that CK1α negatively regulates condensin II through its subunit , Cap-H2 , as depletion or inhibition of CK1α results in stabilization of Cap-H2 protein . Based on the results of stabilization of Cap-H2 when CK1α is inhibited in cultured cells and the increase in Cap-H2 fluorescence on CK1α mutant salivary gland chromosomes ( Figs . 6 and 7A-B ) , we hypothesized that a normal function of CK1α may be to limit chromatin bound Cap-H2 levels . In order to test this , we performed cellular fractionations to ask if chromatin bound Cap-H2 levels were stabilized in the absence of CK1α . For these experiments , we used a Kc cell line that stably expresses an inducible Cap-H2-EGFP and treated them as before with RNAi . CK1α-RNAi depletion resulted in an increase of chromatin bound Cap-H2-EGFP by 40 ±0 . 11% , as compared to chromatin extracts from control-RNAi treated cells ( Fig . 7D-E ) . This result demonstrates that CK1α is negatively modulating both whole cell and more specifically , chromatin bound Cap-H2 protein levels . More importantly , this suggests that CK1α normally inhibits condensin II acitivity in part by limiting the levels of chromatin bound Cap-H2 protein . We previously demonstrated that Slimb and Cap-H2 proteins co-immunoprecipitate [20] . Based on reports of how CK1α is required for Slimb interaction with its targets , for example the Beta-Catenin/Armadillo , Ci and Cdc25 proteins[34 , 45 , 46] , we speculated that CK1α is also required for Slimb interacting with Cap-H2 and promoting Cap-H2 ubiquitination . To test this , we performed immunoprecipitations in CK1α-RNAi depleted Kc cells that stably expressed an inducible Cap-H2-EGFP . To our surprise , when Cap-H2-EGFP was immunoprecipitated , endogenous Slimb also co-immunoprecipitated in both control-RNAi treated and CK1α-RNAi treated cells ( Fig . 7F ) . Slimb did not co-immunoprecipitate with EGFP tag only controls transiently transfected into untreated Kc cells . Similar results were observed in S2 cells transiently transfected with GFP tag only and Cap-H2-EGFP ( S6 Fig . ) . The unexpected result of CK1α depletion having no effect on Cap-H2 and Slimb interaction led us to postulate that an additional kinase ( s ) may be functioning upstream of CK1α to prime Cap-H2 , permitting Slimb interaction . We decided to perform immunoprecipitations in Kc stable cells that were codepleted of PKA , GSK3β , and CK1α . As discussed earlier , PKA and GSK3β have been shown to work with CK1α to phosphorylate Slimb substrates . Surprisingly , Cap-H2-EGFP and Slimb interaction remained unperturbed in the PKA , GSK3β , and CK1α co-depleted cells ( Fig . 7F ) . Furthermore , to see if Cap-H2 was still being ubiquitinated in CK1α depleted cells , immunoprecipitations were performed in RNAi treated S2 cells co-transfected with Cap-H2-EGFP and 3xFLAG-tagged ubiquitin . In both control-RNAi and CK1α-RNAi depleted cells , immunoprecipitated Cap-H2-EGFP was found to be labeled with FLAG-ubiquitin ( Fig . 7G ) . These observations are surprising because Slimb interaction with its protein targets is thought to result in target-protein proteolysis . However , here we observe that although both CK1α and Slimb are required for targeting Cap-H2 for degradation , Slimb can still interact with Cap-H2 when stabilized by CK1α depletion . We speculate on possible explanations in the discussion .
Drosophila condensin II functions in interphase genome organization through its role in regulating chromosome compaction , homolog pairing and dispersal of centromeres [16 , 18 , 19 , 20 , 21 , 22] . In this study , we report a previously unidentified role of Casein Kinase I Alpha ( CK1α ) . We find that CK1α is an important modulator of interphase genome organization , regulating homologous chromosome pairing , centromere clustering , and chromosome compaction in Drosophila cultured cells and in vivo . Furthermore , we show that CK1α affects these processes by attenuating interphase condensin II activity as CK1α function is required for protein turn-over of the Cap-H2 condensin II subunit . Using an RNAi based approach in Drosophila cultured cells , we observed that depletion of CK1α resulted in perturbation of interphase nuclear morphology ( Fig . 1 ) . We also find that CK1α functions to prevent centromere dispersal ( Fig . 2 and S2 Fig . ) , inhibits chromosome compaction ( Fig . 3 ) , and promotes chromosome pairing ( Figs . 3 , 4 , 5 and S2 and S5 Figs . ) . These observations are consistent with the changes seen when Cap-H2 is overexpressed . In cultured Drosophila cells , co-depletion of CK1α with the condensin II subunit Cap-H2 results in suppression of abnormal centromere dispersal , suppression of chromosome hyper-compaction , and suppression of chromosome unpairing ( Figs . 1 , 2 , 3 and S2 Fig . ) . These observations strongly suggest that CK1α and Cap-H2 interact genetically . This interaction is also observed in vivo , as the Drosophila nurse cell chromosome unpairing defect seen in condensin II mutants ( SMC4k00819/+; Cap-H2z30019/+ ) is suppressed by three independent CK1α heterozygous mutations ( Fig . 5 and S5 Fig . ) . Furthermore , CK1α-RNAi depletion or mutations increases chromatin bound Cap-H2 protein levels ( Figs . 6 and 7 ) . This was determined by immunofluorescence of endogenous Cap-H2 protein on polytene chromosomes as well as by sub-cellular fractionation of chromatin bound proteins from Cap-H2-EGFP expressing cells in culture . It is important to emphasize that Cap-H2 and other condensin II gene loss-of-function mutations do have oogenesis phenotypes that are completely suppressed by decreasing the dosage of CK1α by half ( Fig . 5 and S5 Fig . ) . Together , these findings demonstrate CK1α as a novel regulator of interphase condensin II levels and activity . CK1α is a highly conserved serine/threonine kinase involved in Wnt signaling pathways , DNA repair , cell cycle progression , and mRNA metabolism [35 , 47 , 48] . Identification of CK1α furthers our understanding of the mechanisms by which condensin II is regulated . The chromodomain protein Mrg15 is involved in the loading of Cap-H2 , while the E3 Ubiquitin ligase , SCFSlimb ubiquitylates Cap-H2 , removing it from chromatin and targeting it for proteasomal degradation [20 , 21] . Phosphorylation is known to be a prerequisite for Slimb recognition of its target proteins [29 , 49] . CK1α has been shown to require other kinases that prime its recognition site . For example , CK1α can phosphorylate the second S/T in the S/T-X-X-S/T motif but only when the S/T in the first position is already phosphorylated by a different kinase [33 , 50] . Interestingly , the C-terminal end of Drosophila Cap-H2 , TPSDADSGISSMGSSLASTARLK , contains three S-X-X-S sites , where the underlined serines denote potential priming sites and bold serines indicate potential CK1α phosphorylation sites ( Fig . 7C ) . However , our efforts to test a direct interaction between CK1α and Cap-H2 were unsuccessful , as we were unable to produce kinase-active CK1α to determine whether CK1α can phosphorylate recombinant Cap-H2 in vitro . Because depletion of CK1α results in stabilization of Cap-H2 protein and increased condensin II activity ( Figs . 6 and 7 ) , we speculate that Cap-H2 itself is targeted by CK1α but likely requires an additional kinase to prime Cap-H2 phosphorylation , as previously shown for other substrates like Armadillo/β-Catenin [33 , 50] . This hypothesis is further supported by the observation that the Cap-H2-Slimb interaction persists and that Cap-H2 is still ubiquitinated when CK1α is depleted ( Fig . 7F-G and S6 Fig . ) . It can be speculated that the initial steps of Cap-H2 degradation involve an additional regulator , perhaps a kinase priming Cap-H2 at specific residues , permitting Slimb interaction and ubiquitination of Cap-H2 at specific sites . However , the final step for Cap-H2 targeting for degradation and removal from chromatin may be mediated by CK1α . Alternatively , it is possible that CK1α is indirectly regulating Cap-H2 , perhaps through an intermediate . CK1α could be phosphorylating a regulator of Cap-H2 which determines Cap-H2 protein levels . For example , it is possible that although Slimb binding to Cap-H2 is independent of CK1α , this kinase may phosphorylate some other protein or Slimb itself before the Cap-H2 protein can be fully degraded . Clearly how Slimb and CK1α regulate Cap-H2 levels is more complex than previously appreciated . A recent study has shown that Armadillo/β-Catenin is protected from degradation by the Armless protein binding to and inhibiting Ter95 , a component of the SCFSlimb E3-ligase[51] . This raises the possibility that Cap-H2 may similarly be protected from proteolysis , and Slimb-bound Cap-H2 molecules may require CK1α to eliminate this protective function . While we have not been able to show a direct interaction between CK1α and Cap-H2 , it is clear from our experiments in cultured cells and in Drosophila tissues that CK1α limits Cap-H2 protein levels , thereby attenuating interphase condensin II activity . Having multiple regulators of Cap-H2 allows for precise modulation of condensin II activity . We have previously shown that both Cap-H2 and Slimb are chromatin bound [20] . Whether CK1α is also bound to chromatin is unknown . CK1α associates with centrosomes in Chinese hamster ovary cells and CK1α enters the nucleus after DNA damage induction in Drosophila embryos [35 , 52] . If CK1α interaction with Cap-H2 precedes and is required for Slimb mediated Cap-H2 turnover , having a two-step requirement for Cap-H2 degradation would allow for more precise control over the spatial regulation of condensin II activity . We speculate that at chromatin regions where both Slimb and Cap-H2 are localized , condensin II activity can remain high while Cap-H2 remains dephosphorylated within the Slimb recognition site ( Fig . 8 ) . However , condensin II activity can be reduced quickly by localizing CK1α to these specific regions and/or by regulated nuclear import of CK1α[35] , resulting in destruction of Cap-H2 protein . This speculative model of Cap-H2 removal from chromatin is attractive in that it would allow for a “switch-like” change in condensin activity that is responsive to developmental cues and environmental stressors . Furthermore , by differential distribution of Cap-H2 loading factors like Mrg15[21] and inhibitors like Slimb [18 , 20] this model would allow for precise changes in chromosome morphology at the local chromatin level , possibly giving rise to local regions of high and low compaction that may define structures such as topologically associated domains ( TADs ) . It should be noted that there are four predicted Drosophila Cap-H2 splice isoforms . One of these isoforms , Cap-H2-RD , lacks a region of 64 amino acids in the C-terminus , present in the other three isoforms . This raises the possibility that Cap-H2-RD may be resistant to Slimb ubiquitination . We speculate that this isoform could provide areas of constitutive condensin II activity , where the chromatin is unpaired and/or providing local regions of high compaction . In Drosophila it is thought that the genome exhibits region-specific levels of homolog pairing , for example heterochromatin and euchromatin exhibit different somatic pairing properties [18 , 38 , 53] . Having different Cap-H2 isoforms and differential localization of Cap-H2 regulators bound to specific regions of chromatin could provide the necessary landscape to enable differences in regional pairing and compaction state , while regulators such as Slimb and CK1α add plasticity to compaction states by modulating chromatin bound Cap-H2 levels . Condensin II modulation of interphase chromatin compaction has also been shown to affect accessibility of other chromatin proteins . It was observed in Cap-H2 mutant mice , that loss of Cap-H2 resulted in disruption of T-cell differentiation [54 , 55] . Interestingly , chromatin condensation is involved in mouse T-cell differentiation . In response to cytokine signaling , naïve T-cell chromatin must decondense in order to allow for STAT5 transcription factor binding . Furthermore , this decondensation is dependent on the inactivation of condensin II [27] . Clearly , limiting interphase condensin II activity is important in this context , and it remains to be determined what mechanism is acting to repress condensin II function during mammalian T cell activation . It is tempting to speculate that cytokine signaling could trigger the activation of a condensin II antagonist , leading to the decrease in condensin II activity . This would lead to decondensation of chromatin allowing STAT5 access to DNA . Our findings in the Drosophila model suggest that similar interphase condensin II functions may be at play , and CK1α along with Slimb are critical regulators of this condensin II activity . However , at present it is not known if mammalian condensin II activity is regulated by Slimb or CK1α , and it should be noted that mouse and human Cap-H2 do not have clear Slimb binding consensus sequences . It will be of great value to identify additional kinases that may collaborate with CK1α and Slimb to negatively regulate Drosophila condensin II activity , and to further elucidate the biological significance of this interphase condensin II function in Drosophila and other species .
Drosophila cell culture , in vitro dsRNA synthesis , and RNAi treatments were performed as previously described [56] . S2 and Kc cells were cultured in Sf900-II media ( Life Technologies ) supplemented with 1X Antibiotic-Antimycotic ( Life Technologies ) . S2R+ cells were cultured in Schneider’s media ( Life Technologies ) supplemented with 1X Penicillin Streptomycin ( Corning Cellgro ) and 10% Fetal Bovine Serum ( Thermo Scientific; Hyclone ) . RNAi treatments were performed in 6-well tissue culture plates , with 10μg of dsRNA in 1mL media administered to confluent ( 50–90% ) wells . Wells were replenished with fresh 1mL of media and 10μg dsRNA every other day for 4–7 days . dsRNA was made using gene specific primer sequences listed in S1 Table . Control ( SK ) dsRNA was made by amplifying off PCR product from a non-GFP sequence of the pEGFP-N1 vector ( Takara Bio Inc . ) . PCR products were then used as template for dsRNA production using T7 RiboMAX Express Large Scale RNA Production System kit ( Promega , # P1320 ) . dsRNA concentration was then calculated using gel electrophoresis and densitometry analysis ( NIH ImageJ ) . Cell viability experiments were performed with Kc cells treated with dsRNA and cultured as described above with modifications . 750 , 000 cells per well were plated in triplicates into a 24-well tissue culture plate ( Falcon ) with 300μL of media and 3μg of dsRNA per well and treated every other day for 4–6 days . At days 4 and 6 , aliquots of cells were taken for Propidum Iodide and DAPI staining for cell viability analysis . Approximately one third ( 100μL ) of each well was taken and 150μL new media was added containing Propidium Iodide ( 1ng/μL ) and DAPI ( 2μg/mL ) and incubated at 25°c for 30 minutes . Mixture was then spun down and resuspended in 10μL new media and plated onto microscope slides and coverslips to image . Image fields were obtained and used for counting with nuclei co-staining for both Propidium Iodide and DAPI being counted as a “dead” cell . Construct production and transfection were performed as described previously [20] . cDNA encoding Cap-H2-EGFP or Drosophila ubiquitin ( 3x-Flag-Ubiquitin ) were subcloned into the inducible metallothionein promoter pMT vector , as previously described[20] . Transient transfections were performed using the Nucleofector II ( Lonza ) according to manufacturer’s instructions . Transiently transfected cells ( Fig . 7B , F , G ) were induced 24 hours post transfection . Stable S2 cell lines ( Fig . 7A ) were selected by co-transfection with pCoHygro ( Life Technologies ) plasmid and treated for 3–4 weeks with Hygromycin B ( Life Technologies ) . Expression of all constructs was induced by addition of 1mM CuSO4 to the media and processed for downstream experiments 24 hours later . Transiently transfected Kc cells in Fig . 7B were cotransfected with pMT-eGFP and pMT-Cap-H2-eGFP in order to confirm equal transfection efficiency between treatments . Stable Kc cell lines under inducible metallothionein promoter ( Fig . 7D ) were provided by the Wu lab ( Harvard Medical School ) . D4476 , chemical Casein Kinase I inhibitor ( EMD Millipore Calbiochem ) was resuspended in DMSO to 50μM stock dilutions . D4476 was then diluted into cell culture media at 80μM and added to cells ( 60–95% confluent ) in 6-well tissue culture plates . Cells were treated for 8 hours and prepped for immunoblotting or plated onto Concanavalin-A ( Sigma-Aldrich ) coated coverslips for immunostaining , as previously described . Control treated cells were treated with same volume of DMSO under same conditions as drug treatment . Cultured cells were immunostained as previously described [56] . Cells were plated onto Concanavalin-A ( Sigma-Aldrich ) coated cover slips , allowed to adhere onto coverslips for 20min then fixed with 10% formaldehyde ( Ted Pella , INC ) in PBS at room temperature . Cells were then washed with PBS and then 0 . 1% PBS/Triton to permeabilize . Cells were then blocked with Block Solution ( 5% normal goat serum ( Sigma-Aldrich ) , 0 . 1% PBS/Triton , and 1mM Sodium Azide ( Sigma-Aldrich ) ) for 1 hour at room temperature prior to immunostaining . Primary antibodies were diluted in block solution and coverslips were incubated with primary antibody for 1 hour at room temperature . Primary antibody concentrations used were as follows: rabbit anti-CID at 1:400 [20] , mouse anti-Lamin ( Dm0 ) ADL 84 . 12 at 1:200 ( Drosophila Hybridoma Bank , University of Iowa ) , rabbit anti-phosphorylated-Histone-H3 at 1:500 ( Upstate; EMD Millipore ) , and rabbit anti-Cap-H2 at 1:50 [16] . Cells were then washed with 0 . 1% PBS/Triton three times for 5 minutes each at RT . Secondary antibodies ( conjugated to Alexa 488 ( Life Technologies ) , Cy-2 , or Cy-3 ( Jackson ImmunoResearch Laboratories ) were diluted in block solution and coverslips were incubated with secondary for 2 hours at room temperature . Secondary antibody concentrations used were as follows: Alexa 488 at 1:500 , Cy-2 at 1:500 , and Cy-3 at 1:500 . Coverslips were then washed with 0 . 1% PBS/Triton three times for 5 minutes each at RT . Coverslips were then washed with PBS for 5 minutes at RT . 4’ , 6-Diamidino-2-Phenylindole , Dihydrochloride ( DAPI ) ( Life Technologies ) was used at a final concentration of 1μg/mL in PBS for 10 minutes to stain DNA . Coverslips were then washed two times in PBS for 5 minutes . Coverslips were then mounted onto microscope slides and were mounted using Vectashield ( Vector Labs ) and sealed with nail polish and stored at -20 degrees . Whole mount ovaries were immunostained as previously described [21] . Tissues were dissected in PBS and fixed in 4% formaldehyde/PBS for 10 minutes . Tissues were then rinsed with 0 . 1% PBS/Triton and blocked with block solution ( same as used for cell staining ) for 30 minutes . Tissues were incubated with primary antibodies ( same as above ) overnight at 4° rotating on rotator . Tissue was then washed in 0 . 1% PBS/Triton three times for 5 minutes while rotating at room temperature . Tissues were then blocked two times for 15 minutes while rotating at room temperature . Secondary antibodies ( same as above ) were incubated with samples for 2 hours . Tissues were rinsed with 0 . 1% PBS/Triton then PBS and counterstained with DAPI for 10 minutes . Tissues were then mounted onto slides with Vectashield and sealed using nail polish . Salivary gland squashes were performed as previously described [57] with modifications . Salivary glands were dissected in 0 . 7% NaCl and were placed onto Repel silane ( GE Healthcare ) coated cover slips and fixed in 3 . 7% formaldehyde in 45% acetic acid for 2 . 5 minutes . Coverslips with glands were then inverted onto a glass poly-L-lysine ( Sigma ) coated microscope slide and squashed with an orthopedic hammer until nuclei burst . Glands from different genotypes were squashed onto different sections of the same slide to serve as an internal control . The location of the different glands was noted on the slide , as previously described [43] . Slides/coverslips were then dipped into liquid nitrogen and coverslips were removed using a razor blade . Squashed chromosomes were then washed once in PBS at RT . Antibodies ( same as above ) were diluted in PBS , 0 . 1% NP40 ( Sigma-Aldrich ) , and 1% non-fat milk ( Carnation ) and added to slide and covered with cover slip . Primary antibody incubations were performed in humid chamber at 4 degrees overnight . Slides were then washed in 0 . 1% NP40/PBS for 5 minutes and secondary antibodies ( same as above ) were added and covered with cover slip for 2 hours in humid chamber at RT . Slides were then washed three times in PBS and counterstained with DAPI ( 1μg/mL ) for 25 seconds and washed for 5 minutes in PBS . Slides were mounted with coverslip and Vectashield and sealed with nail polish . Whole mount salivary gland and ovary FISH was performed by dissecting tissues into PBS . Tissues were then fixed in 100 mM sodium cacodylate , 100 mM sucrose , 40 mM sodium acetate , 10 mM EGTA , and 3 . 7% formaldehyde for 4 min at RT . Tissues were then rinsed in 2X SSCT ( 0 . 3 M NaCl , 0 . 03 M sodium citrate , pH 7 . 0 , and 0 . 1% Triton ) and treated with 2μg/mL Ribonuclease A ( Sigma-Aldrich ) for 1 hour . 10 minute stepwise washes were performed in 20% , 40% , and 50% formamide ( Sigma-Aldrich ) in 2X SSCT . Tissues were pre-hybridized in 50% formamide/2X SSCT for 2 hours in 37° water bath . 1–2 μl of each probe in hybridization solution ( Dextran sulfate ( Sigma-Aldrich ) , NaCl , sodium citrate , formamide , H2O ) to a total volume of 40 μL was added to 0 . 2 mL PCR tube with tissues . Hybridization was performed using a thermal cycler , 91° for 2 minutes to denature and 37° overnight for hybridization . Tissues were then washed in 50% formamide in 2X SSCT three times for 30 minutes . 10 minute stepwise washes were performed in 40% then 20% formamide in 2X SSCT . Tissues were then washed three times in 2X SSCT for 10 minutes , 10 minutes in 0 . 1% PBS/Triton , and then counterstained with DAPI ( 1 μg/mL ) in PBS for 10 minutes at RT . Two 10 minute PBS washes were performed and tissues were mounted in Vectashield and sealed with nail polish . Cultured cell FISH was performed as previously described [20] . Cells were plated onto Con-A coated coverslips in a well of a 6-well tissue culture plate for 20 minutes and allowed to adhere to coverslips at room temperature . Coverslips were then washed with 1X PBS and fixed in 10% Formaldehyde/PBS for 10 minutes at RT . Coverslips were then washed once with PBS and permeabalized in 0 . 1% PBS/Triton for 10 minutes . Cells were then washed in CSK buffer ( 10mM HEPES , 100mM NaCl , 3mM MgCl2 , 300 mM Sucrose , and Phenylmethanesufonyl fluoride ( PMSF ) ) for 10 minutes and Ribonuclease A ( 100ug/mL ) for 1 hour at RT . Cells were then washed with 0 . 1N HCl for 5 minutes and taken through ethanol series for 5 minutes each ( 70% , 90% , then 100% ) . One 2X SSCT wash was performed and cells were pre-hybed in 50% formamide/2X SSCT at 37 degrees for 2 hours . FISH probe ( 1–2 μL each probe ) was then added to hybridization solution ( total 25 μL ) and this mixture was denatured at 95° for 2 minutes and snap cooled in ice bath . Probe mixture was then added onto microscope slide and coverslips were inverted onto the microscope slide . Coverslips were then sealed with rubber cement and slide/coverslip was denatured at 93° on a heat block for 3 minutes . Slides were then placed in humid chamber and hybridized overnight at 37° . After hybridization was complete , coverslips were detached from slides by immersing in 50% formamide/2X SSCT with shaking for 10 minutes . Coverslips were then placed into 6-well tissue culture plates and washed three times for 30 minutes at 42° ( all heterochromatin probe post washes performed at 37° ) . Ten minute washes at 42° were then performed with 40% then 20% formamide in 2X SSCT . Three 2X SSCT washes were performed for 5 minutes each at RT on shaker . Cells were then counterstained with DAPI ( 1μg/mL ) in PBS for 10 minutes at RT . Coverslips were then washed two times for 10 minutes at RT in PBS . Cells were then mounted with Vectashield and sealed with nail polish . Euchromatic FISH probes were made as previously described [20 , 21] from BAC clones ( Children’s Hospital Oakland Research Institute BACPAC Resources ) as follows: X1 , BACR30C13 and BACR18F10; X2 , BACR20K01 and BACR35A18; X3 , BACR11C13 and BACR07F15; 2L ( 1 ) BACR30M19 and BACR29P12 . For nurse cell and salivary gland FISH: X , BACR20K01 and BACR35A18; 2L , BACR30M19 and BACR29P12 . BAC clones were mapped and picked by using the UCSC genome browser ( http://www . genome . ucsd . edu ) . BAC clones were cultured and DNA was purified using QIAGEN Plasmid Midi Kit ( Qiagen 12143 ) . Purified BAC DNA was amplified using Whole genome amplification kit ( Sigma-Aldrich WGA1 ) , 20 μg of amplified DNA was then digested using a cocktail of AluI , Rsa , MseI , MspI , HaeIII , and BfuCl ( New England BioLabs ) overnight at 37° , ethanol precipitated , and resuspended in 36 μL ddH2O . DNA was denatured at 100°C for 1 min , and then 3'-end-labeled with unmodified aminoallyl dUTP and terminal deoxynucleotidyl transferase ( Roche ) . After incubating for 2 h at 37° , 5 mM EDTA was added to terminate the reaction . DNA was ethanol precipitated , resuspended in 10uL ddH2O , and then conjugated to fluorophores using ARES Alexa Fluor DNA labeling kits ( A-21665 , A21667 , and A-21676; Life Technologies ) for 2 h , according to the manufacturer’s instructions . Probes were then cleaned up using Qiagen PCR clean up kit ( Qiagen ) , ethanol precipitated , and resuspended in 10μL EB buffer ( Qiagen ) . Heterochromatic FISH probes were made using 10 μg of oligonucleotide primers ( Integrated DNA Technologies ) diluted in TE buffer using sequences: for heterochromatin to 2R: 5'-AACACAACACAACACAACACAACACAACAC-3' , for heterochromatin to 3R: 5'-CCCGTACTGGTCCCGTACTGGTCCCGTACTCGGTCCCGTACTCGGT-3' . DNA was denatured at 100° for 1 minute and snap cooled on ice , DNA was then end labeled using 3’-end labeling with fluorophore conjugated dUTP ( cy3 and cy5 ) ( GE Healthcare Amersham ) using terminal deoxynucleotidyl transferase ( Roche ) . Reactions were left in 37° water bath for 2 hours . Reactions were stopped by adding 1 μL of 0 . 25 mM EDTA . Probes were then ethanol precipitated and resuspended in TE buffer . Micrographs were obtained on a Nikon A1RSi confocal microscope using either a Plan Apo 60X 1 . 4 NA or a Plan Apo 100X 1 . 49 NA oil immersion objective using the Nikon Elements 4 . 0 software package . Micrographs were processed using Nikon Elements or ImageJ ( NIH ) . CID and FISH spot counts were performed on maximum z-projections from z-stack images using the counting software in Nikon Elements . 3D FISH distance measurements were performed manually in Nikon Elements using the 3D distance measuring tool by scanning through each Z-slice . The centroid of each FISH signal would be marked and the shortest 3D pairwise distance would then be calculated . For scenarios where the chromosomes are unpaired , resulting in two foci per probe ( 6 FISH signal total ) , pairwise distance measurements were performed on the signals that were closer to each other . The assumption made is that in the situation where the chromosomes are unpaired , it is more likely that the three signals closer to one another are from the same chromosome and that the further three signals are from the other chromosome , as previously described for similar measurements made in mammalian and C . elegans cells [26 , 58 , 59] . Salivary gland polytene squash intensity analysis was performed by obtaining multiple fields of chromosomes from different genotypes on the same slide . Experimental genotypes ( CK1α mutants and RNAi ) were squashed on the same slide as their respective controls ( Oregon R for mutants and 43B>Gal4 for RNAi ) such that they could be imaged using identical settings for downstream analyses [43] . Multiple fields ( 5–6 ) of chromosomes were imaged for each genotype in both DAPI and Cap-H2 channels . Images used for fluorescence intensity measurements were captured such that pixel saturation was limited for all channels . Images were then analyzed using ImageJ as follows: First , all image sets were split into separate channels ( DAPI and Cap-H2 ) and converted to gray scale . Secondly , the mean/average intensity value for the entire DAPI image was measured . Background fluorescence was subtracted by measuring a “blank” square from the same DAPI image and subtracting mean background intensity of this square from the mean grey value of the entire image ( DAPImean intensity—Backgroundmean intensity = DAPIcorrected ) . This produced a background subtracted intensity value for the DAPI image ( DAPIcorrected ) . Next , the same process was performed for the image using the Cap-H2 image in grey scale ( Cap-H2correct ) . Then , the ratio of Cap-H2 to DAPI was calculated ( Cap-H2corrected ÷ DAPIcorrected = Chromatin Cap-H2 ) for each set of images . These values were then averaged for all the image fields obtained for each genotype . Changes in chromatin bound Cap-H2 levels were calculated by comparing the experimental genotype average ratio to its respective control average ratio ( Chromatin Cap-H2CK1α Mutants ÷ Chromatin Cap-H2Controls = Change in chromatin bound Cap-H2 ) . Values in Fig . 6D represent fold-changes in Cap-H2 as compared to their respective controls . Statistical analyses were performed in Microscoft Excel using a two-tailed student t-test assuming unequal variance . Mitotic indexes/percentage of cells staining positive for phosphorylated-histone-H3 were assessed by using CellProfiler Cell Image Analysis Software ( http://www . cellprofiler . org; Broad Insitute ) . Micrographs were captured on a Nikon Eclipse E800 Epifluorescent upright microscope using a Nikon Plan Fluor 20X 0 . 5 NA objective with Olympus DP controller software . For Fig . 7A-B , cell extracts were obtained by pelleting and lysing cells in PBT with protease inhibitor ( Roche ) . Protein concentration was calculated by performing the Bradford protein assay ( Bio-Rad Laboratories ) . Laemmli sample buffer was then added to extracts and boiled for 5 minutes prior to loading onto denaturing gel . For Fig . 7 , antibodies used are as follows: mouse anti-GFP ( JL8 Takara Bio Inc . ) , guinea pig anti-Slimb ( Brownlee et al , 2011 ) , mouse anti-armadillo ( N2 7A1; Drosophila Hybridoma Bank ) , mouse anti-alpha-tubulin ( Dm1; Sigma-Aldrich ) , mouse anti-lamin Dm0 ( ADL84 . 12; Drosophila Hybridoma Bank ) , and rabbit anti-Hisone-H3 ( 06–755; EMD Millipore ) . Immunoprecipitations were performed as previously described [20] . GFP-binding protein ( GBP; [60] ) was fused to the Fc domain of human IgG ( pIg-Tail; R&D Systems ) , tagged with His6 in pET28a ( EMD Millipore ) , expressed in E . coli , and purified on Talon resin ( Takara Bio Inc . ) according to manufacturer’s instructions . GBP was bound to Protein A–coupled Sepharose , cross-linked to the resin using dimethyl pimelimidate , and rocked for 1 h at 22°C; the coupling reaction was then quenched in 0 . 2 M ethanolamine , pH 8 . 0 , and rocked for 2 h at 22°C . Antibody or GBP-coated beads were washed three times with 1 . 5 ml of cell lysis buffer ( CLB; 100 mM Tris , pH 7 . 2 , 125 mM NaCl , 1 mM DTT , 0 . 1% Triton X-100 , and 1X Protease Inhibitor ( Roche ) . S2 , Kc , and Kc stable line expressing inducible Cap-H2-EGFP cells were treated with RNAi for 6 days , as described above in “Cell culture and double-stranded RNAi . ” On day 4 , cells were transfected with inducible GFP tag only , Cap-H2-EGFP , and/or 3x-FLAG-Ubiquitin . On day 5 , transfected cells were induced with 1 mM CuSO4 . After 24 h , transfected cells were lysed in CLB , clarified by centrifugation , and then diluted to 2–5 mg/ml in CLB . Antibody-coated beads were mixed with lysate in 1mL total volume for 90 min at 4°C , washed three times with CLB , and then boiled in Laemmli sample buffer . Kc cells stably expressing an inducible Cap-H2-EGFP ( provided by the Wu lab ) were treated with RNAi as described above but in T75 flasks . Cells were induced 24h prior to harvesting . On day 6 , cells were harvested and then fractionated into whole cell lysate ( WCL ) , cytoplasmic ( S2 ) , nuclear-soluble ( S3 ) , and chromatin ( P3 ) fractions as described previously [20 , 61] . In brief , ~107 cells were collected , washed with cold PBS , and resuspended at 4 × 107 cells/ml in buffer A: 10 mM Hepes , pH 7 . 9 , 10 mM KCl , 1 . 5 mM MgCl2 , 0 . 34 M sucrose , 10% glycerol , 1 mM DTT , and protease inhibitor cocktail ( Roche ) . Cells were lysed by the addition of 0 . 1% Triton X-100 and then incubated on ice for 8 min ( fraction WCL ) . Nuclei were collected by centrifugation ( 5 min , 1 , 300 g , 4°C ) , the initial supernatant was further cleared by high-speed centrifugation ( 5 min , 20 , 000 g , 4°C ) , and the final supernatant was collected ( fraction S2 ) . The pelleted nuclei were washed once in buffer A , and nuclear membranes were lysed for 30 min in 3 mM EDTA , 0 . 2 mM EGTA , 1 mM DTT , and protease inhibitor cocktail ( Roche ) ( buffer B ) . The insoluble chromatin ( fraction P3 ) and soluble ( fraction S3 ) fractions were separated by centrifugation ( 5 min , 1 , 700 g , 4°C ) . The insoluble chromatin pellet was washed once with buffer B and resuspended in SDS loading buffer . The protein concentration of each fraction was determined by a Bradford’s assay ( Bio-Rad Laboratories ) , and 20 μg of protein from each fraction was immunoblotted . Flow cytometry and DNA content analysis on RNAi treated S2 cells was performed as previously described[62] . RNAi treated S2 cells ( 106 ) were pelleted at 1 , 000 g for 5 min , resuspended in 0 . 5 ml PBS , and vortexed while intermittently adding 0 . 5 ml of cold 100% ethanol . Fixed cells were incubated on ice for 20 min , pelleted ( 1 , 000 g for 5 min ) , and resuspended in a 0 . 5 ml propidium iodide ( PI ) -RNase solution ( 50 μg/ml PI + 100 μg/ml RNase Type1 I-A [QIAGEN] in PBS ) . After 20 min , cells were passed through a 12 × 75–mm flow cytometry tube ( Falcon; Thermo Fisher Scientific ) . Cytometric analysis was performed in the Arizona Cancer Center/Arizona Research Laboratories Division of Biotechnology Cytometry Core Facility using a FACScan flow cytometer ( BD ) equipped with an air-cooled 15-mW argon ion laser tuned to 488 nm . List mode data files consisting of 10 , 000 cells gated on forward scatter versus side scatter were acquired and analyzed using CellQuest Pro software ( BD ) . Fly crosses were performed on yeast/molasses/cornmeal media and kept at 25° . CK1αG0492 ( Bloomington Stock Center # 12303 ) , CK1αEP1555 ( Bloomington Stock Center # 17009 ) , CK1α8B12 ( gift from Yashi Ahmed ) [40] , CK1αRNAi VALIUM10 ( Harvard TRiP line: JF7192 , Bloomington Stock Center # 25786 ) , SMC4k00819 ( Bloomington Stock Center # 10831 ) [63] , Cap-H2Z3–0019 [16] , 43B>Gal4 salivary gland specific driver ( gift from Patrick O'Farrel ) [44] , Slimb3A1 , SlimbUU11 alleles were previously described[20] . | The Cap-H2 condensin II subunit is required for interphase condensin II activity . Previous work has shown that low levels of Cap-H2 protein in interphase is achieved by SCFSlimb mediated protein turnover and limits chromatin-bound protein levels . Here we show that Casein Kinase I alpha ( CK1α ) is also a negative regulator of interphase condensin II activity by promoting Cap-H2 destruction and limiting chromatin-bound Cap-H2 levels . Loss of CK1α function leads to aberrant chromosome structures that are suppressed by mutation or depletion of Cap-H2 or other condensin subunits . These observations suggest that normally , interphase condensin II levels must be kept low in order to maintain proper interphase chromosome organization , and this low activity is maintained by targeted destruction of Cap-H2 . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Drosophila Casein Kinase I Alpha Regulates Homolog Pairing and Genome Organization by Modulating Condensin II Subunit Cap-H2 Levels |
As data for microbial community structures found in various environments has increased , studies have examined the relationship between environmental labels given to retrieved microbial samples and their community structures . However , because environments continuously change over time and space , mixed states of some environments and its effects on community formation should be considered , instead of evaluating effects of discrete environmental categories . Here we applied a hierarchical Bayesian model to paired datasets containing more than 30 , 000 samples of microbial community structures and sample description documents . From the training results , we extracted latent environmental topics that associate co-occurring microbes with co-occurring word sets among samples . Topics are the core elements of environmental mixtures and the visualization of topic-based samples clarifies the connections of various environments . Based on the model training results , we developed a web application , LEA ( Latent Environment Allocation ) , which provides the way to evaluate typicality and heterogeneity of microbial communities in newly obtained samples without confining environmental categories to be compared . Because topics link words and microbes , LEA also enables to search samples semantically related to the query out of 30 , 000 microbiome samples .
Microbial communities are present worldwide in almost all possible environments . Because the composition ( structure ) of a microbial community and its surrounding environment are closely related to each other , it is important to understand what kinds of structural patterns are possible and how environmental factors affect community formations . Over the past decade , the structures of tens of thousands of microbial samples derived from various natural environments , including those in symbiosis with humans , have been analyzed . Using these datasets , global patterns of microbial diversity have been characterized that show that community structures constitute distinct clusters among at least certain environments[1–4] . In addition , the community structure of each examined environment has been evaluated using a clearly defined environmental ontology[5 , 6] . However , the granularities ( i . e . , the levels of detail ) of human-classified environmental categories do not necessarily coincide with structural patterns of microbial communities , and this unavoidable arbitrariness in the granularities of environmental labels may bias the interpretation of results of comparative analysis , such as an enrichment analysis of environments among communities[1] . There are three types of incongruences between environmental labels and community structures . First , there are different subtypes in the microbial community structures associated with certain environments , e . g . , enterotypes in the human gut and vaginal community types[7–9] . Second , in contrast to the first case , a nearly identical community structure may be observed across different environmental labels . For example , microbial communities of the surface of the home environment and their inhabitants show highly similar structural patterns[10] . Third , because an environment varies continuously over time and space , it is impossible to define it using a strict segmentation or hierarchical structure . For example , the brackish water of an estuary can have various mixtures of fresh water and seawater , for which the relative proportion continuously shifts[11] . Although an environment is difficult to definitively define owing to uncharacterized factors , these factors can potentially be defined indirectly using microbial community data because microbes respond quickly to environmental changes[12 , 13] , and their community structure reflects the state of the environment[14–16] . Microbial community structures have been analyzed by various data clustering methods . Most of these approaches are based on the evaluation of data densities on high-dimensional feature space in which microbiome samples are distributed . Microbial community structures are complex as they are described by a large number of features ( taxa ) , although not all the features necessarily vary independently . There are groups of taxa that show co-occurrence patterns in samples[17–19] . Herein , we refer to such co-occurrence relationships of microbes as “sub-communities” . Summarizing community structure data as mixing ratios of sub-communities makes it easier to interpret community dynamics according to environmental changes[20] . To extract such partial structures from mixed data , the machine-learning technique denoted a topic model has been extensively studied in recent years since introduction of the Latent Dirichlet Allocation ( LDA ) model[21] . The LDA model is a probabilistic generative modeling approach , mainly used in natural language processing , for discovering the latent ( unobserved ) structures of the dataset . The LDA model and its extended models have been used to analyze microbiomes[20 , 22 , 23] , however , it is often difficult to interpret extracted sub-communities of microbial taxa . Sub-communities have been characterized by evaluating their relationships with occurrences of sample metadata ( i . e . , data describing information about the samples , such as body sites and gender ) after modeling[24] , or by explicitly modeling associations between metadata and sub-communities[20] . These methods cannot be applied unless all samples have standardized metadata with a uniform granularity , and thus tend not to be practical for comprehensive analyses of microbial samples from various environments . All metagenomic data registered in public databases have such metadata , that is , natural language data described by the researchers who registered the samples . The paired dataset of community structures and description documents can be used for modeling the conditional relationship between them . However , natural language descriptions in databases are not always sufficiently described as their content for many samples is often incomplete and has widely variable resolution . For example , in a sample , various information on the host such as race and gender , experimental conditions , the purpose of the research project , etc . are described , but in another sample , it is described only as “human gut metagenome” and is needed to be treated as a sample with missing values . Therefore , for robust modeling , it is necessary to assume stochastic-generating processes not only for community structures but also for documents within the framework of the probabilistic generative model . For such purpose , the Correspondence-LDA ( Corr-LDA ) [25 , 26] model can be applied . Corr-LDA is a probabilistic modeling approach that is used to extract correspondence between various types of elements occurring in the same dataset , for example , the correspondence between sub-regions of pictures and their captions[25] , between topics of blog documents and their annotation tags[26] , or between brain regions and their cognitive functions[27][28] . In this research , we attempted to find relationships between patterns of microbial community structures and patterns of “environments” that the human recognizes and describes . To this end , we applied the Corr-LDA model to pairs of taxonomic compositions and natural language sample descriptions for tens of thousands of sequenced 16S rRNA gene amplicon samples reanalyzed by the unified analysis pipeline . Using this dataset , “topics” extracted by the Corr-LDA model would represent the core elements of environmental mixtures . By integrating training results , we developed an interactive web application denoted LEA ( Latent-Environment Allocation ) , which is freely available at http://leamicrobe . jp . The extracted connections between microbial sub-communities and subsets of English words via topics are applicable to various analyses . LEA enables researchers to do the following: 1 ) clarify the relationship between environments and patterns of microbial community structures . 2 ) predict the “latent environments” of new samples from , for example , the ocean , a diseased gut , or another unexpected environment , and quickly compare new samples with tens of thousands of existing samples based on their environmental similarity , which makes it easy to detect dysbiosis of the microbiome in the human gut or contaminants in natural environments . 3 ) search for samples in the >30 , 000-sample dataset based on an ecological perspective , without depending on exact word matching of queries and sample descriptions . In this paper , we show the patterns found in the human gut and vaginal communities as an example of separations and connections of extracted “environments” , and show how the LEA global map and a semantic search method on the map make it easy to explore patterns in microbial community structures . In addition , as an example of environmental predictions for newly acquired microbiome samples , we show the LEA mapping results for the datasets of the human gut microbiome and the microbiome derived from various natural environments .
We collected sequenced 16S rRNA gene amplicon samples from the MicrobeDB . jp database and performed a phylogenetic analysis using VITCOMIC2[29] , which is the metagenomic analysis pipeline improved from VITCOMIC[30] , on all samples . This resulted in a dataset containing 30 , 718 samples with genus level information on their taxonomic composition linked to a document containing sample description information . For this dataset , model inference runs were performed by Corr-LDA with a varying number of topics . The perplexity , which is the performance evaluation index of the model ( smaller values indicate a better performance ) , was sufficiently small for the model with 80 topics ( S1 Fig ) . In the following sections , we discuss the results for the model inferred with 80 topics . The inferred word subsets and microbial sub-communities for each topic are shown in S2 and S3 Figs . Each topic has a unique subset of words and a microbial sub-community . The structure of a microbial community sample that has a large proportion of a certain topic is likely to contain microbes in the sub-community of that topic , and the description of the sample is likely to contain words in the word subset of that topic . For most topics , the word subset associated with the topic represents a single natural environment or a symbiotic environment with humans . Based on the topic composition of each sample , the similarities among the samples were visualized using parametric t-SNE[31] . In Fig 1 , the dots represent the 30 , 718 samples and the pictures represent 80 topics . A sample that is mapped near a picture indicates that the sample has a large proportion of that topic . On the map in Fig 1C , the sample ( SRS425923 ) , which is located approximately midway between the two pictures , has the two topics ( topic #37 and #52 ) mixed in similar amounts . The 80 topics can be regarded as latent environments that affect the formation of microbial community structures . Topics form several clusters with dense connections formed by many samples but with sparse or no connections between clusters . Topics can be roughly divided into gut , skin , vagina , oral cavity , ocean , and soil . In addition , there are several isolated topics , which include a coral reef , a mosquito , phyllosphere , etc . The gut microbiome in healthy adult humans are reported to consist of three[7] or four[24] community types . However , whether truly discrete clusters exist as individual gut microbial communities remains in doubt[32] . For the gut community types ( enterotypes ) , the key genera characterizing each community type have been identified—Bacteroides , Prevotella , and Ruminococcus[7]—although the abundance of key genera varies between samples instead of being discretely clustered[32 , 33] . Therefore , unlike discrete clusters , e . g . , blood types , the compositions of microbial communities are continuously shifting , perhaps as a result of environmental factors . Such continuous variation of the structure of a microbial community can be discerned by our method . 22 topics are related to the gut according to the word subsets associated with each of the topics ( S8 Fig ) , including those with a large proportion of Bacteroides , Ruminococcus , and Prevotella ( topics #79 , #51 , and #24 . Fig 1B ) . However , most of the samples do not reside near a single topic but instead occupy an intermediate position between multiple topics , meaning that the samples are a mixture of several topics . Because many samples with intermediate properties owing to multiple topics exist , there is variability across a limited area of the gut microbiome . Bacteroides is often found in the guts of people who eat diets rich in protein and fat , and Prevotella is often found in the guts of vegetarians[12 , 34] . In fact , words denoting herbivores , e . g . , “pig” , “swine” , “horse” , “bovine” , and/or “rumen” , are frequently found in the Prevotella-rich topic and in topics that are peripherally connected to the Prevotella-rich topic ( Fig 1B ) . Regarding the vaginal flora , three related topics were found ( Fig 1C ) : the Lactobacillus-rich topic ( #37 ) ; the topic including Gardnerella , Sneathia , and Atopobium ( #52 ) ; and the Shuttleworthia-rich topic ( #43 ) . Vaginal community types ( Community State Type; CST ) have previously been examined in detail with five CSTs recognized to date: four ( CSTs I , II , III , and V ) in which Lactobacillus species dominates and one ( CST IV ) with various obligate or facultative anaerobes and very few Lactobacillus[9 , 35] . The two topics detected in our model are consistent with the above results ( Fig 1C ) . Because we used community structure data found at the genus level , we cannot distinguish between CSTs I , II , III , and V , so these communities were identified as a single topic dominated by Lactobacillus . For the second topic corresponding to CST IV in which Gardnerella and Atopobium dominate , the associated samples likely were obtained from African-American women ( as estimated by the word subset of topic #52 ) . Interestingly , samples in which the Lactobacillus-rich and Gardnerella-rich topics are mixed in various proportions are frequently found , as indicated by the many dots that connect these two pictures ( Fig 1C ) . Vaginal bacterial communities are known to be stable throughout pregnancy and to be relatively stable throughout the menstrual cycle although changes in the Lactobacillus spp . populations have been observed[36 , 37] . Therefore , an environmental gradient of unidentified factors may exist in the vagina , which would cause a community structure to exist as an intermediate state between two topics . Another topic related to the vaginal environment is a topic dominated by Shuttleworthia . The presence of Shuttleworthia may be related to bacterial vaginosis[38] or to squamous intraepithelial cervical lesions[39] , but its ecology is not well understood . Interestingly , the continuous transition of samples to the Shuttleworthia-rich topic links only with the Gardnerella-rich topic ( Fig 1C ) . LEA can predict latent environmental topics of newly acquired samples using the Bayesian prediction method with the identified 80 topics ( see Methods ) . By examining the word subsets associated with the mixed topics , the environment in which new samples are found can be estimated . In addition , by training the dimension-reduction function of t-SNE in our system using a neural network procedure , it is possible to arrange the locations of new samples on the global map ( Fig 1A ) according to their topic compositions without changing the coordinates of previously mapped samples . The topic prediction of a new sample and its placement on the map are implemented by the LEA web application . By uploading the taxonomic assignment file of VITCOMIC2 , the placement of the sample on the map can be viewed in a few seconds . As examples of LEA mapping results , we have analyzed the dataset of a time-series human gut microbiome analysis[40] , which consists of fecal samples obtained every day from two male subjects from the US ( subjects A and B ) . The results are shown in S4 Fig . The LEA visualization reproduces the results of David et al . [40] , such as the stability of the gut microbiome of subject A over the course of the experiment with the exception of his time in Southeast Asia , and the change in the gut microbiome of subject B caused by an infection . Such results can be easily obtained using the LEA web application . In a meta-analysis of a large-scale dataset , the existence of systematic bias due to the difference in methods across studies often becomes a problem . We tried to address this problem by processing all samples with a unified information analysis pipeline , but there is a possibility that a further upstream , sample preparation protocol could be a confounding factor . In particular , a bias due to differences in DNA extraction methods often becomes a problem[41] . To assess the impact of different DNA extraction methods of the human gut microbiome analysis on the locations on LEA global map , we conducted LEA environment predictions for the Microbiome Quality Control ( MBQC ) dataset[42] . This dataset contains 16S amplicon sequencing data from human stool samples , chemostats , and artificial microbial communities . For the same biological sample , there are multiple sequencing data analyzed with different wet laboratories or different DNA extraction methods . The results are shown in S10 Fig . First , most of the samples derived from human feces in the MBQC dataset were properly mapped to the “gut” area of the LEA global map ( S10A Fig ) . As a whole , there was no tendency for samples processed with a specific DNA extraction kit to be mapped only to a specific topic . Therefore , separation of topics on LEA is not necessarily influenced by differences in experimental protocols . However , considering samples that have a same biological origin , some samples were mapped to the nearly same position on the map , and the others were mapped on the location biased by the DNA extraction kits ( S10B–S10S Fig ) . The direction of biases probably differs depending on the position of the true taxonomy composition . Therefore , topics may partially contain systematic bias due to differences in studies , and caution is necessary for interpretation . As an example of LEA involving a natural environment , Fig 2 shows the topic predictions for 38 samples of microbial communities obtained over a short period of time and at a high density from various points along the Tamagawa river in Japan . The upstream region of the river begins in a deep mountainous region; the middle region flows through a densely-populated zone where there is water from sewage treatment plants and from tributaries that joins the river; and the downstream-most region flows into Tokyo Bay . On May 26 and 27 , 2015 , we sampled the surface water of the river at 38 points ( S5 Fig , S1 Table ) and identified the microbes contained in the samples by VITCOMIC2 after sequencing of their PCR-amplified 16S rRNA genes . The genus level taxonomic compositions are shown in Fig 2A . Limnohabitans is the major genus found in the samples from the river . The microbial community structure of the river continuously shifted as it flowed to the estuary , but sample 200 had a greatly different structure . Sample 200 was obtained just under the sewage treatment facility , and its composition probably reflects the microbiome of the treated water . The community structures of samples 10 , 20 , and 30 , which were obtained from brackish water in the estuary , also differed greatly from those collected elsewhere along the river . The topic predictions for the river samples are shown in Fig 2B . After performing LEA , two topics related to “river” were found: one was topic #3 , which frequently occurs together with words such as “Baltic sea , ” “lake , ” and “river , ” and the second was topic #53 , which is associated with the words , “river , ” “wastewater , ” and “urban . ” The aforementioned words belong to the dominant topics in Fig 2B . Topic #3 is primarily associated with the upstream region of the river and topic #53 with all areas of the river . The relative proportions of these two topics gradually change as the river flows downstream . Given the word subsets associated with the two topics , topic #3 represents freshwater ecosystems , such as lakes and rivers , and topic #53 represents river areas adjacent to cities . Samples 10 and 20 ( from the estuary ) are largely associated with topics #11 and #63 , which represent the ocean , and , along with sample 30 , are associated with topic #56 , which represents activated sludge . For the Tamagawa river , about half its water that flows into the estuary is treated water[43] . Thus , our results suggest that the mixing of the river water with seawater greatly changes the community structure and that the river’s ecosystem is greatly affected by it interaction with the urban area . Topic #45 is associated with the upstream region of the Tamagawa river ( sample 360 to 250 ) ; the words associated with this topic include “pet” and an indoor environment ( Fig 2B ) . Many of the 16S rRNA sequences associated with topic #45 belong to Blastomonas , a genus associated with domestic wastewater , which is found in tap water , faucets , and shower hoses and is resistant to disinfection[44–46] . Advanced sewage treatment facilities are not found in this region of the Tamagawa river , suggesting that untreated household wastewater is being dumped into the river . Most of the Tamagawa river samples were mapped near “freshwater” topics #53 and #3 on the global map ( Fig 2C ) , although the topics of sample 200 , taken near the sewage treatment plant , and samples 10 , 20 , and 30 , taken from the estuary , diverged , to some extent , from the freshwater topic . Specifically , sample 10 were mapped within “ocean” topics . In this way , the LEA web application can place a new sample appropriately on the global map of existing samples and enables visual and intuitive operation to evaluate its characteristics , e . g . , deviations from the expected environments . For further testing of LEA using external dataset , we conducted environmental predictions of microbiome data derived from a highly diverse environment produced by the Earth Microbiome Project ( EMP ) [47] . One of the good points about this dataset is that every sample is given an environmental label based on a controlled vocabulary , the EMP Ontology ( EMPO ) . EMPO is a hierarchical framework that captures the major axis of the microbial community diversity and is used to assign samples of EMP to its habitat[47] . Therefore , by comparing the result of LEA mapping with each EMPO label , we can estimate the accuracy of environmental prediction by LEA . For each of the lowest layer label ( level 3: most specific habitat name ) of EMPO , we examined the location of the samples given that label on the LEA map . The results are shown in S11 Fig . Environmental prediction results have well captured the influence of salinity known as the main axis that determines the community structure[2] . For most samples of water , sediments , biofilms , and soils , saline samples were mapped around the ocean area , and non-saline samples were mapped to freshwater or soil area ( S11A–S11H Fig ) . Regarding the samples derived from the host-associated environments , it was observed that the mapping pattern varied depending on the host species even with the same EMPO label . Also , the EMPO label “Plant surface” intuitively evokes leaf surface of land plants , but most of the EMP samples labeled with “Plant surface” mapped to “ocean” area on LEA . This is because most of the EMP samples used in this study with the label “Plant surface” are derived from the kelp as the host ( S11J Fig ) . In such a case , environmental prediction by LEA gives interpretable results ( microbial communities on the kelp surface reflects the oceanic community structure pattern , etc . ) . When the environmental ontology and the community structure pattern seem to conflict , LEA can be used to infer the reason from the mapping results . The topic-model approach can be used to semantically search documents related to a user’s query[48 , 49] . By using the trained model parameters in LEA , existing samples can be searched using natural language such as “forest soil” , or “hot spring” . Instead of needing to search for samples by exactly matching the queried words and the description information associated with samples , we can find the sample using latent environmental topics , using the probability of each sample to generate the query sentence as the score of the sample . As an example , Table 1 shows the top five scoring samples obtained by querying “What kind of microorganisms are in an oil sands tailings pond ? ” Oil sands tailings ponds are slag ponds accompanying oil sand development and are highly toxic environments as they contain heavy metals , naphtha , bitumen , and other toxic chemicals . Tailing ponds have heterogeneous environments , being aerobic at their surfaces and anaerobic at their bottoms . Many members of the class Methanomicrobia , including Methanoculleus , Methanolinea , Methanosaeta , Methanobrevibacter , and Methanocorpusculum , which are methanogenic archaea found in the sediment of tailing ponds , contribute to the decomposition of hydrocarbons[50 , 51] . Given the query , “What kind of microorganisms are in oil sands tailings ponds ? ” , LEA returned the samples derived from oil sand tailing ponds and oil-water mixtures ( Table 1 ) . In addition , LEA returned the sample derived from ocean sediments , although the description of this sample did not contain words such as “oil , ” “sand , ” or “pond . ” Although the microbial community structures of these samples varied in terms of their taxonomy , almost all were composed of methanogenic archaea . Within the machine-learning process , these members of Methanomicrobia are considered simply as variables in the microbial community structure data and their shared characteristics are not recognized ( although humans would recognize properties common to all of them given that “Methano-” is at the beginning of each of their names ) . All high-scoring samples were associated with a large proportion of topic #8 related to methanogenesis ( S2 Fig ) . Therefore , the fact that samples containing many methanogenic archaea were retrieved after querying for “oil sands tailing ponds” indicates that LEA can automatically extract the following two linkages: 1 ) the association between words such as “hydrocarbon , ” “oil , ” “tailing , ” and “methane , ” and the latent environmental topic that represent “methanogenesis , ” and 2 ) the association between methanogenic archaea of various genera and the latent environmental topic that represent “methanogenesis” .
For this study , we applied a correspondence topic model to more than 30 , 000 samples of microbial community structure data and extracted the latent environments of each sample as topics . By doing so , we obtained microbial sub-communities that can be regarded as “base variables” to describe an entire dataset and associated word subsets that characterize the environments corresponding to the base variables . By visualizing each sample , which is expressed as a linear combination of these base variables , in two-dimensional space , LEA clarifies continuous variation of the microbial community structures linking two or more environments . The difference between continuously connected environments and an isolated environment might mean that only a few samples have been characterized that bridge the isolated environments . Such environments currently include wastewater , the phyllosphere , and environments related to insect symbiosis . Conversely , human-related environments have been vigorously sampled , and therefore we believe that the visualization reported in this manuscript represents a nearly complete picture of those environments related to healthy human adults . Using the extracted environmental topics , LEA can infer what mixture of core environments influenced the taxonomic compositions of newly acquired samples . In the river microbiome analysis , we showed that samples taken from the brackish water area of the Tamagawa river can be expressed as a mixture of a “freshwater” topic , a “seawater” topic and a “wastewater” topic . Environmental prediction of new samples is performed by a Bayesian approach similar to that used in a microbial source tracking algorithm[52] , but using topic sub-communities extracted from a large-scale dataset as source communities , instead of using the samples pre-specified by a user as sources . This allows to compare new samples virtually with tens of thousands of samples related to diverse natural environments and human body sites . Environmental prediction of new samples may be done by fixing the granularities of environmental labels to be used and comparing with samples to which those labels are added in advance[53] . In such a method , however , it is difficult to set the level of granularities , especially when there are multiple structural patterns of microbial communities in a single environment . When analyzing the dynamics of community structures in a single environment , for example , the time series analysis of human gut microbiome or the spatial distribution of river microbiome , it is more useful to use fine-grained environmental labels than to use simple labels such as “river” or “human gut” . Our method clarifies the structural patterns naturally existing in various environments and provides the way to evaluate how new samples transit among them . By using a neural network algorithm that maps the data to a two-dimensional space , LEA can position new samples onto the existing global map . This mapping system can be regarded as a microbial global positioning system[54] used to specify the position of a new sample based on the positions of existing sample and allows a user to intuitively evaluate the properties of new samples . Dysbiosis , a deviation from the ordinary distribution of a microbial community structure that exists in symbiosis with humans , has been discussed in relation to diseases , especially those of the gut[55] . Because a newly acquired sample , such as one from an ill patient , can be located anywhere on the map , identifying the ideal end-point from a clinical perspective and defining its vector may be useful information when choosing a specific treatment that can transition its community structure to another state[54] . To perform comparative metagenomics based on environmental information , a huge amount of environmentally labeled data ordered as a dataset is required . However , manually labeling such data is nearly impossible , as the amount of available data is increasing too rapidly at present . In addition , as microbial community structures from new environments are characterized , much work will be needed to design the ontologies of the corresponding environmental labels at the appropriate granularities while incorporating all new environments . Furthermore , because binary environmental labels ( presence or absence of an environmental property ) are often used to characterize the samples , it is not possible to manually and appropriately evaluate samples that have intermediate properties associated with several environments . Our method automatically extracts the relationship between microbes and their environments by assuming that the microbial community structure and the natural language description for a given sample are both generated from a state in which several environments are mixed . The accuracy of the model should increase as more training data are incorporated . Prior to extending our method for future work , several problems must be solved . First is how many topics are needed to model microbiomes in highly diverse environments . The number of topics in this study , 80 , is an arbitrarily determined value in a sense . In fact , the prediction accuracy of the model for the validation set shows that 80 topics are still inadequate and that a more accurate model can be constructed by setting the more number of topics ( perplexity , S1 Fig ) . However , increasing the number of topics may lead to overfitting , and too large a number of topics may make the map difficult to visualize and interpret . Therefore , we aimed to explain the data with as few topics as possible while keeping the overall prediction accuracy . We are not claiming that microbial communities can be explained by a combination of 80 patterns . The model used in this study is a practical choice to facilitate the interpretation of the whole picture of the microbial community structures and to provide a tool to explore interesting patterns . In the future , as the number of samples acquired from various environments increases , it will be necessary to set a larger number of topics . Nevertheless , from the results of experiments with a large number of topics , the characteristics of the environments considered in this paper ( i . e . , the gut and vagina ) are robust , and an increase in the number of topics would mainly lead to the generation of a topic related to a single research study ( e . g . , “whale skin” as part of a new topic separated from topic #14 , which contains “coral reef” ) . Second is a sampling bias among environments . Depending on the environment , the number of samples studied so far varies greatly . Since LEA determines topics based on the prediction accuracy of the validation set samples , LEA tends to express an environment with a large number of samples in high resolution by placing many topics in it , while environments with a small number of samples tend to be compressed into a limited number of topics . Currently , the environment related to humans , especially to the intestinal tract , have high resolution , but the environment related to freshwater or hot springs has relatively few samples and the separation of topics can be insufficient ( S9 Fig ) . Therefore , attention should be paid to interpretation concerning LEA predictions for such natural environments in which relatively a few number of samples were studied . By incorporating large-scale data of microbiomes sampled from the natural environment , such as data from the Earth Microbiome Project[47] , high-resolution topics will be obtained for those environments in the future . The third problem is , though common to any meta-analysis , a bias due to the difference in experimental methods . In the environmental prediction results of the MBQC dataset by LEA , it was hardly the case that the difference in experimental methods yields a large difference in coordinates on the LEA global map . Nonetheless , depending on the area on the map , such a bias may have the effect of placing the sample close to a particular topic . Ideally , there should be as much data as possible processed in a unified experimental protocol . We do not know if all topics used for our research specifically express a set of related environmental parameters , except for topics such as those representing “methanogenesis . ” In the future , the environmental parameters that determine the microbial community structure will need clarification that can be accomplished by examining the relevance of various metadata and topic compositions and transitions .
Genus-level taxonomic composition data for 48 , 873 sequenced 16S rRNA gene amplicon samples were obtained from MicrobeDB . jp ( http://microbedb . jp/MDB/ ) , which is an integrated , publically available database for microbes in which all metagenomic data registered in the International Nucleotide Sequence Database Collaboration Sequence Read Archive ( SRA ) prior to August 2014 are stored . MicrobeDB . jp includes reanalyzed taxonomic compositions of all samples using the unified analysis pipeline . Briefly , for all original sequence data registered in the SRA , we trimmed adapter sequences and filtered out low-quality sequences , sequences derived from the PhiX genome , and sequences derived from the human genome . For all high-quality sequences in each sample , we searched for similar sequences in the VITCOMIC2[29] database ( http://vitcomic . org ) using CLAST[56] and then performed taxonomic assignments on these sequences . Finally , the number of sequences assigned to a specific genus were summed for each sample . As a result , we obtained a dataset for which each SRS ID ( identifier associated with a sample in SRA ) had a genus-level taxonomic composition . All samples with <1 , 000 sequences with an assigned genus were discarded , and samples with >10 , 000 assigned sequences were subsampled as 10 , 000 sequences . To obtain metadata for the samples , XML files for the SRS IDs registered prior to August 2014 were downloaded from the SRA database . The XML file for each SRS ID contains descriptions of the properties of the corresponding sample , such as pH values and a description of the environment from which the sample was obtained . We extracted the text sandwiched between all tags ( e . g . , research titles , scientific names , geographical locations , and sample descriptions ) in those XML files , lemmatized all words , and organized them according to a bag-of-words model ( with counts of the number of times each word appears ) . We removed all English stop words from the extracted text . Additionally , all words corresponding to any of the following conditions were removed: 1 ) words that contain numbers or symbols that are not alphabetic ( in many cases these were the project-specific sample identifiers ) ; 2 ) words in which the letters “A” , “T” , “C” , or “G” occupied ≥70% of the word length ( often tag sequences or primer sequences ) ; 3 ) words generally used in many samples , e . g . , genome and metagenome ( S2 Table ) ; 4 ) words used only for a single research study; and 5 ) words that appeared <20 times or in >30% of the samples in the dataset . Samples with all words removed according to the aforementioned conditions were themselves discarded . Finally , we integrated the taxonomic composition and sample description datasets to construct a dataset consisting of only samples containing both features . This dataset contains 30 , 718 samples , each of which has a taxonomic composition ( a count of each genus in the sample ) and a description of the properties of the sample ( bag-of-words ) . The number of genera in the final dataset ( vocabulary of taxonomic composition data; S3 Table ) is 1675 , and the number of unique words in the dataset ( vocabulary of sample description data; S4 Table ) is 764 . The final dataset is publicly available at http://palaeo . nig . ac . jp/Resources/lea2018/ . We assumed that each sample had a multinomial distribution of latent environments and that both the observed genera and the sample description document were generated according to the mixing ratio of those environments or topics . To infer topics , we applied Corr-LDA[25 , 26] to the dataset . In our LEA system , Corr-LDA model was applied to our dataset as described below . Each sample d ( d = 1 …D ) has taxonomic composition data wd = {wdn} ( n = 1…Nd ) and description data td = {tdm} ( m = 1…Md ) . Here , Nd is the number of sequences with an assigned taxonomy in sample d , wdn is the taxonomy assigned to the nth sequence in sample d , Md is the number of words in the description given to sample d , and tdm is the mth word in sample d . In a topic model , it is assumed that all elements in the data have a latent topic . The latent topic of the nth sequence in sample d is defined zdn ∈ {1 …K} , and K is the number of latent topics that have been pre-specified . The latent topic of the mth word in sample d is defined cdm ∈ {1 …K} , with K again representing the number of latent topics , which is the same for both sequences and words . Topics assigned to each sequence and each word is undefined prior to analysis and inferred using the entire dataset . For the entire dataset , the joint probability distribution for the taxonomic composition data W , the description data T , the latent topics Z for sequences , and the latent topics C for words was written as follows: P ( W , T , Z , C|α , β , γ ) =P ( Z|α ) P ( W|Z , β ) P ( C|Z ) P ( T|C , γ ) where α , β , and γ are hyper-parameters of prior distributions for topics , taxa , and words , respectively . We assumed that latent topics Z for genera appearing in each sample d is generated according to the multinomial distribution θd and that the prior of θd is the asymmetric Dirichlet distribution having αz ( z = 1 …K ) as hyper-parameters . θd can be integrated out and P ( Z|α ) can be written as follows: P ( Z|α ) =∏d=1D∫P ( Zd|θd ) P ( θd|α ) dθd= ( Γ ( ∑z=1Kαz ) ∏z=1KΓ ( αz ) ) D∏d=1D∏z=1KΓ ( Nzd+αz ) Γ ( Nd+∑z=1Kαz ) where Nzd is the number of sequences that are assigned to a topic z in sample d and Г is the gamma function . The genera appearing in a sample is generated according to a multinomial distribution φz when its latent topic is z . φz can be interpreted as a sub-community of genera associated with latent topic z . We assumed the symmetric Dirichlet prior for φz . φz can also be integrated out and P ( W|Z , β ) can be written as follows: P ( W|Z , β ) =∏z=1K∫P ( W|z , φz ) P ( φz|β ) dφz= ( Γ ( βV ) Γ ( β ) V ) K∏z=1K∏w=1VΓ ( Nzw+β ) Γ ( Nz+βV ) where Nzw is the number of sequences assigned to genus w with a topic assigned to z , Nz is the number of sequences assigned to a topic z , and V is the number of unique genera in the dataset . Topics for words in sample description data were generated according to the following: cdm∼Multinomial ( {NzdNd}z=1z=K ) where Nd is the number of genus-assigned sequences in sample d . Thus , topics for words are conditional on topics assigned for genera . A word appearing in a sample is generated according to the multinomial distribution ψc when its latent topic is c . ψc can be interpreted as a subset of words representing latent topic c , and we assumed the symmetric Dirichlet prior for ψc . As in the case of P ( W|Z , β ) , P ( T|C , γ ) can be written as follows: P ( T|C , γ ) =∏z=1K∫P ( T|c , ψc ) P ( ψc|γ ) dψc= ( Γ ( γT ) Γ ( γ ) T ) K∏z=1K∏t=1SΓ ( Mzt+γ ) Γ ( Mz+γS ) where Mzt is the number of words t with a topic assigned to z , Mz is the total number of words assigned to a topic z , and S is the number of unique words in the dataset . We assumed the asymmetric Dirichlet prior only for the topic multinomial distribution , and the symmetric Dirichlet prior for the genera and word multinomial distributions because samples for which their microbial community structure had been analyzed previously are more likely to have been acquired from human symbiotic environments , suggesting that a bias might also exist for the topic occurrence probabilities . It was reported that setting an asymmetric Dirichlet prior for a topic distribution is effective for robust inference of a topic model[57] . The posterior distributions of the latent topic Z for genera and the latent topic C for words were approximated by the collapsed Gibbs sampling method[26 , 58] . Hyper-parameters ( α , β , and γ ) were updated in each step during the Gibbs sampling by the fixed-point iteration method[59] . To determine the number of topics K , we randomly divided the dataset into 25 , 718 samples as a training set and a group of 5 , 000 samples as a test set , and then the test set perplexity was calculated using the results from the training set with varying numbers of topics . Perplexity is an index for measuring the predictive performance of a held-out test set , and the smaller the value , the better the performance . We ran five Markov chains using different initial values and inferred the model parameters . Next , using the model parameters inferred from the training set , the topic composition of each sample in the test set was estimated using 50% of the sequences in each sample , and the generation probability of genera assigned to remaining 50% of the sequences was calculated . S1 Fig shows the average perplexities and standard deviations obtained by inference with five Markov chains for 5 to 300 topics . Setting too large a number of topics reduces the interpretability of the training results and might cause overfitting for existing samples , so we fixed the number of topics as 80 . Finally , we ran a Markov chain using 30 , 718 samples with the number of topics set at 80 and acquired the topic composition θ for each sample , the genera probability φ for each topic ( microbial sub-community in each topic ) , and the word probability ψ for each topic ( word subset corresponding to each topic ) when the joint likelihood converged after a sufficient number of Gibbs iterations . The implementation of Corr-LDA used in this study is available at https://github . com/khigashi1987/CorrLDA . Parameters used were -I 1000 -K 80 ( 1000 Gibbs iterations and 80 topics ) . Word subsets and microbial sub-communities associated with each topic are listed in S2 and S3 Figs . All word-cloud images were generated using the word-cloud generator in Python ( https://github . com/amueller/word_cloud ) . The structure of sub-communities differs greatly between most topic pairs ( S6A Fig ) . The topic pair with the most similar sub-communities is formed by the soil-environment topics #19 and #54 , although the structures of these topics are still greatly different ( S6B Fig ) . Similar comparison on word subsets of topics shows that there are many more similar topic pairs with respect to word probabilities ( S6C Fig ) . The topic pair with the most similar word subsets is formed by the vagina-associated topics #43 and #52 ( S6D Fig ) . In both topics , two words , “female” and “vaginal” , dominate greatly in their probabilities and reflect similar environmental concepts ( “vagina” ) . However , in a topic #52 , two words , “african” and “american” , have relatively large probabilities , whereas in a topic #43 the probabilities of those words are very small ( S6D Fig ) . Unlike microbial sub-communities of topics , it is natural that there are multiple similar topics for word-subsets . That is because there are some environments which have different patterns of the community structures but are difficult for the human to distinguish and describe those , as with the case of the human gut environment ( enterotypes[7] ) . Nonetheless , like these vagina topics , some topics may reflect slight differences that appear in the sample descriptions . As for what kinds of environment the topics show , it is strongly influenced by the sampling bias by the previous research . The vigorously studied environment can be modeled at the high resolution , resulting in a large number of topics associated with the environment . We investigated the number of topics related to a specific environment by calculating the generation probability of specific words for each topic . If the topic had a probability of generating the word “gut” above 5% , we considered that topic to be related to the gut-associated environment . Similarly , we calculated the sum of the generation probabilities the words “oral” , “cavity” , “dental” , “plaque” , “gingiva” , “tonsil” and “saliva” for the oral cavity-associated environment , the word “skin” for the skin-associated environment , the words “vaginal” and “vagina” for the vagina-associated environment , the words “marine” , “sea” , “ocean” , “seawater” and “saline” for the ocean-associated environment , the words “freshwater” , “lake” and “river” for the freshwater-associated environment , the words “soil” , “agricultural” and “field” for the soil-associated environment , and the words “hot” and “spring” for the hot spring-associated environment for each topic . As a result , there were many topics associated with the gut environment ( 22 topics ) , and the number of topics related to the natural environment tended to be small ( S8 Fig ) . When examining the number of samples with more than 50% of each “environment-associated topics” , similar trends were observed ( S9 Fig ) . After the above process , all samples in the dataset were expressed as 80-dimensional , real-valued vectors showing topic compositions . In general , for visualization of 80-dimensional vectors , it is effective to arrange sample points in two or three-dimensional space by dimensionality reduction , and various dimensionality reduction methods , such as principal component analysis or the multidimensional scaling method , can be applied . For this approach , we used t-SNE ( t-distributed Stochastic Neighbor Embedding ) [60] , which embeds sample points in a low-dimensional space while maintaining the local structures between sample points in the original high-dimensional space . However , simple t-SNE method is inadequate for newly acquired samples . When predicting the topic composition of a new sample and comparing it with existing samples , the distance calculation and the cost minimization for the entire dataset must be performed again with the new sample , which might cause a random change of the coordinates in the low-dimensional space each time a new sample is added . Therefore , we adopted the parametric t-SNE method[31] , which trains a function with the same behavior as t-SNE using a neural network . This neural network inputs an 80-dimensional vector and outputs two-dimensional coordinates . Weights of the neural network are trained with the same loss function as for the normal t-SNE . We constructed a four-layer , feed-forward neural network ( the number of nodes is 80 , 160 , 160 , 640 , and 2 ) . We used the Rectified Linear Unit as the activation function of the nodes in all layers except the last one and the linear activation function in the last layer and trained the weights using the mini-batch stochastic gradient descent method . We implemented parametric t-SNE by Theano[61] and Keras ( https://github . com/fchollet/keras ) and used Adam[62] as the optimizer . After a sufficient number of epochs , we obtained the coordinates of all samples by inputting topic compositions of samples into the neural network and then visualized samples in a scatter plot . At the same time , we obtained the coordinates of topics in two-dimensional space by inputting one-hot vectors ( 80-dimensional vectors with a single 1 value and 0 for all other values ) into the neural network . On those coordinates in a scatter plot , we mapped pictures corresponding to the word subset of each topic . Thus , the sample plotted in a position close to a given picture means that the mixing ratio of that topic is very high for the sample . All pictures used in this study are in the public domain . The topic composition of a new sample was estimated using the taxonomic composition of the new sample , the hyper-parameter α , and the genera generation probability φ , learned from the training samples . First , the DNA sequence data from a new sample was analyzed by VITCOMIC2 and converted into taxonomic composition data . The estimation of the topic composition in a Bayesian approach requires consideration of all possible assignments of sequences to topics , but direct estimation of the posterior distribution of the topic composition is intractable[52] . In LEA , the posterior distribution of the topic composition of a new sample is approximated by performing Gibbs sampling with randomly initializing topics assigned to sequences in the new sample . For the genus w assigned to the nth sequence of a new sample d , the latent topic z was sampled according to the following conditional distribution: P ( zdn=k|W , Z∖dn , φ , α ) ∝φkwNkd∖dn+αkNd∖dn+∑z=1Kαz where φkw is the generative probability of genus w when zdn is k , αk is the hyper-parameter of the topic probability; Nkd\dn is the number of sequences assigned to topic k , except for the nth sequence in sample d; and Nd\dn is the total number of sequences in sample d minus one . This distribution is similar to the sampling formula used in SourceTracker[52] , but LEA uses microbial sub-communities of 80 topics as source communities , instead of using a sample set specified by the user as the source in SourceTracker . LEA estimates the topic composition of the new sample using only sequence information because we assume that the primary use of LEA is to predict the environment of the sample for which description information is not available , or , more importantly , to predict contamination of the unexpected environment . After a sufficient number of Gibbs-sampling iterations , the topic composition of the new sample was expressed as an 80-dimensional vector and converted to the coordinates on a two-dimensional map by the feed-forward neural network learned with training samples . By arranging the transformed coordinates of the new sample on the same two-dimensional map as the existing samples , we could compare the features of the new sample with those of all samples in our database . We provide REST ( representational state transfer ) -style APIs ( application programming interfaces ) to access all the model parameters of LEA and the environmental prediction function of the user sample at http://snail . nig . ac . jp/leaapi/ . With the HTTP POST method , environmental predictions for the new samples can be calculated from the command line interface without going through the web application . There is also an API that provides information on the instability of the posterior distribution calculation of topic proportions for the new samples . The function computes topic proportions by drawing 100 samples from 100 independent Gibbs sampling chains with different initial settings and by averaging 100 samples . This also provides standard deviations of each topic proportion estimates . These values can be regarded as “goodness of fit” of the new samples to the LEA model . In the calculation results on the Tamagawa river microbiome , the topic proportions shown in Fig 2 was stable at any point ( S7 Fig ) . The training results from the 30 , 718 samples and their visualization are available as an interactive web application at http://leamicrobe . jp . The back end of the application is written in C language and Python , and the front end is written in JavaScript . Users can examine the taxonomic and topic compositions of each sample , and the sub-community and word subset of each topic on the global map . By uploading the taxonomic composition data obtained via VITCOMIC2 , it is possible to place user samples on the global map after several seconds of calculation for estimating topic compositions . Because LEA accepts only taxonomic composition data generated by VITCOMIC2 , VITCOMIC2 must be used to obtain taxonomic compositions from raw sequence data . Furthermore , with the use of the training results in our model , it is possible to perform a sample search that does not depend on exact matching between the query text and sample metadata . The sample-retrieval process uses the word generation probability ψ in each topic and the topic composition θ in each sample as learned with the existing samples . The search query consists of free words , i . e . , several English words or English sentences . First , the search query is divided into words . If the search query contains words that do not exist in the vocabulary of LEA ( 764 words ) , those words are simply discarded . LEA also carries out preprocessing of query words ( removal of English stop words and lemmatization ) same as the training step of LEA model . After that , the score of sample d is calculated using the following equation: Score ( d ) =P ( q|d ) =∏n=1N∑z=1KP ( qn|z ) P ( z|d ) =∏n=1N∑z=1Kψzqnθdz where q = {qn} ( n = 1 to N ) is a search-word set and N is the number of valid words in the search-word set . This means that we use the probability that the sample d generates the search query q as the score of the sample d . LEA calculates scores of all samples in LEA dataset ( 30 , 718 samples ) and sorts them in descending order . By entering free words in the search window of the web application , it is possible to highlight samples with a high score , meaning that those samples are semantically related to the search query . If the taxonomy name of a microbe ( or its synonym ( s ) ) is included in the search query , the score is scaled according to the abundance of that microbe in each sample . If the search query contains the name of higher taxa than a genus level , the score is scaled according to the sum of abundances of microbes below the input taxa ( taxonomy based on NCBI Taxonomy ) . This makes it possible to search samples by queries such as “Gemmatimonadetes in the ocean” . As examples that show how LEA mapping can be used , we acquired four microbiome datasets . 1 ) Intestinal microbiota data from two subjects over the course of a year[40] . 2 ) Intestinal microbiota data obtained by various DNA extraction methods from Microbiome Quality Control Project[42] . 3 ) Microbiota data from various natural environments from Earth Microbiome Project[47] . 4 ) Microbiota data from the upper region to the estuary of the river Tamagawa . The intestinal microbiota data are those reported by David et al . ( PRJEB6518 ) [40] . In that study , every day for all or most of a year , the microbial 16S rRNA genes from the feces of two healthy adult men ( subjects A and B ) were PCR amplified and sequenced . The sequence data were downloaded from the European Nucleotide Archive ( ENA ) , and the taxonomic compositions of the samples were analyzed by VITCOMIC2 . LEA was used to predict the topic compositions from the taxonomic composition data and input the topic compositions into the neural network to obtain coordinates of the samples on the global map . For the topic composition prediction , samples with <1 , 000 sequences assigned to genera were discarded , and samples with >10 , 000 sequences were sub-sampled to 10 , 000 sequences . For comparison , the dataset of Japanese gut microbiome samples by Nishijima et al . ( PRJDB3601 ) [63] was also downloaded , their taxonomic compositions were estimated by VITCOMIC2 , their topic compositions were estimated by LEA , and the samples were mapped on the global map . The number of gut samples was 314 for subject A , 184 for subject B , and 265 for the Japanese population . To assess the impact of different DNA extraction methods of the human gut microbiome analysis on the locations on LEA global map , we used the Microbiome Quality Control ( MBQC ) dataset reported by Sinha et al . ( PRJNA260846 ) [42] . This dataset contains 16S amplicon sequencing data from human stool samples , chemostats , and artificial microbial communities . For the same biological sample , there are multiple sequencing data analyzed with different wet laboratories or different DNA extraction methods . The sequence data were downloaded from the SRA , and the taxonomic compositions of the samples were analyzed by VITCOMIC2 . LEA was used to predict the topic compositions and obtain coordinates of the samples on the global map . For the topic composition prediction , samples with <1 , 000 sequences assigned to genera were discarded , and samples with >10 , 000 sequences were sub-sampled to 10 , 000 sequences . The number of samples was 1558 for human stool samples , 228 for chemostats , 133 for fecal artificial communities , and 130 for oral artificial communities . For the microbiome data from the highly diverse natural environments , we used the subset of Earth Microbiome Project ( EMP ) dataset[47] . The original dataset of EMP contains more than 27 , 000 samples of 97 studies , but they also provide the subset of the sample list which gives as equal as possible representation across EMP Ontology ( EMPO ) -level 3 sample types and across studies within those sample types . We downloaded the table of sample identifiers for the EMP subset of 2 , 000 samples ( emp_qiime_mapping_subset_2k . tsv ) from the FTP site of the EMP project ( ftp://ftp . microbio . me/emp/release1 ) . Raw sequence data of each sample were downloaded from the ENA , and the taxonomic compositions of the samples were analyzed by VITCOMIC2 . LEA was used to predict the topic compositions and obtain coordinates of the samples on the global map . For the topic composition prediction , samples with <1 , 000 sequences assigned to genera were discarded , and samples with >10 , 000 sequences were sub-sampled to 10 , 000 sequences . The number of samples used in the LEA mapping was 1760 . For the spatial microbiome distribution , we sampled the Tamagawa river in Japan . Its source is at the peak of Mt . Kasatori , and it flows through the Yamanashi , Tokyo , and Kanagawa prefectures . The river’s basin area is 1 , 240 km2 , its total length is 138 km , and the altitude of its source is 1 , 953 m . We sampled the river’s water on May 26 and 27 , 2015 at 38 points from its upper region to its estuary ( S5 Fig , S1 Table ) . Sampling was carried out during sunny weather and no precipitation had occurred for the 4 days before sampling . Samples were collected from the river’s surface water ( defined as flowing water no deeper than ~30 cm from the river surface ) using a ladle at the riverbank without disturbing the sediment . At least 500 ml of water was sampled at each site and then immediately injected into a sterile polypropylene bottle . Samples were transported in an insulated box containing a refrigerant and were moved to a refrigerator at 4°C within the day . The water of each sample was filtered through a 0 . 2-μm membrane within 5 days of sampling . Each filter was quickly placed in a sterile tube and stored frozen at –20°C . DNA was extracted from the material on each filter using the PowerWater DNA Isolation Kit ( MO BIO Laboratories , Carlsbad , CA ) with bead homogenization on a Micro Smash MS-100R ( TOMY , Tokyo ) at 3 , 000 rpm for 30 s . PCR amplification was performed using the primers 342F and 806R , which are V3–V4 region universal primers for prokaryotic 16S rRNA genes[64] , and Ex Taq DNA polymerase ( Takara , Shiga , Japan ) with a denaturation step at 98°C for 2 min , followed by 30 cycles at 98°C for 10 s , 50°C for 30 s , and 72°C for 40 s , and a final extension step at 72°C for 10 min . PCR products were purified with the Agencourt AMPure XP system ( Beckman Coulter ) and sequenced using an Illumina MiSeq sequencer ( Illumina , San Diego , CA , USA ) . The taxonomic composition of each sample was analyzed by VITCOMIC2 , and topic compositions for the samples were predicted by LEA . The Tamagawa river microbiome data has been deposited in DDBJ BioProject database; accession number: PRJDB5936 . | In the past decade , microbiomes from various natural and human symbiotic environments have been thoroughly studied . However , our knowledge is limited as to what types of environments affect the structure of a microbial community . In the first place , how can we define “environments” , in particular , the environmental entities that are often continuously varying and difficult to discretely categorize ? We assumed that environments could be represented from microbiome data because the structure of microbial communities reflect the state of the environment . We applied a probabilistic topic model to a dataset containing taxonomic composition data and natural language sample descriptions of >30 , 000 microbiome samples and extracted “latent environments” of the microbial communities , which are core elements of environmental mixtures . Integrating the training results of the model , we developed a web application to explore the microbiome universe and to place new metagenomic data on this universe like a global positioning system . Our tool shows what kinds of the environment naturally exist and are similar to each other on the perspective of the structural patterns of microbiome , and provides the way to evaluate the commonality and the heterogeneity of users’ microbiome samples . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"taxonomy",
"ecology",
"and",
"environmental",
"sciences",
"surface",
"water",
"microbiome",
"rivers",
"community",
"structure",
"microbiology",
"data",
"management",
"metagenomics",
"aquatic",
"environments",
"bodies",
"of",
"water",
"microbial",
"genomics",
"research",
"and",
"analysis",
"methods",
"hydrology",
"sequence",
"analysis",
"computer",
"and",
"information",
"sciences",
"bioinformatics",
"medical",
"microbiology",
"marine",
"and",
"aquatic",
"sciences",
"controlled",
"vocabularies",
"biological",
"databases",
"microbial",
"taxonomy",
"community",
"ecology",
"freshwater",
"environments",
"sequence",
"databases",
"ecology",
"earth",
"sciences",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics"
] | 2018 | Latent environment allocation of microbial community data |
The giant non-fimbrial adhesin SiiE of Salmonella enterica mediates the first contact to the apical site of epithelial cells and enables subsequent invasion . SiiE is a 595 kDa protein composed of 53 repetitive bacterial immunoglobulin ( BIg ) domains and the only known substrate of the SPI4-encoded type 1 secretion system ( T1SS ) . The crystal structure of BIg50-52 of SiiE revealed two distinct Ca2+-binding sites per BIg domain formed by conserved aspartate or glutamate residues . In a mutational analysis Ca2+-binding sites were disrupted by aspartate to serine exchange at various positions in the BIg domains of SiiE . Amounts of secreted SiiE diminish with a decreasing number of intact Ca2+-binding sites . BIg domains of SiiE contain distinct Ca2+-binding sites , with type I sites being similar to other T1SS-secreted proteins and type II sites newly identified in SiiE . We functionally and structurally dissected the roles of type I and type II Ca2+-binding sites in SiiE , as well as the importance of Ca2+-binding sites in various positions of SiiE . Type I Ca2+-binding sites were critical for efficient secretion of SiiE and a decreasing number of type I sites correlated with reduced secretion . Type II sites were less important for secretion , stability and surface expression of SiiE , however integrity of type II sites in the C-terminal portion was required for the function of SiiE in mediating adhesion and invasion .
Salmonella enterica is a food-borne Gram-negative pathogen which causes self-limiting gastroenteritis . To survive inside the host , Salmonella possesses sophisticated virulence factors and protein secretion systems [1] . A Salmonella pathogenicity island ( SPI ) 1-encoded type 3 secretion system ( T3SS ) is necessary for invasion [2] . This protein secretion system is capable to secrete a distinct cocktail of effector proteins , which manipulate the host cell . In order to establish the initial contact to the apical side of polarized epithelial cells and to enable translocation by the SPI1-T3SS , Salmonella deploys the SPI4-encoded T1SS and the giant non-fimbrial adhesin SiiE [3] . The SPI4 locus encodes SiiE and the T1SS for secretion of SiiE , with SiiF being the inner membrane transport ATPase , SiiD acting as periplasmic adaptor protein ( PAP ) , and outer membrane secretin SiiC [4] . SiiE , a 595 kDa non-fimbrial adhesin , is the only known substrate for the SPI4-T1SS [4] . SiiE mediates the first intimate contact to the host cell through binding to glycostructures containing N-acetyl-glucosamine and/or α2 , 3-linked sialic acid [5] . This contact positions the SPI1-T3SS to efficiently translocate effector proteins which lead to actin remodeling and macropinocytosis of the bacteria . As a T1SS substrate protein , SiiE possesses a C-terminal secretion signal [4] . The adhesin is transiently retained within the secretion system and at later time points present in the supernatant [6] . The two accessory proteins SiiA and SiiB are located in the inner membrane presumably forming a proton-conductive channel . This channel may use the proton motive force ( PMF ) at the cytoplasmic membrane to regulate the retention of SiiE , either through sensing the physiological state of the cell or by inducing conformational changes to binding partners [7] . The adhesin SiiE is composed of an N-terminal domain containing β-sheet and coiled-coil repeats , followed by 53 repeats of bacterial immunoglobulin ( BIg ) domains [6] . BIg52 and BIg53 are separated by a putatively unfolded element termed insertion . Sequence alignments of all 53 BIg domains revealed that a prototypical BIg domain possesses 6 conserved aspartate ( D ) or glutamate ( E ) residues , of which 5 form two binding sites for Ca2+ ions . Recently we solved the crystal structure of SiiE BIg50-52 . Despite not having explicitly added any Ca2+ ions during protein production , the crystal structure revealed that up to two Ca2+ ions are bound per BIg domain in SiiE [8] . The SiiE-wide conservation of D and E residues that are involved in Ca2+ binding suggests that SiiE binds about 100 Ca2+ ions per molecule [8] . Ultrastructural analysis showed that chelation of Ca2+ ions of purified secreted SiiE molecules distorts the linear rod-like structure of SiiE , indicating a stabilizing effect of Ca2+ ions . Ca2+ binding has been demonstrated for other T1SS substrate proteins , such as adhesins , the antifreeze protein of Marinomonas primoryensis ( MpAFP ) [9] or SpaA of Corynebacterium diphtheria [10] . Repeat in Toxin ( RTX ) proteins are a family of T1SS-secreted toxins and E . coli HlyA und Bordetella pertussis adenylate cyclase CyaA are well studied RTX toxins . The RTX motif is a glycine-rich nonapeptide involved in Ca2+ binding . For both HlyA and CyaA , binding of Ca2+ ions was shown to support the secretion by the T1SS [11–14] . The BIg domains of SiiE possess two distinct types of Ca2+-binding sites that are distinct from the RTX motif . The conserved D residues are numbered according to the multi-sequence alignment of SiiE BIg domains shown by Griessl et al . [8] . Type I Ca2+-binding sites are positioned at the interface of two BIg domains and contain three D residues , namely BIgn-1117D and BIgn43D and 97D . Type I Ca2+-binding sites are frequently found in BIg domain proteins . In contrast , type II Ca2+-binding sites are specific to SiiE and built by two D residues within one BIg domain ( BIgn16D and 24D ) . Please note that each Ca2+ ion is coordinated by 6 ligands and therefore other residues and water molecules are involved as well in Ca2+ ion binding , in addition to the conserved D residues [8] . We also observed conserved tryptophan residues , i . e . 74W , in most of the BIg domains . This residue is distal to the Ca2+-binding sites , but may be involved in interaction of SiiE with glycostructures as observed for transport proteins [15] or fimbrial adhesins [16] . The roles of the distinct type I and type II Ca2+-binding sites for secretion of SiiE and function as adhesin are not known . Here , we report the functional dissection of SiiE Ca2+-binding sites . We found that with an increasing number of conserved D residues exchanged to the non-charged amino acid serine ( S ) , the amounts of secreted SiiE were dramatically decreased . Exchanges of single D residues or of either single type I or type II Ca2+-binding sites showed no effect , while exchange of multiple type I or type II Ca2+-binding sites showed a more dramatic effect when type I Ca2+-binding sites are missing . Our data demonstrate a critical role of type I sites to support transport of SiiE through the T1SS , while type II sites are important to structure secreted SiiE and to maintain a BIg domain conformation that enables interaction with cognate ligands on the host cell surface .
The giant adhesin SiiE possesses 53 BIg domains , most of which contain five conserved D or E residues that coordinate binding of two Ca2+ ions . We investigated the role of Ca2 +-binding sites in BIg domains for secretion of SiiE . The Gaussia luciferase ( GLuc ) [17] was used as reporter for quantification of amounts of secreted SiiE ( Fig 1 , S1 Fig ) . Compared to Firefly luciferase , GLuc is ATP independent , and more robust and progressive [18] . To quantify the secretion , the reporter GLuc BIg50-53 was constructed by fusing GLuc to the C-terminal moiety of SiiE , i . e . BIg domains 50–53 , the insertion and the C-terminal secretion signal ( Fig 1A ) . Another construct was generated in which all five conserved D residues forming the type I and type II Ca2+-binding sites in BIg51 and BIg52 were exchanged to S ( D/S exchange ) , termed GLuc BIg50-53Δ2 . The number of deleted Ca2+-binding sites is indicated by Δn . A further reporter fusion consisting of GLuc and BIg47-53 was generated and D/S exchanges of various extent were introduced ( S1C Fig ) . The synthesis and secretion of GLuc fusions with WT or mutant SiiE portions was compared using Salmonella WT and the siiF-deficient strain unable to form a functional T1SS . GLuc activities in the lysate and supernatant obtained after 6 h of subculture represent the cytosolic and surface-bound , or secreted portion of GLuc fusions , respectively ( S1 Fig , Fig 1B ) . GLuc activities in lysates were similar for GLuc BIg50-53 and GLuc BIg50-53Δ2 reporters , indicating similar rates of synthesis and stability ( Fig 1B ) . Secreted GLuc activity for the SiiE WT reporter was 46 . 7-fold lower in the ΔsiiF background , demonstrating SPI4-T1SS-dependent secretion of the reporter . In the SPI4-T1SS-proficient background , secreted GLuc activity for the GLuc BIg50-53Δ2 reporter was 122 . 4-fold lower than for the GLuc BIg50-53 reporter . We also analyzed secretion of a GLuc fusion protein containing BIg47-53 ( S1C and S1D Fig ) . Here , five D/S exchanges in GLuc BIg47-53Δ2 did not reduce secretion . Also , the exchanges resulting in GLuc BIg47-53Δ4 were without effect on the secretion of the reporter , while additional D/S exchanges for deletion of another two Ca2+-binding sites caused a five-fold reduced secretion of GLuc BIg47-53Δ6 . The complete removal of 10 Ca2+-binding sites in GLuc BIg47-53Δ10 resulted in 73 . 7-fold reduced secretion , similar to levels of the WT reporter fusion in ΔsiiF background . We conclude that Ca2+-binding sites in BIg SiiE are required for the secretion of SiiE . The removal of two Ca2+-binding sites in a secretion reporter with four BIg was sufficient to ablate T1SS-dependent secretion , while removal of at least 6 Ca2+-binding sites was required to affect secretion of a reporter fusion with 7 BIg . To analyze the role of Ca2+-binding sites in SiiE for SiiE-dependent virulence functions of Salmonella , we transferred mutant alleles with D/S exchanges of various extent into chromosomal siiE using λ Red recombineering ( Fig 2A ) . The resulting strains synthesized mutant forms of SiiE with D/S exchanges resulting in removal of 2 , 5 , 6 or 10 Ca2+-binding sites ( Fig 2B ) . No SiiE was detected for the ΔsiiE strain and compared to WT SiiE , variable amounts of mutant SiiE were observed . Compared to the WT , all mutant strains investigated contain lower amount of cell-associated SiiE . Since whole bacterial lysates were analyzed , one cannot distinguish between SiiE present in cytosol , or SiiE retained on the bacterial surface . Mutant forms of SiiE with reduced surface retention will lead to lower amounts of cell-associated SiiE , although levels of SiiE synthesis are comparable to WT SiiE . SiiE retention on the bacterial surface and secretion into culture supernatant was analyzed by dot blots of whole cells , and protein precipitated from culture supernatants , respectively ( Fig 2C and 2D ) . Our previous work demonstrated that SiiE is mainly retained on the bacterial surface at 3 . 5 h of subculture , and predominantly released into the supernatant at 6 h and later of subculture [6] . Of the various mutant strains analyzed , only the strain producing SiiE BIg52Δ2 showed SiiE retention after 3 . 5 h of subculture similar to WT . After 6 h subculture levels of SiiE retention of WT and all mutant strains were as low as the negative control . The number of mutated Ca2+-binding sites correlated with reduction of secreted SiiE at 3 . 5 and 6 h of subculture . With increasing numbers of D/S exchanges , lesser amounts of secreted SiiE were detected . Amounts of SiiE BIg47-52Δ10 were as low as the negative control , indicating a complete loss of secretion for this SiiE mutant ( Fig 2D ) . We compared secretion of WT and mutant SiiE quantified by dot blot analyses to GLuc activities of the GLuc-SiiE reporter ( Fig 2E , S1C Fig ) . GLuc reporters for SiiE BIg52Δ2 and SiiE BIg47Δ1 BIg51-52Δ4 resulted in GLuc activities similar to GLuc-SiiEWT . If introduced in chromosomal siiE , the mutations resulted in reduced amounts of secreted SiiE . Secretion of SiiE BIg47Δ2 BIg51-52Δ4 was highly reduced in both assays , while no secretion of SiiE BIg47-52Δ10 was detected in GLuc and dot blot assays . The data demonstrate that Ca2+-binding sites in BIg are important for the secretion of SiiE , and that amounts of secreted SiiE decreases with an increasing number of D/S exchanges in BIg domains . We next determined the effect of deletion of Ca2+-binding sites on SiiE-dependent virulence functions , i . e . adhesion to polarized epithelial cell followed by SPI1-T3SS-mediated invasion ( Fig 2F ) . Only SiiE BIg52Δ2 conferred invasion of MDCK cells at a level comparable to WT SiiE . All other mutant SiiE we investigated resulted in highly reduced invasion , comparable to the siiF-deficient strain that is unable to secrete SiiE . We conclude that Ca2+-binding sites in the C-terminal part of SiiE are essential for secretion and function of the adhesin . Removal of Ca2+-binding sites in more than one BIg in this moiety results in loss of function . The C-terminal moiety of SiiE contains the signal for T1SS secretion , is secreted first and is likely to be exposed most distal to the cell envelope . We next tested the functional relevance of Ca2+-binding sites in the middle or N-terminal portions of SiiE . Strains were generated with mutations in chromosomal siiE resulting in deletion of Ca2+-binding sites in BIg2 , BIg40 , or BIg1-5 ( Fig 3A ) . Since 117D of a previous BIg domain ( BIg ( n-1 ) ) forms a type I Ca2+-binding site with 43D and 97D of a subsequent BIg domain ( BIg ( n ) ) , 117D of BIg1 and BIg39 were exchanged instead of 117D of BIg2 and BIg40 , resulting in SiiE BIg2Δ2 and SiiE BIg40Δ2 , respectively . To control the precision of the Red recombineering method applied here and the absence of unwanted attenuating mutations , we used a siiE mutant strain and restored the WT sequence . This strain , termed WTrestored , showed SiiE-dependent phenotypes as the WT strain . Strains expressing mutant chromosomal siiE were tested for protein synthesis . No expression could be detected for the negative control ΔsiiE . All mutant strains synthesized SiiE of correct size ( Fig 3B ) . The deletion of Ca2+-binding sites only in BIg2 or BIg40 had no or only small effects on surface retention of SiiE ( Fig 3C ) , secretion ( Fig 3D ) , or SiiE-dependent invasion ( Fig 3E ) . Secretion of SiiE BIg1-5Δ10 was slightly reduced , but there was still more secretion after 6 h of subculture than after 3 . 5 h of subculture . Interestingly , the level of SiiE retention was also reduced to approximately 50% of WT and maintained at the same level at 6 h of subculture . Destruction of all Ca2+-binding sites in BIg1-5 ( BIg1-5Δ10 ) led to a 74 . 6-fold decreased invasion of polarized cells ( Fig 3E ) , while the same extent of deletions in BIg47-52 ( BIg47-52Δ10 ) resulted in 392 . 2-fold reduced invasion ( Fig 2F ) . Compared to SiiE BIg47-52Δ10 , reduction of retention and secretion is less pronounced for SiiE BIg1-5Δ10 . If extracellular Ca2+ ions facilitate secretion of SiiE , the secretion might come to a halt earlier for BIg47-52Δ10 than for SiiE BIg1-5Δ10 . In addition to the conserved D or E residues involved in Ca2+ binding , 47 of 53 BIg domains possess a conserved tryptophan residue at position 74 . To test a potential role of these conserved tryptophan residues in SiiE function , we performed W to F ( W/F ) exchanges in 1 , 2 , or 3 BIg in the C-terminal moiety of SiiE ( S2A Fig ) . These mutations only resulted in minor changes of the amounts of SiiE retained and secreted at 3 . 5 h or 6 h of subculture ( S2B , S2C and S2D Fig ) . Functionally , none of the mutant forms of SiiE with various degrees of W/F exchanges resulted in reduced invasion of polarized epithelial cells ( S2E Fig ) , indicating that conserved 74W residues in the C-terminal moiety of SiiE are neither important for secretion and retention of SiiE , nor for the SiiE-dependent adhesion and invasion . T1SS substrate proteins are secreted in an unfolded state [19] . For example , secretion of E . coli HlyA was highly reduced if the protein was modified in a way that allowed folding in the cytosol [20] . To further investigate parameters known to affect secretion of T1SS substrate proteins , we tested if folding rate influences secretion as for HlyA . We fused the C-terminal portion of SiiE harboring the secretion signal to MalE . Distinct point mutations in the MalE portion of the fusion protein led to different folding rates as previous established by Bakkes et al . [20] . Secretion of various MalE-SiiE fusion proteins was analyzed at 3 . 5 h and 6 h of subculture . Similar amounts of fusion proteins were detected in the culture supernatant at both time points ( S3 Fig ) . This indicates that the rate of intracellular folding did not affect SiiE secretion , supporting that binding of extracellular Ca2+ ions by the secreted portion of BIg domains is more important . In order to assess the role of Ca2+ ions for the conformational stability , molecular dynamics ( MD ) simulations of WT and mutant SiiE were performed . We focused on BIg domains 50–52 for the following reasons: ( i ) a high-resolution crystal structure is available for this portion of SiiE , ( ii ) the fragment is sufficiently small to allow for extensive MD simulations , and ( iii ) the role of Ca2+-binding sites in BIg50-52 constructs was experimentally investigated ( Fig 1 ) . The role of Ca2+ ions was assessed from inspection of the tilt and twist angles defining the relative orientation of BIg 51 and 52 ( Fig 4A–4D ) . The percentage of tilted or twisted structures detected over the simulation time is summarized in Fig 4E . A comparison of WT and mutant forms revealed that mutation of both type I and type II sites caused an enhanced tilting of the structure and thus a less extended domain arrangement compared to the WT structure . The stronger effect was observed for the type I site , which is consistent with the direct location in the domain interface . Notably , the concomitant mutation of both sites had an additive effect resulting in the highest portion of tilted structures among all systems investigated . Mutation of type I and type II site did not only affect tilting , but also resulted in an enhanced twisting of the domain pair ( Fig 4E ) . However , in contrast to tilting , type I and type II site had a similar effect on domain twisting and there was no additive effect upon mutation of both sites . These data support the role of Ca2+-binding sites in increasing the rigidity of SiiE . To assess the relative Ca2+ binding affinities of the type I and type II sites , steered molecular dynamics ( SMD ) simulations were performed . The setup is schematically depicted in Fig 5A and 5B . The Ca2+ ions were independently removed from both sites and 10 simulations were performed for each site . For the type II site , all 10 work plots display an overall similar shape ( Fig 5C ) . Up to a distance of ~ 15 Å , the work linearly increases reflecting the disruption of the interactions between the Ca2+ ion and its protein ligands . For larger distances there is only a marginal further increase of the work , indicating that dissociation is almost complete at a distance of ~ 15 Å . The work required in the 10 SMD runs of the type II site ranges from 246–329 kcal x mol-1 . For the type I sites , the qualitative appearance of the curves is similar ( Fig 5D ) ; however , less work is required for the removal of the ion from this site . The resulting work ranges from 134–212 kcal x mol-1 , which is significantly lower than for the type II site . SiiE variants with altered type I and type II Ca2+-binding sites in the C-terminal moiety of SiiE ( SiiECterm ) were characterized in detail in in vitro experiments . These variants encompass BIg domains 48–53 , the insertion and the C-terminal segment that includes the secretion signal ( Fig 6A ) . We have previously shown that recombinant protein production of a fragment that covers BIg domains 50 to 52 in E . coli and subsequent purification of the fragment without the addition of any Ca2+ ions leads to a protein sample in which all Ca2+-binding sites are fully occupied [8] . This observation was corroborated not only by the final electron density map of the solved crystal structure , but also by analyzing in detail the anomalous scattering signal and via X-ray fluorescence measurements [8] . Here , we now used inductively coupled plasma—atom emission spectroscopy ( ICP-AES ) to verify the presence and/or absence of Ca2+ ions in protein variants that were produced following a similar purification protocol as previously described for the BIg domains 50 to 52 protein fragment [8] . In case of WT SiiECterm the occurrence of 8 to 10 Ca2+-binding sites was expected . In the ICP-AES experiment the protein was analysed at a concentration of 6 . 5 mg x ml-1 ( 88 . 1 μM ) , which results in a theoretical maximum calcium content of 28–35 μg x ml-1 ( 705 . 2–881 . 5 μM ) . The experimentally determined calcium concentration was 23 μg x ml-1 which amounts to 81 . 8% to 65 . 44% of the expected theoretically maximum content ( Table 1 , S4 Fig ) . Variant SiiECterm with type I and type II Ca2+-binding sites mutated ( SiiECterm BIg48-52Δ8type I + II ) was measured at a concentration of 4 . 2 mg x ml-1 ( 57 . 63 μM ) . For this variant , a calcium signal below the 0 . 5 μg x ml-1 calibration standard and outside the calibration range was detected ( S4 Fig ) . Thus , no calcium binding is observed for variant SiiECterm BIg48-52Δ8type I + II ( Table 1 ) . These measurements show that in case of the WT protein the Ca2+-binding sites are almost fully occupied in the recombinantly produced protein sample whereas the substitution of defined aspartic residues against serines in the SiiECterm BIg48-52Δ8type I + II results in a protein that is devoid of any calcium binding . Conversely , variants with some of the Ca2+-binding sites disrupted should display reduced Ca2+ binding if one assumes the absence of any cooperativity between the binding sites . To experimentally address individual roles of type I and type II Ca2+-binding sites for the conformational stability of SiiE , we performed circular dichroism ( CD ) measurements . Highly similar FarUV spectra were recorded for all variants , namely the SiiECterm construct with WT sequence , mutations of type I , type II and of both type I and type II Ca2+-binding sites ( Fig 6B ) . Estimation of the secondary structure content using the BeStSel server suggests that the SiiECterm wild-type variant consist of around 50% β-sheets , 10% turns and 40% others ( e . g . random coil ) . These values are close to those derived from the available SiiE BIg-domain crystal structure ( BIg domains 50–52 , PDB entry: 2YN5 ) . This suggests that the so far structurally uncharacterized domains ( BIg domains 48–49 and BIg domain 53 ) display similar secondary structures as domains 50 to 52 . Most importantly , however , the highly similar spectra and concomitant results from the secondary structure analysis of the variants studied here demonstrate that the secondary structure composition of the proteins is not altered , thus excluding pronounced mis- or unfolding , when mutating the type I and/or type II Ca2+-binding sites ( S4 Table ) . The thermal scanning CD measurements revealed distinct effects of type I and type II site mutations on the conformational stability of SiiECterm ( Fig 6C ) . All protein variants exhibit a decrease in ellipticity upon heating , which suggests that instead of a thermally induced unfolding an increased secondary structure and/or β-sheet formation via aggregation occurs . However , the magnitude of this transition is lower for SiiECterm WT and SiiECterm BIg48-52Δ4type II than for SiiECterm BIg48-52Δ4type I and SiiECterm BIg48-52Δ8type I + II . Also , Tonset of these structural changes is higher for WT and Δ4type II variants at 72°C and 69°C , respectively , than for Δ4type I and Δ8type I + II variants at 50°C and 45°C , respectively . Thus , while the thermal scanning CD curve of SiiECterm BIg48-52Δ4type II resembles that of SiiECterm WT , the spectra of proteins with mutations of type I and both type I and II sites indicated that the proteins are more prone to aggregation . To further investigate the conformational stability , the SiiECterm variants were subjected to native PAGE ( Fig 6D and 6E ) . All variants covering the C-terminal moiety of SiiE migrated as a single band under mild conditions ( Fig 6D ) . To further investigate the aggregation behavior , individual samples of each variant were incubated at increasing temperatures and subsequently analyzed by native PAGE . SiiECterm WT was resistant to aggregation up to 80°C and SiiECterm BIg48-52Δ4type II behaved similar to WT protein , although pronounced aggregation was detected at a slightly lower temperature ( Fig 6E ) . A ladder-like pattern indicates that SiiECterm BIg48-52Δ8type I + II started to form oligomers and aggregated already at 50°C . Aggregation of SiiECterm BIg48-52Δ4type I resembled SiiECterm BIg48-52Δ8type I + II , although slightly delayed ( Fig 6D ) . The analysis of the aggregation behavior therefore reflects the results of the thermal scanning CD measurements . Faster migrating protein species are visible in the native PAGE of the Δ4type I and Δ8type I + II mutants . To control whether unwanted proteolysis might have caused the occurrence of these so-called lower bands , SDS-PAGE analysis was performed for SiiECterm samples after incubation at various temperatures ( S5 Fig ) . Neither WT SiiECterm nor any of the mutant forms indicate a temperature-dependent occurrence of proteolytic fragments . The increased migration behavior thus results from a partial collapse of the expected linear overall structure of SiiE into a more globular domain arrangement as the result of the removal of the type I Ca2+-binding sites that are located in the interface between BIg domains . Next , the conformation and compactness of the SiiECterm variants was probed by limited proteolysis using α-Chymotrypsin and Proteinase K ( S6 Fig ) . Multi-domain proteins with flexible domain surface loops and/or interdomain linkers are expected to be more prone to proteolytic cleavage than proteins with very rigid domain architecture . Resistance against proteolytic cleavage by α-Chymotrypsin was clearly reduced for SiiECterm BIg48-52Δ8type I + II in comparison to WT protein or protein with only type I or type II binding site mutations . This suggests that both types of Ca2+-binding sites help to stabilize the fold against proteolytic degradation . The effect is possibly enhanced by the close spatial proximity of the two binding sites [8] . Resistance against proteolytic cleavage by Proteinase K was also reduced most for SiiECterm BIg48-52Δ8type I + II similarly to the proteolysis using α-Chymotrypsin . However , the stability of the individual type I or type II Ca2+-binding site mutants was also decreased , although to a lesser extent than for the variant with both binding sites mutated . Finally , we set out to functionally dissect the roles of type I and type II Ca2+-binding sites in SiiE . We generated mutant alleles by site-directed mutagenesis for single aa exchanges in BIg51 and BIg52 ( S7 Fig ) , or exchanges of all residues of either the type I or the type II Ca2+-binding sites within BIg52 , within BIg47-52 , or BIg1-5 ( Fig 7 ) . Mutation of single aspartate residues or D/S exchanges in single Ca2+-binding sites in chromosomal siiE did not affect SiiE synthesis , surface retention , secretion , or SiiE-dependent invasion ( S7B , S7C , S7D and S7E Fig ) . In contrast , if either type I or type II Ca2+-binding sites are missing within the five C-terminal BIg domains 47–52 , invasion is reduced to the level of the negative control ( Fig 7E ) . For both mutants , retention ( Fig 7C ) and secretion ( Fig 7D ) was reduced . This reduction was more pronounced for SiiE BIg47-52Δ5type I than for SiiE BIg47-52Δ5type II . Retention was fully abolished for SiiE BIg47-52Δ5type I , as well as for BIg47-52Δ10 for all time points tested , while secretion was reduced to 50% of WT SiiE . For SiiE BIg47-52Δ5type II retention after 3 . 5 h of subculture was reduced to 40% of WT SiiE and after 6 h and later time points no surface retention was detected , similar to WT SiiE . Removal of type I or type II Ca2+-binding sites in BIg1-5 had only minor effects on retention and secretion of SiiE . In contrast to SiiE BIg1-5Δ5type II , SiiE BIg1-5Δ5type I was retained at late time points at a level similar to BIg1-5Δ10 ( 8 and 24 h ) . SiiE BIg1-5Δ5type II was also retained at late time points , but to a lesser extent than SiiE BIg1-5Δ5type I or SiiE BIg1-5Δ10 . The mutation of five type I sites in BIg1-5 resulted in highly ( 43 . 5-fold ) reduced invasion , while removal of five type II Ca2+-binding sites in the same moiety ( BIg1-5Δ5type II ) did not reduce invasion of polarized epithelial cells ( Fig 7E ) . These results suggest distinct roles of type II Ca2+-binding sites in N- and C-terminal portions of SiiE . To further analyze the role of type II Ca2+-binding sites for function of SiiE , we removed type II Ca2+-binding sites by D/S exchanges in BIg31-35 , BIg 36–40 , BIg 41–45 , or BIg 46–50 . The invasion of MDCK cells of strains expressing these mutant forms of siiE was compared to invasion by strains with WT SiiE , SiiE BIg1-5Δ5type II and SiiE BIg47-52Δ5type II ( Fig 8 ) . We observed that strains producing SiiE with type II sites removed in BIg31-35 , BIg 36–40 , BIg 41–45 , BIg 46–50 or BIg47-52 all exhibited reduced invasion compared to strains with SiiE WT or SiiE BIg1-5Δ5type II . The reduction of invasion was pronounced if BIg domains in the C-terminal region were affected , and smallest reduction of invasion was observed for the strain with SiiE BIg30-35Δ5type II . The fact that SiiE BIg1-5Δ5type II , but not SiiE BIg47-52Δ5type II still mediates binding to , and invasion of MDCK cells indicates that type II Ca2+-binding sites are more important for the correct local conformation of the protein , which might be necessary for proper binding . We conclude that in SiiE BIg1-5Δ5type II the C-terminal part is correctly folded and can mediate binding to the host cell , while this is not the case for SiiE BIg47-52Δ5type II .
Our comprehensive mutational and functional analyses revealed a role of conserved aspartate residues in BIg domains of SiiE in secretion and adhesin function of this giant adhesin . An increasing number of exchanges of conserved aspartate residues resulted in decreased secretion of SiiE . This observation was made for a C-terminal plasmid-encoded portion of SiiE , as well as for chromosomally encoded variants of SiiE with the same amino acid exchanges . For chromosomally encoded SiiE BIg52Δ2 , the secretion was reduced . This reduced amount of secreted SiiE was not associated with higher levels of SiiE retention , or reduced SiiE-dependent invasion . Also , 5 D/S exchanges in BIg2Δ2 or BIg40Δ2 in chromosomally encoded SiiE did not lead to reduced invasion . Based on these results we conclude that the number of functional SiiE molecules on the bacterial surface is still high enough to mediate apical adhesion and subsequent invasion . We found that deletion of two Ca2+-binding sites by 5 D/S exchanges in the N- or C-terminal parts of SiiE showed no or only mild phenotypic difference , indicating that the remaining Ca2+-binding sites in adjacent domains can compensate for a certain degree of loss of Ca2+-binding properties of SiiE . Upon deletion of 5 or more Ca2+-binding sites , we observed loss of SiiE retention , dramatically decreased amounts of secreted SiiE and attenuated invasion . Thus , the lack of 5 Ca2+-binding sites in the C-terminal portion of SiiE could not be compensated by the function of residual domains . We propose a model in which binding of extracellular Ca2+ ions promotes directionality in the secretion of SiiE ( Fig 9A ) . If many consecutive D residues are missing , Ca2+ binding is ablated and the secretion is reduced . Lack of a few Ca2+-binding sites is not critical since Ca2+ ions will bind to the next available Ca2+-binding site of BIg domains that are already outside of the T1SS . If too many Ca2+-binding sites are missing , the next available Ca2+-binding sites are still within the channel of the T1SS and not accessible for the extracellular Ca2+ ions ( Fig 9A ) . Dependent on the position of the missing Ca2+-binding sites within SiiE , secretion of SiiE is arrested at a certain stage . If Ca2+-binding sites were removed in BIg1-5 , SiiE was also surface-retained at 6 h of subculture and later , indicating that the secretion process stopped . Exchanges in the C-terminal moiety , namely BIg47-52Δ10 led to secretion stalling early in the process of secretion , so that surface expressed or secreted SiiE was highly reduced . If fusion proteins covering a short portion of SiiE are investigated ( Fig 1 ) , lack of already a few Ca2+-binding sites leads to a dramatic decrease in secreted SiiE . Possibly , not only the number of lacking Ca2+-binding sites is responsible for this , but also the overall number of available Ca2+-binding sites . Thomas et al . [13] recently described a Ca2+ driven folding that may facilitate secretion in E . coli pro-HlyA . HlyA contains RTX motifs that bind Ca2+ . If no Ca2+ is bound , or if Ca2+ is chelated by e . g . EDTA , the protein remains unfolded [13 , 21 , 22] . A similar mechanism could be considered for secretion of SiiE , with binding of extracellular Ca2+ ions initially facilitating secretion and later supporting the proper conformation and interaction with ligands . Such an interaction could , for example , occur with the carboxyl group of SiiE ligand α2 , 3-linked sialic acid . From other T1SS substrates it is known that they are secreted in an unfolded state . For SiiE , the intracellular folding rate is not the most critical parameter for the initialization of the secretion process , as seen for the E . coli HlyA system [13] ( S3 Fig ) . Since the SPI4-encoded T1SS possesses two unique accessory proteins SiiA and SiiB , these proteins could promote the start of SiiE release until the protein becomes accessible to extracellular Ca2+ ions . SiiE has to be transiently retained on the bacterial surface in order to function as an adhesin . The proton channel formed by SiiA and SiiB or possibly other interactions between subunits of the T1SS could act as the retention signal , which has to be stronger than the extracellular Ca2+ ions . More experiments are needed to fully understand the retention process of this exceptional adhesin . Why does SiiE possess two distinct types of Ca2+-binding sites ? Conserved D residues forming type I Ca2+-binding sites can be found in other bacterial adhesins or secreted enzymes , like BapA from Salmonella enterica , LapF from Pseudomonas putida , or the PKD domain from the plant cell wall-targeting endoglucanase of Clostridium thermocellum [8 , 23–25] . In contrast , type II Ca2+-binding sites are specific for SiiE . Since type I and type II Ca2+-binding sites are structurally different , also distinct functions have to be considered . We speculate that type II Ca2+-binding sites are required for proper fine-tuning of the conformation of SiiE , while type I Ca2+-binding sites promote an overall rigidification of the tertiary structure of SiiE and thereby secretion through the T1SS . To elucidate the impact of type I and type II Ca2+-binding sites , several binding sites within the C-terminal portion were exchanged . All recombinantly produced C-terminal variants showed similar CD spectra showing that the secondary structure content was not altered in these variants . However , compared to type II sites , removal of type I sites had more dramatic effects on secretion and retention , independent of the region in SiiE in which the mutations were introduced . This observation would be in line with a role of type I sites in supporting secretion ( Fig 9B ) . Interestingly , mutations BIg47-52Δ5type I , and BIg1-5Δ5type I led to loss of function of SiiE in supporting invasion of polarized cells . Loss of SiiE function was observed for mutation BIg47-52Δ5type II , while mutation BIg1-5Δ5type II only caused minute alteration in invasion of MDCK cells . Thus , type II Ca2+-binding sites in the N-terminal region of SiiE are dispensable for secretion and surface expression of SiiE . BIg domains of LapF and BapA are more similar to SiiE BIg50 , harboring only a type I Ca2+-binding site [8] . Until now , no distinct role for BIg50 could be identified . Since it is much shorter than the other SiiE BIg domains , one Ca2+-binding site could be sufficient to stabilize this domain , whereas longer BIg domains need two Ca2+-binding sites to be stabilized . We previously reported that Ca2+ ions stabilize a rigid , linear conformation of SiiE , suggesting a role of Ca2+-binding sites for the overall protein stability [8] . The MD simulations of the present study showed that mutations of type I as well as type II sites destabilize SiiE and cause more frequent deviations from the geometry of the crystal structure . In particular , changes of the tilt angle result in less extended conformations ( Fig 4B ) that are expected to exhibit a reduced SiiE functionality . It is also interesting to note that the majority of the tilt angles observed during simulation are in the range of 15° to 30° . These rather small tilt angles suggest that larger deviations from the extended geometry require kinking at multiple sites at the same time . This is in line with the experimental data that mutations of multiple type I sites are required in order to disturb SiiE function . The observation that tilting is predominantly enhanced by mutation of the type I site suggests that these sites might also be more critical for the structural rigidification of SiiE and is in line with the observation that disruption of the type I sites leads to an earlier onset of temperature-induced aggregation of SiiE . Although the work calculated by SMD simulations cannot readily be converted into binding affinities , it suggests that type II sites exhibit higher Ca2+ affinity compared to type I sites ( Fig 5 ) . This is in line with a sequential mechanism of Ca2+ binding during secretion , in which initially the type II sites get occupied ( probably during the folding of the individual domains ) , and subsequently Ca2+ binding of the type I sites stabilizes the extended domain arrangement of SiiE . We propose that local conformational distortions in the C-terminal BIg domains due to the BIg47-52Δ5type II mutations may result in loss of binding to host cell ligands , despite sufficient amounts of SiiE being surface expressed . For SiiE BIg1-5Δ5type II , the C-terminally located BIg domains are correctly folded and proficient in ligand binding . This model also implies that BIg domains located in the N-terminal and perhaps also in the central portion of SiiE are not directly involved in binding to ligands on apical membranes of host cells ( Fig 9C , 9D and 9E ) . Although Ca2+ ions promote SiiE secretion and folding , we do not assume that they are also directly involved into binding to the target structure . SiiE binds to GlcNAc- and sialic acid-containing structures [5] . Griessl et al . [8] also showed that the majority of SiiE BIg domains possess a conserved tryptophan residue , however , exchanges of this aromatic residue in BIg50-52 did not influence SiiE specific phenotypes ( S2 Fig ) . Future mutational and functional analyses may identify the residues in C-terminal BIg domains of SiiE that directly contribute to the interaction with ligands . The 'C-terminus first secretion' of T1SS substrate proteins has been formally demonstrated for HlyA [26] . The C-terminal RTX domain of B . pertussis CyaA has 5 repeat blocks containing GGxGxDxxx motifs that form β-rolls coordinating Ca2+ binding . A total of 40 Ca2+ ions bound per CyaA molecule have been estimated [27] . Recent analyses of T1SS secretion of CyaA and related RTX toxins demonstrated a vectorial push-ratchet mechanism [12 , 14] . In the bacterial cytosol , the RTX domain is intrinsically disordered . After exit of the T1SS duct , the β-rolls sequentially bind Ca2+ at the outside of the cell envelope , initiate folding of the C-terminal domain and thereby provide directionality and secretion support for the rest of the protein [28 , 29] . The proposed model is in line with earlier findings , namely that ATP-hydrolysis and membrane potential are only required for the initiation of secretion , while further secretion is driven by the folding of portions of the substrate protein that have left the T1SS [11 , 14] . Interestingly , a recent analysis revealed that HlyA secretion efficiency is independent from Ca2+ concentration ranging from 0 to 5 mM Ca2+ in the external medium [30] . This observation would argue against a role of Ca2+ in supporting T1SS secretion of HlyA , however , the local Ca2+ pool in the outer membrane of secreting bacteria also has to be considered . Although Ca2+ binding by type I and type II Ca2+-binding sites in SiiE is mediated by structurally distinct motifs , we propose a function similar to RTX repeats regarding disordered-to-folded transition and directionality of secretion . While Ca2+-binding sites in CyaA are restricted to the C-terminal RTX domain , Ca2+-binding sites are present in all BIg domains of SiiE . This would be in line with a dual function of Ca2+ binding in promoting secretion , as well as stabilizing the ligand-binding competent overall conformation of SiiE .
Salmonella enterica serovar Typhimurium ( S . Typhimurium ) NCTC 12023 was used as wild-type strain in this study and all mutant strains are isogenic to this strain . The characteristics of strains used in this study are listed in Table 2 . Bacterial strains were routinely grown in LB broth or on LB agar containing antibiotics if required for selection of specific markers . The Ca2+ concentration in LB media is not defined . Carbenicillin or kanamycin were used at 50 μg x ml-1 , and tetracycline or chloramphenicol were added to a final concentration of 20 or 10 μg x ml-1 respectively , if required for the selection of phenotypes or maintenance of plasmids . Madin-Darby Canine Kidney Epithelial ( MDCK ) cells are an immortalized cell line initially derived from renal tube of a cocker spaniel . MDCK Pf subclone used for the generation of polarized epithelial cell monolayers was kindly provided by Department of Nephroplogy , FAU Erlangen-Nürnberg . MDCK cells were used for the generation of polarized epithelial cell layer . Cell culture conditions were previously described by Wagner et al . [6] . Briefly , cells were cultures in MEM with Earle’s salts , 4 mM Glutamax , non-essential amino acids and 10% heat-inactivated fetal calf serum . The Ca2+ concentration in this medium is 1 . 8 mM . The synthetic DNA fragments ( GeneArt or IDT ) were subcloned in blunt end restriction sites of the pJET1 . 2 vector backbone . The synthetic DNA fragment SiiE-BIg49half-52D-S was first cloned into pWRG454 ( pWSK29::PsiiAGlucM43LM110L::siiE-BIg50-53 ) via HindIII , NheI digest and ligation . To obtain the whole SiiE-BIg49-52D-S , the second synthetic DNA fragment SiiE-BIg47-49half D-S was cloned into pWSK29::PsiiAGlucM43LM110L::siiE-BIg49half-52D-S via HindIII digestion and ligation . Orientation was checked by PstI , ClaI diagnostic digest and subsequent sequencing confirmed the construct . The Q5 site-directed mutagenesis kit ( NEB ) was used to create plasmid with exchanges in codons for single conserved aspartate residues or exchanges for type I or type II Ca2+-binding sites in SiiE . Primers included new recognition sites for restriction enzymes through silent mutations . After performing colony PCR , the PCR fragment was digested with an appropriate restriction enzyme to confirm the silent mutations . Clones have also been confirmed by sequencing . The construction of scar-less in-frame deletions in siiE using the I-SceI site was previously described by Blank et al . [32] and the protocol modified by Hoffmann et al . [33] was applied here . Briefly , the I-SceI aph resistance cassette was amplified from pWRG717 using primers with 20 bp homology and 40 bp 5’ overlap . The resistance cassette was inserted by Red-mediated recombination within the desired region . The plasmid pWRG730 was used instead of pKD46 and features a heat-inducible promoter for red genes . For preparing competent cells of WT [pWRG730] , cells were grown in LB Cm10 to OD600 of 0 . 4–0 . 5 in a baffled flask in a shaking water bath at 150 rpm at 30°C . Subsequently , cells were immediately transferred to another water bath pre-heated to 42°C and incubated for 12 . 5 min at 100 rpm . After incubation of heat-induced cells on ice for 15 min , competent cells were prepared as described before [34] . After DpnI-digestion and purification , the PCR product was electroporated into Salmonella and transformants were selected on LB Km25 agar plates . Transformants were checked by colony PCR and confirmed clones were streaked on LB Cm10 plates with or without 100 ng x ml-1 anhydrotetracycline ( AHT , Sigma-Aldrich ) . Clones with the highest inhibition on AHT containing plates at 30°C were selected for further procedures . PCR fragments were amplified from plasmids containing exchanges of conserved aspartate codons in siiE or synthetic DNA were used . These fragments contain 5’ overlaps which are homolog to the insertion site of the chromosomally integrated I-SceI aph cassette . The strain containing the I-SceI aph cassette and harboring pWRG730 was then transformed by electroporation with either purified PCR product or synthetic DNA fragments . Serial dilutions from 10−1 to 10−4 were plated on LB Cm10 plates containing 100 ng x ml-1 AHT and incubated overnight at 30°C . The next day , large colonies were re-streaked on LB Cm10 AHT100 plates . Clones were verified by testing sensitivity to kanamycin and by colony PCR . The Gaussia luciferase assay was performed as previously described by Wille et al . [17] . Bacterial strains were diluted 1:31 in LB from O/N cultures and grown at 37°C for 3 . 5 h . Aliquots of 1 ml of bacterial culture were collected , cells pelleted and resuspended in 1 ml of sterile LB . After an additional washing step with sterile LB , optical density was measured and adjusted to OD600 of 1 in 500 μl of 3% PFA in PBS . After fixation of bacterial cells for 15 min at RT , cells were pelleted ( 10 , 000 x g; 5 min ) and resuspended in 500 μl PBS . Five microliters of bacterial suspensions were spotted on a nitrocellulose membrane which has been pre-wetted with PBS and dried again before adding bacteria . After drying of the spots , membranes were blocked with 5% dry milk powder in TBS/T ( TBS; 0 . 1% Tween20 ) for at least 30 min . For detection of SiiE on the bacterial surface , antiserum against the C-terminal moiety of SiiE [35] was diluted 1:10 , 000 in blocking solution and applied to the membrane . LPS was detected using antiserum against Salmonella O-antigen ( Becton-Dickinson ) at the same dilution . After incubation O/N at 4°C , membranes were washed thrice with TBS/T and bound primary antibodies were detected with anti-rabbit IRDye 800CW ( LI-COR ) at a dilution of 1:20 , 000 in PBS/T ( PBS; 0 . 1% Tween20 ) . Subsequently , membranes were incubated for 1 h at RT in the dark and washed thrice with PBS/T . Membranes were rinsed in PBS and signals were quantified using the Odyssey Imaging System ( LI-COR Biotechnology ) . Bacterial strains were diluted 1:31 in LB from an O/N culture and grown at 37°C for 6 h . The OD600 was measured . Aliquots of 2 ml of bacterial culture were taken and pelleted . Supernatants were filter sterilized ( 0 . 2 μm Millex filter units , Millipore ) . 200 μl of 100% TCA was added to 1 . 8 ml of filtered sterilized supernatant and incubated O/N at 4°C . The precipitated proteins were pelleted by centrifugation at 4°C for 45 min . After two washing steps with 1 ml ice-cold acetone ( 14 , 000 x g , 30 min , 4°C ) the precipitate was air dried and afterwards resuspended in 25 μl PBS per OD600 of 1 . Dot blot analysis was carried out as described before without detection of LPS . For Western blot analysis , O/N cultures were diluted 1:31 in LB and grown for 3 . 5 h at 37°C . OD600 was measured , and 150 μl were pelleted in 2 min at 4°C at 16 . 000 x g . The pellet was resuspended in OD600 x 50 μl in 1 x SDS sample buffer and incubated at 100°C for 5 min . 15 μl of each sample were loaded onto 3–8% gradient gels ( NuPage ) and electrophoretically separated for 1 h at 150 V . Semi-dry blotting was performed using a 0 . 2 μm nitrocellulose membrane with 64 mA/blot for 4 h . After blocking with 5% milk/TBS/T , the membrane was incubated first with a primary antibody against SiiE and subsequently with a secondary HRP-conjugated antibody against rabbit IgG . The signals were determines using ECL reagent ( ThermoScientific ) and the ChemiDoc system ( BioRad ) . Invasion assay was performed as previous described by Wagner et al . [6] . Briefly , O/N cultures of Salmonella strains were diluted 1:31 in LB and grown for 3 . 5 h in test tubes with aeration in a roller drum . The cultures were diluted in MEM medium to obtain a multiplicity of infection ( MOI ) of 5 and this inoculum as added to the MDCK cells . After infection for 25 min cells were washed three times with PBS to remove non-internalized bacteria , and medium was replaced by medium containing 100 μg x ml-1 gentamicin to kill remaining extracellular bacteria . After incubation for 1 h , cells were washed again with PBS , lysed by addition of 0 . 5% sodium desoxychlate in PBS , and colony forming units were determined by plating serial dilutions of the lysates onto agar plates . All SiiECterm variants ( p4033 , p4034 , p4462 , p4463 , S1 Table ) were cloned into the pGEX-6P-1 vector ( GE Healthcare ) and expressed as GST fusion constructs in E . coli BL21 ( DE3 ) ( Novagen ) . The bacteria were chemically transformed with the expression plasmid and transformants selected on LB agar plates containing 100 μg x ml-1 ampicillin . The expression was done in terrific broth containing 100 μg x ml-1 ampicillin . A starter culture was inoculated with a single colony and grown over night at 37°C and 180 rpm . On the next day , the main expression cultures were inoculated to an OD600 of 0 . 08 and incubated at 37°C and 180 rpm . At OD600 0 . 6 to 0 . 7 the temperature was reduced to 20°C and at OD600 1 . 0 to 1 . 2 the protein expression was induced by adding 0 . 5 mM IPTG . After induction , the proteins were expressed for 20 h at 20°C and 180 rpm , the bacteria harvested by centrifugation and the pellets stored at -80°C until used for purification . Briefly , the GST-tagged proteins were captured by Glutathione Sepharose affinity column ( GE Healthcare ) using standard buffers given in the manual . The GST tag was cleaved by adding GST tagged HRV 3C protease at a mass ratio of 1 to 250 ( protease-to-fusion protein ) and the tag , undigested fusion proteins and the protease were extracted by a second Glutathione Sepharose purification . As a final purification step , the proteins were separated by size exclusion chromatography using a HiLoad 26/60 Superdex 200 pg column ( GE Healthcare ) and a buffer with 25 mM Tris-HCl , 150 mM NaCl , pH 8 . 0 . No calcium was added to the buffers during the chromatographic purification . The protein was concentrated to 8 mg x ml-1 , frozen in liquid nitrogen and stored at -80°C until use . The calcium content of the SiiECterm variants was analysed by inductively coupled plasma—atom emission spectroscopy . All proteins were purified as described above , but without the second Glutathione Sepharose step after tag cleavage . The purified SiiECterm proteins were concentrated and dialysed against the same batch of buffer ( 5 mM Tris-HCl , pH 8 . 0 , ratio of sample-to-buffer volume 1:100 ) . Calcium standard for ICP ( Sigma-Aldrich ) was diluted to 0 . 5 , 5 and 50 μg x ml-1 with the same buffer . The ICP-AES analyses were performed using a Ciros CCD ( Spectro Analytical Instruments GmbH ) . Three individual measurements of the same sample were conducted for each variant and the mean calculated . Limited proteolysis was performed in order to investigate the influence of the mutations on the conformation and compactness of the proteins [36] . The proteolysis experiments were conducted at 20°C and 550 rpm in a benchtop shaker . The assay was done in a buffer containing 25 mM Tris-HCl , 150 mM NaCl , pH 8 . 0 and the SiiE protein concentration was adjusted to 1 mg x ml-1 . 10 μg α-Chymotrypsin or 0 . 5 μg Proteinase K ( Proti-Ace & Proti-Ace 2 Kit , Hampton Research , Aliso Viejo , USA ) were added per mg of SiiE protein . Aliquots of 18 μl were taken prior ( 0 min ) , and at various time points after protease addition ( 1 min , 5 min , 15 min , 30 min , 1 h , 2 h , 4 h , 7 h , O/N ) . The samples were mixed with 6 μl 4 x SDS-PAGE loading buffer and boiled at 95°C for 5 min to stop the cleavage reaction . The heat-treated samples were briefly spun down and stored at -20°C . The 10 μl of each sample were analyzed by SDS-PAGE using 15% polyacrylamide gels . Gels were stained with Coomassie Blue . The conformation and stability of SiiE variants were probed using circular dichroism ( CD ) spectroscopy . All measurements were done with a Jasco J-815 spectropolarimeter ( Jasco , Tokyo , Japan ) using a cuvette with a 0 . 1 cm path length . CD spectra were recorded at 20°C in the far UV region between 185 and 260 nm in 10 mM potassium phosphate buffer , pH 8 . 0 . Protein concentrations of 15 μM were used for SiiECterm variants . The band width was set to 1 . 0 nm , the scan speed to 20 nm x min−1 , data integration time to 1 sec , data pitch to 0 . 1 nm and sensitivity to standard . Each measurement was averaged across ten accumulations and the protein spectra corrected for the sample buffer . The stability of the proteins was compared by thermal scanning analysis . Changes in the secondary structure composition were investigated between 20°C and 96°C by monitoring the CD signal at wavelengths of 222 nm for SiiECterm proteins . A band width of 1 . 0 nm , data integration time of 8 sec , heat rate of 1°C x min-1 , sampling rate of one data point per 0 . 2°C and standard sensitivity was used for all thermal scanning experiments . Conversion of the data to concentration- and length-independent mean residue weight ( MRW ) ellipticities [θ]MRW was done as described previously [37] . The secondary structure analysis of the CD-spectra and of the Ig-domain structure of SiiE BIg domains 50 to 52 ( PDB entry: 2YN5 ) was done with single spectrum analysis and the secondary structure and beta-sheet decomposition for PDB-structures tools of the BeStSel server , respectively ( http://bestsel . elte . hu/ssfrompdb . php ) [38] . The wavelength range between 190 nm and 250 nm of the CD spectra was used for estimation of the secondary structure content . Native polyacrylamide gel electrophoresis ( PAGE ) was used to analyze the aggregation tendency of the SiiE variants . All protein samples were adjusted to 0 . 3125 mg x ml-1 and 20 μl samples of each protein were incubated at various temperatures ( 20 , 30 , 41 . 7 , 50 , 60 , 70 , 80 and 90°C ) for 5 min , briefly spun down and chilled on ice for 1 min . 5 μl of 5 x native PAGE-buffer ( 0 . 25% ( w/v ) Bromophenol blue , 4 . 5% ( w/v ) sucrose ) were added to each sample to achieve a final protein concentration of 0 . 25 mg x ml-1 and 15 μl ( 3 . 75 μg protein ) were loaded per sample . Native PAGE was done using 7 . 5% native polyacrylamide gels and native PAGE running buffer ( 50 mM Tris , 384 mM glycine ) . The gel runs were done at 5 mA per gel for 5 h and 6–8°C in the cold room and gels were stained with Coomassie Blue . For the SDS-PAGE analysis , 4 μl of 4 x SDS sample buffer were added to 16 μl of each sample , the mixture boiled at 95°C for 5 min and briefly centrifuged . 10 μl of each of the samples were analyzed on 15% acrylamide gels . The gel runs were done at 200 V for 60 min . The crystal structure of SiiE wild-type BIg domains 50–52 ( PDB code 2YN5 , chain A ) was used for all computational studies . Based on the wild-type system , three mutants were modelled that lack either the type I , the type II , or both types of Ca2+-binding sites . Mutants were generated by replacing the respective coordinating aspartate and glutamate residues by serine . All systems were neutralized by adding an appropriate amount of sodium counter ions . Each system was placed in a periodic TIP3P water box [39] extending at least 12 Å in all directions from the solute . All simulations were done with Amber 14 [40] using the ff99SB force field [41] . Long-range electrostatics were calculated with the particle mesh Ewald ( PME ) approximation [42] . Shake was used to constrain hydrogen atoms during equilibration and simulation [43] . Minimization , equilibration and MD calculations were carried out with the pmemd module of AMBER . Minimizations were run for 10 , 000 steps and switched from steepest descent to conjugate gradient after 500 cycles . During equilibration the system was gradually heated from 30 K to 310 K in 60 ps with backbone restraints of 2 . 0 kcal/mol Å2 and then relaxed at 310 K with backbone restraints of 0 . 2 kcal/mol Å2 for another 20 ps . Two independent production runs of 300 ns were generated for each system using the weak-coupling algorithm [44] and a Berendsen barostat in an NPT ensemble . For the SMD simulations ten restart files containing atomic coordinates and velocities were taken from wild-type MD simulation ( at intervals of 10 ns ) . Ca2+ ions were pulled from type I and type II sites of SiiE individually by a harmonic potential with the spring energy constant of 50 kcal/mol Å2 . The center of this potential was moved away from the center of mass ( CoM ) of the backbone atoms of one BIg domain with a constant velocity of 0 . 2 Å/ps . In this way the force was evenly distributed on the protein and no positional restraints had to be used . | The interaction of Salmonella enterica with polarized epithelial cells depends on the function of SiiE , a 595 kDa adhesin containing 53 repeats of a bacterial immunoglobulin ( BIg ) domain . SiiE is secreted and surface-expressed by a cognate type I secretion system ( T1SS ) . We found that BIg domains contain amino acid ( aa ) residues forming binding sites for Ca2+ ions . Two types of Ca2+-binding sites can be distinguished , termed type I and type II sites . We performed a structural and functional dissection of Ca2+-binding sites of SiiE . After mutation of aa residues forming type I and/or type II Ca2+-binding sites , we investigated the secretion , surface expression and function as adhesin for interaction with polarized epithelial cells of the SiiE variants . We found that Ca2+-binding sites are critical for supporting the secretion of SiiE . Integrity of type I sites in any position of SiiE is essential for efficient secretion and surface expression . In contrast integrity of type II sites is less important for secretion . However , loss of type II in the C-terminal , most distal portion of SiiE ablated SiiE-mediated adhesion , while loss of the type II sites in middle or N-terminal portions of SiiE had less or no effect on SiiE function . We propose a novel mechanism of Ca2+-dependent secretion and conformational fine tuning of SiiE as a large T1SS substrate with a central role in the interaction of S . enterica with host cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"bacteriology",
"protein",
"transport",
"medicine",
"and",
"health",
"sciences",
"crystal",
"structure",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"pathogens",
"cell",
"processes",
"condensed",
"matter",
"physics",
"microbiology",
"organic",
"compounds",
"physiological",
"processes",
"bacterial",
"diseases",
"mutation",
"secretion",
"systems",
"enterobacteriaceae",
"amino",
"acids",
"crystallography",
"bacteria",
"bacterial",
"pathogens",
"adhesins",
"aromatic",
"amino",
"acids",
"infectious",
"diseases",
"mutant",
"strains",
"solid",
"state",
"physics",
"microbial",
"physiology",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"chemistry",
"salmonella",
"physics",
"biochemistry",
"bacterial",
"physiology",
"protein",
"secretion",
"cell",
"biology",
"organic",
"chemistry",
"virulence",
"factors",
"physiology",
"secretion",
"tryptophan",
"biology",
"and",
"life",
"sciences",
"genetics",
"physical",
"sciences",
"organisms"
] | 2017 | Structural and functional dissection reveals distinct roles of Ca2+-binding sites in the giant adhesin SiiE of Salmonella enterica |
Environmental stressors can severely perturb cellular homeostasis and compromise viability . To cope with environmental stressors , eukaryotes have developed distinct signaling programs that allow for adaptation during different stress conditions . These programs often require a host of post-translational modifications that alter proteins to elicit appropriate cellular responses . One crucial protein modifier during stress is the small ubiquitin-like modifier SUMO . In many cases , however , the functions of stress dependent protein SUMOylation remain unclear . Previously , we showed that the conserved Saccharomyces cerevisiae Cyc8-Tup1 transcriptional corepressor complex undergoes transient hyperosmotic stress-induced SUMOylation and inclusion formation , which are important for appropriate regulation of hyperosmotic-stress genes . Here , we show the osmostress-responsive MAP kinase Hog1 regulates Cyc8 SUMOylation and inclusion formation via its role in the transcriptional activation of glycerol biosynthesis genes . Mutations that ablate Cyc8 SUMOylation can partially rescue the osmosensitivity of hog1Δ cells , and this is facilitated by inappropriate derepression of glycerol-biosynthesis genes . Furthermore , cells specifically unable to synthesize the osmolyte glycerol cause transient Cyc8 SUMOylation and inclusions to persist , indicating a regulatory role for glycerol to reestablish the basal state of Cyc8 following adaptation to hyperosmotic stress . These observations unveil a novel intersection between phosphorylation and SUMOylation networks , which are critical for shifting gene expression and metabolic programs during stress adaptation .
Cellular stresses are abiotic perturbations that can severely and irreversibly damage biomolecules and essential cell structures . All organisms experience cellular stress and subsequently must adjust a variety of cellular programs to reestablish homeostasis . Many of these include signal transduction networks , metabolic pathways , gene expression programs , cell-cycle progression , and protein quality control systems . Resiliency in the face of stress is crucial to cellular survival with the deterioration of adaptive measures thought to underlie a variety of age-related human diseases such as cancer , heart disease , and neurodegeneration [1–3] . Post-translational protein modifications are critical for the responses to cellular stress , and often occur rapidly upon the onset of stress . One of the major modifications that have been observed to occur during a wide range of cellular stresses is the addition of the small-ubiquitin-like-modifier ( SUMO ) to proteins , with many of the targets of SUMOylation being specifically modified during distinct stress conditions [4 , 5] . While many studies have documented the targets of stress-dependent SUMOylation , consequences of many SUMOylation events still remain poorly characterized [6] . Furthermore , it is still unclear what cellular factors regulate the dynamics of stress-dependent SUMOylation during cellular adaptation [7] . In the budding yeast Saccharomyces cerevisiae , exposure to hyperosmotic stress initiates a rapid transient wave of SUMOylation [8–10] . Yeast adaptation to hyperosmotic conditions is largely facilitated by the high osmolarity glycerol ( HOG ) signaling pathway , which converges on the yeast orthologue of mammalian p38 mitogen-activated protein kinase ( MAPK ) , Hog1 [11 , 12] . Upon activation , Hog1 initiates a multifaceted adaptive program that results in alterations in gene expression , temporary cell-cycle arrest , and a metabolic shift toward the synthesis and retention of the intracellular osmolyte glycerol [11] . Furthermore , loss of Hog1 function extends the duration of hyperosmotic stress-induced SUMOylation [8] . However , the targets of this extended SUMOylation and its functional purpose have yet to be understood . We previously discovered that the primary targets of hyperosmotic stress-induced SUMOylation are the transcriptional corepressor proteins Tup1 and Cyc8 [10] . These proteins are rapidly and transiently SUMOylated following exposure to hyperosmotic conditions , and mutations that limit their SUMOylation drastically alter transcriptional patterns during adaptation to stress . Interestingly , the Cyc8-Tup1 complex forms reversible nuclear inclusions upon hyperosmotic shock , and the persistence of these inclusions correlates tightly with the duration of SUMOylation and the loss of transcriptional repression . Self-association of the Cyc8-Tup1 complex into inclusions is driven in large part by a disordered prion domain in Cyc8 [10] , which is highly glutamine-rich [13] . Moreover , Cyc8 can be induced to form and propagate as a prion , and cells bearing the prion form of Cyc8 show derepression of target genes consistent with a partial loss-of-function phenotype [14] . Recent studies have shown that a variety of disordered or prionogenic proteins form dynamic inclusion bodies under stress conditions , but it remains unclear what facilitates a shift from soluble , liquid-like states to insoluble , solid-like states [15 , 16] . Here , we explore how Cyc8 inclusions and SUMOylation are modulated by the Hog1-dependent accumulation of the intracellular osmolyte glycerol during hyperosmotic stress .
Hog1 has been shown to limit the accumulation of high molecular weight SUMO conjugates after the induction of hyperosmotic stress [8 , 10] , but the proteins that comprise the Hog1-regulated SUMO conjugates had not been identified . We previously demonstrated that the Cyc8-Tup1 corepressor complex comprised the major targets of SUMOylation following hyperosmotic stress exposure [10] . Given this , we hypothesized that Hog1 regulates the duration of Cyc8-Tup1 SUMOylation following exposure to hyperosmotic stress . To test this , we performed metal affinity purification of SUMOylated proteins on yeast cells expressing modified version of the yeast SUMO protein , His6-FLAG-Smt3 , from the endogenous SMT3 promoter after exposure to 1 . 2M sorbitol . We followed the purification by examining the SUMOylation state for Cyc8 or Tup1 via Western analysis . While parent cells showed transient SUMOylation of both Cyc8 and Tup1 , hog1Δ cells showed prolonged SUMOylation of both complex members , with the SUMOylation persisting to at least 60 minutes post stress induction ( Fig 1A and 1B , WT vs hog1Δ ) . Hog1 function depends on both phosphorylation-dependent activation using an established TXY motif and catalytic function through a defined kinase domain [17] . Because Cyc8 SUMOylation is the key initiating event for SUMOylation of the Cyc8-Tup1 complex [10] , we wanted to verify that these individual Hog1 functions were necessary for the regulation of Cyc8 SUMOylation . To do so , we generated plasmids encoding intact ( WT ) , activation-deficient ( TGY-AGA ) , or kinase-deficient ( D144A ) Hog1 and expressed them ectopically in hog1Δ cells . The hyperosmotic sensitivity of the mutants was measured by a spot titer assay on synthetic medium containing an elevated concentration of KCl ( S1 Fig ) . Each hog1 mutant strain showed the same sensitivity as the hog1Δ strain . To assess the effect of the hog1 mutations on Cyc8 SUMOylation , SUMOylated proteins were purified from these strains by metal affinity chromatography before and after exposure to 1 . 2M sorbitol , and Cyc8 SUMOylation was examined by Western analysis . While ectopic expression of WT Hog1 rescued the delay in Cyc8 deSUMOylation , both activation-dead and kinase-dead Hog1 showed prolonged Cyc8 SUMOylation similar to that of the complete HOG1 gene deletion ( Fig 1C and 1D ) . Upon exposure to hyperosmotic stress , Hog1 feedback can inhibit the pheromone response MAPK cascade to prevent inappropriate activation of the mating pathway , which shares the upstream MAPKKK Ste11 [11] . As a result , hog1Δ cells can undergo inappropriate activation of the mating pathway upon exposure to hyperosmotic stress . To ensure that our observations on Cyc8 SUMOylation were not due to crosstalk , we generated ste11Δ cells and examined the distribution of high molecular weight SUMO conjugates by Western analysis . Compared to the prolonged SUMOylation observed in hog1Δ cells , ste11Δ cells showed the same transient SUMOylation profile as parent cells , indicating that crosstalk with the mating pathway was not responsible for the delay in Cyc8 deSUMOylation ( S1 Fig ) . Hog1 functions in the hyperosmotic stress response by altering transcription , translation , and post-translation programs in the cell [11] . Because Cyc8 is a transcription corepressor , we wanted to determine whether the accumulation and/or persistence of SUMOylated Cyc8 following hyperosmotic stress was regulated by active transcription . We chose pharmacological manipulation because that allows for transcription to be inhibited rapidly and acutely . To do this , we used the transcription inhibitors thiolutin ( THL ) and 1 , 10-phenanthroline ( PHN ) . These drugs block transcription through distinct mechanisms: THL directly inhibits RNA polymerase while PHN is a potent chelator of metal ions that function as cofactors for RNA polymerases [18 , 19] . We found that acute pretreatment of cells with both drugs prolonged Cyc8 SUMOylation following the onset of hyperosmotic stress . In both cases after pretreatment , Cyc8 SUMOylation was maintained at least to 60 minutes following hyperosmotic stress induction ( Fig 2A and 2B ) . While both drugs prolonged the duration of Cyc8 SUMOylation , the relative levels of hyperosmotic stress-induced Cyc8 SUMOylation were different: THL pretreatment showed initial Cyc8 SUMOylation equivalent to vehicle pretreatment , whereas PHN pretreatment showed initial Cyc8 SUMOylation much less than vehicle treatment . As stated previously , PHN is a metal chelator that forms stable complexes with a variety of divalent metals including Zn2+ . SUMO ligases use Zn2+ as an essential cofactor [20] . Thus , the lower levels of initial Cyc8 SUMOylation seen after PHN pretreatment prior to hyperosmotic stress induction likely reflects inhibition of SUMO ligase activity , which would explain the observed differences in relative SUMOylation following THL and PHN pretreatment . We previously reported that Cyc8 forms transient inclusions during hyperosmotic stress that correlated with Cyc8 SUMOylation kinetics and dynamics [10] . As such , we hypothesized that transcription inhibition that prolonged SUMOylation of Cyc8 would also prolong the persistence of Cyc8 inclusions . To test this , we pretreated parent cells expressing Cyc8-eGFP from the endogenous CYC8 locus with vehicle , THL , or PHN , challenged them with hyperosmotic stress , and imaged by fluorescence microscopy ( Fig 2C ) . Treatment with either THL or PHN significantly prolonged the lifetime of Cyc8 nuclear inclusions during hyperosmotic stress ( Fig 2D ) . Thus , both transient Cyc8 SUMOylation and inclusion kinetics/dynamics require active transcription , suggesting that a regulatory factor must be produced through gene expression to elicit appropriate Cyc8 SUMOylation and behavior during hyperosmotic stress . Cyc8 SUMOylation was prolonged when transcription was inhibited , so we explored the consequences of eliminating Cyc8 SUMOylation . We previously revealed that hyperosmotic stress-dependent Cyc8 SUMOylation primarily occurs at a cluster of four lysine residues ( position 735 , 736 , 738 , and 748 ) located C-terminal to the TPR region [10] . We created four Lys-to-Arg mutations that ablate SUMOylation of Cyc8 , which we call Cyc84KtoR . These mutations reside in the Cyc8 C-terminal region that , when deleted in its entirety , partially suppressed the osmosensitivity of hog1Δ cells , and was correlated with derepression of hyperosmotic stress-responsive genes [21] . Because this large deletion , from residues 389–966 , may contain other important regulatory regions , we asked if only the mutation of the Cyc8 SUMOylation sites could suppress the osmosensitivity of hog1Δ cells using a spot titer assay . While mutation of the Cyc8 SUMOylation sites did not alter growth of HOG1 cells , loss of Cyc8 SUMOylation partially suppressed the osmosensitivity of hog1Δ cells ( Fig 3A ) , similar to previous studies using the Cyc8 C-terminal truncation mutant . We verified the appropriate reduction of Cyc8 SUMOylation in Cyc84KtoR mutant cells by performing metal affinity purification of SUMOylated proteins and probing for Cyc8-3HSV by Western analysis ( Fig 3D ) . Spot titer assays do not reveal quantitative growth rates . Therefore , we analyzed growth rates under hyperosmotic stress using a Bioscreen C automated growth curve analyzer ( Fig 3B ) . When HOG1 is present , the SUMOylation-deficient Cyc84KtoR had little effect on cells' ability to grow under stress . In hog1Δ cells , the presence of SUMOylation-deficient Cyc84KtoR accelerated the cells’ growth rate when compared to SUMOylation-proficient Cyc8 . We calculated doubling times for all yeast strains using the Yeast Outgrowth Data Analyzer software [22] , and found that loss of Cyc8 SUMOylation significantly reduced the relative doubling time of hog1Δ cells when grown during hyperosmotic stress ( Fig 3C ) , similar to the spot titer assay . Altogether , loss of Cyc8 SUMOylation is sufficient to reestablish growth of hog1Δ cells in hyperosmotic conditions . Loss of Cyc8 SUMOylation in hog1Δ cells partially restored the viability of hog1Δ cells during hyperosmotic stress . Because Cyc8 is a transcription corepressor that engages chromatin-bound transcription factors in target gene promoters , we wanted to examine the effects of the hog1Δ allele and loss of Cyc8 SUMOylation on Cyc8 chromatin occupancy . We chose to interrogate Cyc8 promoter occupancy at the glycerol biosynthetic enzyme GPD1 gene for two reasons . First , while Hog1 activation by hyperosmotic stress upregulates the expression of >100 genes upon hyperosmotic stress [23] , only the increased expression of GPD1 is required for cellular survival during hyperosmotic stress via its role in the biogenesis of the intracellular osmolyte glycerol [24] . Second , GPD1 is a known target of Cyc8-mediated repression under non-stress conditions [25] . We hypothesized that altered Cyc8 occupancy at GPD1 when Cyc8 SUMOylation was ablated could lead to increased production of glycerol in hog1Δ cells . To test this hypothesis , we examined Cyc8 or SUMOylation-deficient Cyc84KtoR association at the GPD1 promoter using chromatin immunoprecipitation ( ChIP ) and quantitative PCR ( ChIP-qPCR ) in HOG1 and hog1Δ cells ( Fig 4A ) . In HOG1 cells , Cyc8 became enriched at the GPD1 promoter within 10 minutes of stress exposure and returned to baseline levels after 60 minutes . In hog1Δ cells , Cyc8 had nearly identical enrichment at the GPD1 promoter prior to stress onset and during the initial phase of adaptation ( 10 minutes ) . However , Cyc8 continued to be highly enriched at the GPD1 promoter 60 minutes after stress onset in hog1Δ cells . SUMOylation-deficient Cyc84KtoR showed different GPD1 promoter occupancy that Cyc8 . In HOG1 cells , Cyc84KtoR had slightly increased GPD1 promoter occupancy prior to the onset of stress . During the initial phase of stress exposure , Cyc84KtoR was enriched at the GPD1 promoter , but it was reduced compared with Cyc8 . After 60 minutes of stress exposure , Cyc84KtoR occupancy at the GPD1 promoter returned to baseline , similar to Cyc8 . In hog1Δ cells , Cyc84KtoR had greater GPD1 promoter occupancy prior to the onset of stress . There was no further enrichment at the GPD1 promoter for Cyc84KtoR in hog1Δ cells during the course of stress exposure . We next examined whether Cyc8 SUMOylation deficiency correlated with alterations in glycerol production that might explain the partial suppression of osmosensitivity of hog1Δ cells . To do this , we generated lysates from the various cells across a time course of hyperosmotic stress and analyzed the levels of glycerol present by a colorimetric assay ( Fig 4B ) . In cells with HOG1 intact and either Cyc8 or Cyc84KtoR , there was rapid , robust glycerol accumulation upon the onset of hyperosmotic stress . In hog1Δ cells , there was markedly reduced glycerol accumulation , but SUMOylation-deficient Cyc84KtoR partially rescued glycerol accumulation over the course of hyperosmotic stress . The accumulation of approximately 50% of glycerol in SUMOylation-deficient Cyc84KtoR hog1Δ cells during exposure to hyperosmotic stress is consistent with the partial suppression of osmosensitivity observed in Fig 3 . Although we observed elevated intracellular glycerol in Cyc84KtoRhog1Δ cells , we wanted to ensure that glycerol biosynthesis was necessary for the observed suppression of osmosensitivity . Therefore , we deleted GPD1 from Cyc84KtoR hog1Δ cells and compared growth of these cells to control cells under hyperosmotic stress conditions by spot titer assay ( Fig 4C ) . The additional deletion of GPD1 from Cyc84KtoR hog1Δ cells did not alter growth on normal media . However , Cyc84KtoRhog1Δgpd1Δ cells were highly osmosensitive and now equivalent to hog1Δ cells . Taken together , we conclude that altered GPD1 promoter occupancy of Cyc84KtoR can change the capacity to synthesize glycerol in the absence of Hog1 , and this leads to the partial suppression of osmosensitivity in hog1Δ cells by Cyc84KtoR . Loss of Cyc8 SUMOylation partially rescued glycerol biosynthesis in hog1Δ cells . Therefore , we investigated if glycerol biosynthesis regulated Cyc8 SUMOylation . Glycerol biosynthesis is a two-step process in S . cerevisiae . First , dihydroxyacetone phosphate ( DHAP ) is reduced to glycerol-3-phosphate ( G3P ) in the rate-limiting step by the redundant glyceraldehyde-phosphate dehydrogenases 1 and 2 ( Gpd1/Gpd2 ) [26] . Following , G3P is dephosphorylated by glycerol-phosphate phosphatases 1 and 2 ( Gpp1/Gpp2 ) to yield glycerol [26] . To examine whether glycerol biosynthesis regulated hyperosmotic stress-induced SUMOylation , we deleted components of the glycerol biosynthesis machinery and analyzed global SUMOylation kinetics by Western analysis . We found that deletion of individual components of either step of glycerol biosynthesis did not alter SUMOylation kinetics from that of parent cells ( Fig 5A ) . However , when the glycerol biosynthesis pathway was fully ablated by combinatorial deletion of GPD1 and GPD2 , hyperosmotic stress-induced SUMOylation was prolonged to at least 60 minutes after exposure . G3P is a necessary precursor to the production of lysophosphatidic acid ( LPA ) , a phospholipid derivative with myriad cellular functions [27] . To verify that our findings were due to loss of glycerol and not of LPA , we generated cells deleted for the downstream glycerol biogenesis enzymes Gpp1 and Gpp2 ( Fig 5B ) . As with the upstream Gpd proteins , only when both GPP1 and GPP2 were deleted did we prolong hyperosmotic stress-induced SUMOylation from that of parent cells . These data indicated that glycerol biosynthesis is a key regulator of hyperosmotic stress-induced SUMOylation . To verify that the observed delay in deSUMOylation kinetics was specific to Cyc8 , we purified SUMOylated proteins from parent , hog1Δ , and gpd1Δgpd2Δ cells by metal affinity purification and examined Cyc8 SUMOylation by Western analysis ( Fig 5C ) . As seen before , parent cells show a transient pattern of Cyc8 SUMOylation wherein Cyc8 is transiently SUMOylated upon hyperosmotic stress . As seen before in Fig 1A , deletion of HOG1 delays the deSUMOylation of Cyc8 , resulting in a significant amount of Cyc8 remaining SUMOylated at 60 minutes in hog1Δ cells ( Fig 5C ) . Cells that are gpd1Δgpd2Δ show robust SUMOylation of Cyc8 at 60 minutes , more so than in hog1Δ cells . It has been shown that hog1Δ cells accumulate glycerol under hyperosmotic stress , but do so at significantly slower rates than parent cells [28] . To illustrate the differences in glycerol accumulation between the different deletion cells , we performed colorimetric assays ( S2 Fig ) . As expected , parent cells rapidly and robustly accumulated glycerol over the course of an hour under hyperosmotic stress , whereas hog1Δ cells accumulated glycerol slowly with a maximal accumulation of approximately 66% that of parent cells . By contrast , gpd1Δgpd2Δ cells accumulated virtually no glycerol over the observed time course . To verify the glycerol accumulation was due to active transcription , we treated parent cells with THL or PHN prior to exposure to hyperosmotic stress . These treatments significantly reduced glycerol accumulation ( S2 Fig ) , further strengthening the hypothesis that stress-induced Cyc8 SUMOylation persists until glycerol has accumulated to sufficient levels and that active transcription of glycerol biosynthesis genes is necessary for the accumulation . To establish that Cyc8 deSUMOylation is linked to glycerol biosynthesis and not to the action of glycerol activating another kinase signaling pathway , we chose to broadly inhibit intracellular kinases prior to hyperosmotic stress . We found that widespread kinase inhibition by pretreatment with the broad-spectrum kinase inhibitor staurosporine did not change Cyc8 SUMOylation dynamics during adaptation to hyperosmotic stress ( S3 Fig ) . Importantly , we performed this experiment at a concentration of staurosporine found to inhibit PKA , but not Hog1 ( S3 Fig ) . Thus , kinase pathways outside the Hog1 MAPK do not appear to be involved in regulating the dynamics of Cyc8 SUMOylation . Due to our observation of prolonged Cyc8 SUMOylation in the absence of glycerol biosynthesis , we wanted to investigate whether glycerol accumulation regulated the resolution of Cyc8 inclusions . Using parent , hog1Δ , and gpd1Δgpd2Δ cells expressing Cyc8-eGFP , we performed fluorescence microscopy experiments across a time course of hyperosmotic stress ( Fig 5D ) . As we observed previously , Cyc8 forms nuclear inclusions in parent cells that are resolved within 15 minutes . By contrast , hog1Δ cells show elevated prevalence of inclusions and slightly delayed resolution , with inclusions persisting to at least 30 minutes . In cells that are gpd1Δgpd2Δ , we observed robust inclusion formation with no observable resolution across the entire 90 minutes of observation . By quantifying the relative number of cells with Cyc8 inclusion for each background across the time course of stress , there was a significant increase in Cyc8 inclusion number and lifespan in the absence of glycerol accumulation ( Fig 5E ) . Both acute THL and PHN treatment prolonged Cyc8 inclusions ( Fig 2C and 2D ) , and prevented glycerol accumulation ( S2B Fig ) . However , it was unclear whether the action of these transcription inhibitors on glycerol biosynthesis was a causative reason for the extended persistence of Cyc8 inclusions . To decipher this , we monitored Cyc8 inclusions by fluorescence microscopy on parent or gpd1Δgpd2Δ cells expressing Cyc8-GFP after washing out the reversible transcription inhibitor PHN ( Fig 6A ) . Yeast cells were treated with PHN for 5 minutes to inhibit transcription . As previously seen , PHN treatment was insufficient to induce significant amounts of Cyc8 inclusion formation alone ( Fig 6B ) . After PHN treatment , cells were subsequently challenged with 1 . 2M sorbitol for 30 minutes to form Cyc8 inclusions , then the cells were moved to media containing 1 . 2M sorbitol with or without PHN . Cyc8 inclusions disappeared over the course of 60 minutes in parent cells after removal of PHN , but were maintained up to 120 minutes in gpd1Δgpd2Δ cells after removal of PHN . Cyc8 inclusions did not resolve in either parent or gpd1Δgpd2Δ cells when PHN was maintained in the media ( no washout ) . Quantification of the total number of cells displaying Cyc8 inclusions following drug removal showed a significant loss of Cyc8 inclusions only in parent cells by 30 minutes , reaching basal levels by 120 minutes after removal of PHN ( Fig 6C ) . Cells incapable of producing glycerol , however , showed no resolution of Cyc8 inclusions . Under the same experimental design , we monitored intracellular glycerol content of parent or gpd1Δgpd2Δ cells after removal of PHN ( Fig 6D ) . As expected , parent cells showed a significant accumulation of intracellular glycerol following removal of PHN , which was inversely correlated to the amount of Cyc8 inclusions seen previously . In line with this , parent cells had greatly reduced glycerol accumulation when PHN was maintained in the media . Cells that were gpd1Δgpd2Δ showed virtually no intracellular glycerol , regardless of the presence of PHN . Finally , we tracked Cyc8 SUMOylation in parent or gpd1Δgpd2Δ cells after transcriptional restart ( Fig 6E ) . As we showed in Fig 2 , transcription inhibition was insufficient to induce Cyc8 SUMOylation in either strain prior to the onset of stress . Application of the hyperosmotic stressor initiated Cyc8 SUMOylation in both parent and gpd1Δgpd2Δ cells as observed after 30 minutes of exposure . As expected , maintenance of transcriptional blockade resulted in sustained Cyc8 SUMOylation , regardless of the presence of glycerol biogenesis enzymes ( Fig 6E , no wash ) . In parent cells , removal of transcriptional block ( Fig 6E , wash ) resulted in deSUMOylation of Cyc8 , tightly correlating with the rate of glycerol accumulation . In gpd1Δgpd2Δ cells , however , Cyc8 SUMOylation was maintained even upon removal of PHN . Altogether , these results support the idea that accumulation of the osmolyte glycerol is the necessary signal for the resolution of Cyc8 inclusions and Cyc8 deSUMOylation during exposure to hyperosmotic stress .
Cellular stress responses involve integration of multiple signaling pathways to adapt the cell’s metabolism to the stress . Here , we examined the integration of signaling pathways during hyperosmotic stress and reveal a novel intersection between Hog1 MAPK-dependent phosphorylation and SUMOylation of a key transcription corepressor complex Cyc8-Tup1 . In turn , these signaling events work in concert to orchestrate spatiotemporal regulation of the Cyc8-Tup1 complex , which undergoes a phase transition to concentrate at specific promoters . These biochemical and cell biological processes are necessary for timely and efficient expression of survival genes , namely those responsible for the production of the key osmolyte glycerol , whose intracellular accumulation provides a feedback loop needed to reset the system following the adaptive process . We summarize these observations in Fig 7 . Environmental stressors are diverse and each stress inhibits cellular functions in unique ways . Cellular stressors can rapidly cause irreparable damage to biomolecules and cellular structures , so it is paramount that cells respond to stress with expediency and efficiency . As such , organisms have evolved fast biophysical mechanisms to expedite slower biochemical responses during adaptation , as seen with the yeast proteins Pab1 and Sup35 that condense into focal structures during stress conditions [15 , 29] . Hyperosmotic stress causes significant water efflux from cells , decreasing the total cellular volume and increasing macromolecular crowding [30] , which has been implicated as an important factor in the condensation of protein-RNA complexes into phase separated droplets in vitro [31] . Moreover , it is well documented that hyperosmotic stress drives the formation of various cytoplasmic membraneless organelles , including stress granules in yeast and P granules in Caenorhabditis elegans [32 , 33] . These dynamic bodies are primarily composed of proteins rich in low complexity sequences that are particularly enriched in polar residues such as asparagine and glutamine [34] . Cyc8 contains a glutamine-rich prion domain that we have previously shown significantly contributes to Cyc8 inclusion formation [10] . The data presented here are consistent with a model that the Cyc8 inclusions are membraneless organelles that form under stress conditions , with control of Cyc8 inclusions regulated in part through SUMOylation and osmolyte concentration . In accord with this model , we found that glycerol biosynthesis was a key regulator of the persistence of Cyc8 inclusions , as their lifetime was inversely correlated with cellular glycerol content after exposure to hyperosmotic stress . Compatible osmolytes have been implicated as regulators of cytoplasmic inclusions during osmotic stress in multiple organisms [32 , 33] . Yeast cells accumulate glycerol following exposure to hyperosmotic stress to reestablish ionic balance , retain water , and counteract molecular crowding [35] . Our data suggest that intracellular osmotic conditions modulate Cyc8 inclusion coalescence and dissolution and concomitant SUMOylation and deSUMOylation . However , we note that there are non-mutually exclusive alternative models that could also account for the observations presented here . Glycerol is capable of preventing protein aggregation in vitro by altering protein-solvent interactions and promoting a larger radius of hydration [36] . In addition , glycerol has been shown to act as a chemical chaperone that promotes the native folding state in proteins susceptible to misfolding [37] . In line with these previous observations , we observe Cyc8 inclusions forming only upon onset of hyperosmotic stress , and resolving only after sufficient accumulation of intracellular glycerol . It is formally possible that glycerol acts as a chemical chaperone to dissolve Cyc8 inclusions; however , our prior studies on Cyc8 inclusions do not show phenomena consistent with stable protein aggregation [10] . Moreover , chemical chaperones are hypothesized to promote native folding , not by direct binding , but rather by stabilizing a protein’s hydration shell [38] . Therefore , we think it is unlikely that glycerol is a direct ligand for Cyc8 , but rather facilitates inclusion dissolution by rehydrating the nucleoplasm and reducing macromolecular crowding in the organelle . Mechanistically , it is unclear whether the formation and subsequent resolution of Cyc8 inclusions occurs spontaneously or is mediated by some as yet unidentified factor . While widespread kinase inhibition by pretreatment with the broad-spectrum kinase inhibitor staurosporine did not change Cyc8 SUMOylation dynamics during adaptation to hyperosmotic stress ( S3 Fig ) , it is possible that some other factor not observed in these studies guides Cyc8 into and out of inclusions . In turn , glycerol accumulation may affect some other signaling pathway that functions to modulate Cyc8 inclusion dynamics . Deciphering further the biophysical and signaling contributions of glycerol effects on Cyc8 dynamic SUMOylation and inclusion formation will be a topic of future study . In the larger context of transcription , previous work has demonstrated that SUMO and the SUMOylation machinery are enriched at the promoters of inducible genes , and both Tup1 and Cyc8 SUMOylation have been implicated in controlling the timing of inducible genes [39–41] . A wide variety of transcriptional/translational modulators are known to be SUMOylated , and the consequences of these SUMOylation events are equally diverse . Similar to our observations with Cyc8 , some of these SUMO substrates are enriched in nuclear inclusions or have prion-like propensities for aggregation , including Arabidopsis TCP proteins and human CPEB3 [42 , 43] . Also , recent advances in super-resolution microscopy and gene editing have shown a well-characterized binding partner of Cyc8 –the Mediator complex–undergoing phase-separation at inducible loci in a transcription-dependent manner [44 , 45] . Given this , it is plausible that biophysically driven inclusion formation of aggregation-prone transcription and translation factors is thematic throughout nature , and occurs in response to diverse stimuli . Once initial scaffolds are constructed , post-translational modifications like SUMOylation allow for the potential recruitment of necessary interactors or regulate the duration of the inclusions to elicit the appropriate temporal biological responses ( such as the recruitment of transcriptional machinery or chromatin modifiers ) . Our work provides new insights on genetic and metabolic factors that can regulate the stress-induced formation of dynamic inclusion bodies in transcription factors , and may shed new light on the interplay of different post-translational modifications during the temporal adaptation to stress .
Yeast strains and plasmids used in this study are listed in Tables 1 and 2 . Standard yeast genetic methods were used for these studies [46] . All gene deletions were verified by colony PCR and phenotypic analyses when available . Cells were grown to a density of ~1 . 5x107 cells/ml at 30°C in yeast complete ( YC ) media prior to stress induction . All 0 time point samples were collected before stress induction . For induction of hyperosmotic stress , equal volumes of culture and YC+2 . 4M sorbitol were combined for a final concentration of 1 . 2M sorbitol . For transcription inhibition , cells were treated at room temperature for 5 minutes with either 4ug/ml thiolutin ( Sigma Aldrich ) , 500ug/ml 1 , 10-phenanthroline ( Sigma Aldrich ) , or an equal volume of DMSO vehicle control prior to the induction of hyperosmotic stress . For broad-spectrum kinase inhibition , cells were treated with staurosporine at the indicated concentrations or an equal volume of DMSO at 30°C for 1 hour prior to the induction of hyperosmotic stress . 50ml aliquots of cells were collected at each time point after stress and flash frozen in liquid nitrogen . Harvested cells were lysed by vortexing with glass beads at 4°C in 1ml denaturing lysis buffer ( 8M urea , 50mM Tris pH 8 . 0 , 0 . 05% SDS with 2mM PMSF and 20mM NEM ) . An aliquot representing 5% of the input was set aside . Cell lysates were incubated with TALON resin ( Novagen ) overnight at 4°C . The resin was washed 3x with wash buffer ( 8M urea , 50mM Tris pH 8 . 0 , 200mM NaCl , 0 . 05% SDS , 5mM imidazole ) . SUMOylated proteins were eluted from the column by addition of loading buffer ( 8M urea , 10mM MOPS , 10mM EDTA , 1% SDS , 0 . 01% bromophenol blue , pH 6 . 8 ) and incubation at 65°C for 10 minutes . SUMOylated proteins were resolved by SDS-PAGE using 4–20% gradient gels . Western analyses were performed with mouse anti-FLAG ( 1:2500 , Sigma ) , mouse anti-HSV ( 1:2500 , Novagen ) , mouse anti-HA ( 1:2500 , Sigma ) , mouse anti-actin ( 1:2500 , Abcam ) , rabbit anti PKA consensus phospho-motif ( RRXS*/T* ) ( 1:2500 , CST ) , or rabbit phospho-p38 ( 1:2500 , CST ) . Chromatin immunoprecipitation was performed as previously described [47] . Briefly , 50ml aliquots of cells were grown to a density of ~1 . 5x107 cells/ml at 30°C in rich medium prior to stress induction . A pre-stress sample was collected prior to induction of stress . Hyperosmotic stress was induced by addition of sorbitol to a final concentration of 1 . 2M , and samples were collected at the indicated times . Protein-DNA crosslinking was initiated by addition of formaldehyde to a concentration of 1% and incubated at room temperature for 15 minutes with continuous swirling . The reaction was quenched by addition of glycine to 0 . 125M for 5 minutes at room temperature . Samples were washed in 1X TBS , collected by centrifugation , and flash frozen in liquid nitrogen . Cells were lysed by bead-beating in 5ml breaking buffer ( 100mM Tris pH 7 . 9 , 20% glycerol , 2mM PMSF ) at 4°C . Insoluble chromatin was collected by centrifugation and resuspended in 700uL ChIP lysis buffer ( 50mM HEPES pH 7 . 5 , 140mM NaCl , 1% Triton X-100 , 0 . 1% deoxycholate , 2mM PMSF ) and sonicated to shear chromatin . A sample prior to immunoprecipitation was collected to represent input controls . Immunoprecipitation was performed on 200ul sheared chromatin by incubation with mouse anti-HSV bound to protein A-sepharose beads overnight at 4°C . Beads were washed twice each with ChIP lysis buffer , high salt ChIP lysis buffer ( 50mM HEPES pH 7 . 5 , 500mM NaCl , 1% Triton X-100 , 0 . 1% deoxycholate , 2mM PMSF ) , ChIP wash buffer ( 10mM Tris pH 8 . 0 , 250mM LiCl , 0 . 5% NP-40 , 0 . 5% deoxycholate , 1mM EDTA , 2mM PMSF ) , and 1X TE buffer . Immunoprecipitated protein-DNA complexes were eluted by incubation in ChIP elution buffer ( 50mM Tris pH 8 . 0 , 1% SDS , 10mM EDTA ) at 65°C for 10 minutes . Samples were treated Proteinase K ( New England Biolabs ) and RNase at 50°C for one hour , followed by removal of crosslinking by incubation at 65C overnight . Input and elute DNA was isolated using QIAprep spin columns ( Qiagen ) and analyzed by qPCR . Quantitative real-time PCR was performed using a 7500 Fast Real-Time PCR system ( Applied Biosystems ) . Reactions were performed on ChIP inputs and eluates using PowerUP SYBR Green Master Mix ( Applied Biosystems ) with specific primers for the GPD1 UAS or ACT1 control . ΔCt values were calculated for each condition and corrected by their respective ACT1 ΔCt . Results were converted to ΔΔCt and normalized to each cell’s respective pre-stress time point . Primer sequences for GPD1: 5’–TCTCACCTCTCACCGCTGAC– 3’ , 5’–AGACTTGCTCAAACCCCAGGAG– 3’ . Primer sequences for ACT1: 5’–TGGCCGGTAGAGATTTGACTGACT– 3’ , 5’–TCGAAGTAAGGCGACGTAACAT– 3’ . 10ml aliquots of cells were grown to a density of ~1 . 5x107 cells/ml at 30°C in rich medium . A pre-stress sample was collected prior to induction of stress . Hyperosmotic stress was induced by addition of sorbitol to a final concentration of 1 . 2M , and samples were collected at the indicated time points . Cells were collected by centrifugation , resuspended in 100uL 1X TBS , and incubated at 95°C for 10 minutes . Supernatant was collected by centrifugation and glycerol concentration in the resulting fraction was measured using a commercial enzymatic assay kit ( Sigma Aldrich ) . Aliquots of cells at each time point after hyperosmotic stress were removed , fixed in 4% paraformaldehyde solution for 15 minutes at room temperature and then washed with PBS . Cells were imaged on a Nikon Eclipse 90i with a100X objective , filters for GFP ( HC HiSN 0 Shift filter set with excitation wavelength ( 450–490 nm ) , dichroic mirror ( 495 nm ) , and emission filter ( 500–550 nm ) ) , and a Photometrics Cool Snap HQ2 cooled CCD camera with NIS-Elements acquisition software . All blots were scanned using an Epson Perfection V350 Photo scanner at 300 dpi . All images were processed with a Mac iMac or Pro computer ( Apple ) using Photoshop CS or CS4 ( Adobe ) . | The ability to sense and react to diverse environmental cues is a central aspect in the maintenance of cellular homeostasis . In response to harsh conditions , cells must rapidly deploy specific stress responses in order to adapt , survive , and proliferate . To ensure optimal spatial and temporal control over stress responses , many proteins undergo biophysical and biochemical alterations . More specifically , these alterations include conformational changes and post-translational modifications–such as phosphorylation , ubiquitination , and SUMOylation–that alter the function , localization , and interactome of target proteins . In this study , we show that the Hog1 MAPK regulates SUMOylation and biomolecular condensation of the yeast transcription corepressor complex Cyc8-Tup1 during exposure to hyperosmotic stress . In turn , this signaling relationship functions to effectively rewire yeast metabolism toward the biosynthesis of the compatible osmolyte glycerol , which serves as the ultimate signal to reset this genetic circuit . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"cellular",
"stress",
"responses",
"monomers",
"cell",
"processes",
"fungi",
"sumoylation",
"mapk",
"signaling",
"cascades",
"polymer",
"chemistry",
"stress",
"signaling",
"cascade",
"proteins",
"chemistry",
"yeast",
"biochemistry",
"signal",
"transduction",
"eukaryota",
"cell",
"biology",
"post-translational",
"modification",
"glycerol",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"physical",
"sciences",
"osmotic",
"shock",
"cell",
"signaling",
"organisms",
"signaling",
"cascades"
] | 2019 | Osmolyte accumulation regulates the SUMOylation and inclusion dynamics of the prionogenic Cyc8-Tup1 transcription corepressor |
An important factor influencing the transmission dynamics of vector-borne diseases is the contribution of hosts with different parasitemia ( no . of parasites per ml of blood ) to the infected vector population . Today , estimation of this contribution is often impractical since it relies exclusively on limited-scale xenodiagnostic or artificial feeding experiments ( i . e . , measuring the proportion of vectors that become infected after feeding on infected blood/host ) . We developed a novel mechanistic model that facilitates the quantification of the contribution of hosts with different parasitemias to the infection of the vectors from data on the distribution of these parasitemias within the host population . We applied the model to an ample data set of Leishmania donovani carriers , the causative agent of visceral leishmaniasis in Ethiopia . Calculations facilitated by the model quantified the host parasitemias that are mostly responsible for the infection of vector , the sand fly Phlebotomus orientalis . Our findings indicate that a 3 . 2% of the most infected people were responsible for the infection of between 53% and 79% ( mean – 62% ) of the infected sand fly vector population . Our modeling framework can easily be extended to facilitate the calculation of the contribution of other host groups ( such as different host species , hosts with different ages ) to the infected vector population . Identifying the hosts that contribute most towards infection of the vectors is crucial for understanding the transmission dynamics , and planning targeted intervention policy of visceral leishmaniasis as well as other vector borne infectious diseases ( e . g . , West Nile Fever ) .
An important factor influencing the transmission dynamics of Vector-Borne Diseases ( VBDs ) is the contribution of hosts with different parasitemia ( no . of parasites per ml of blood ) to the infected vector population ( the Host Infectiousness Profile , HIP ) [1] , [2] . Identifying the hosts that contribute most to the infection of the vectors is crucial for understanding the transmission dynamics of VBDs , as well as for planning intervention strategies targeting the relevant infected host groups [1] , [2] . Calculating this contribution demands quantitative information on the distribution of different parasitemias within the host population and their infectiousness to the vectors . Quantifying the distribution of hosts with different parasitemias within the population is usually achieved through mass screening using molecular , parasitological , or immunological approaches [3] , [4] , [5] . Today , quantifying the infectiousness of hosts with different parasitemias relies on xenodiagnosis or artificial infection experiments which directly determine the infectiousness of the host blood by measuring the proportion of vectors that become infected after feeding on it . However , such experiments can be prohibitively expensive , and require experienced staff and adequate insectary facilities . Moreover , xenodiagnosis on human volunteers is frequently disallowed by institutional ethical review boards or Helsinki committees . Xenodiagnosis or artificial infection experiments are therefore limited to a small number of both , hosts and vectors [2] , [4] , [6] . Thus , efficient estimation of host infectiousness is hindered by small sample sizes . Clearly , a quantitative mathematical model designed to interpolate or extrapolate the infectiousness of hosts with different parasitemias without necessitating direct measurement , would constitute a crucial addition to our disease-fighting arsenal . In this study we develop such a mechanistic model , the first of its kind , that enables the calculation of HIP ( Host Infectiousness Profile ) from given host parasitemia distribution for a wide range of VBDs , and apply it to a data set of human volunteers , potential hosts of Visceral Leishmaniasis ( VL ) in northern Ethiopia [3] . Among VBDs , VL is the second most important killer ( 500 , 000 cases and 70 , 000 deaths annually ) after malaria [7] , [8] . VL is caused by infection with Leishmania ssp . with most cases ( ∼90% ) occurring in the Indian sub-continent , East Africa , and South America [7] , [9] . The sand fly ( Diptera: Psychodidae ) females become infected when they bite infectious humans ( anthroponotic transmission ) or other mammalian hosts ( zoonotic transmission ) [7] , [8] . The Leishmania parasites proliferate in the lumen of its gut and are transmitted to a naïve host when the sand fly female imbibes a subsequent blood-meal [7] , [8] . Although most of the people infected with VL remain free of disease , they may be infectious to biting sand flies and , thereby , contribute to the propagation of the disease . Yet , the potential role of these asymptomatic carriers to the transmission remains largely unknown [10] , [11] , [12] . The main goal of the current study was to estimate the contribution of L . donovani carriers with different blood parasitemias to the infection of Phlebotomus orientalis , the sand fly vector of VL in northern Ethiopia , by developing a customized mathematical model .
To develop the infectiousness model , we assumed that the infectiousness of a host for a biting insect depends exclusively on its blood parasitemia ( although parasites may exist in other tissues , such as the spleen or the skin , e . g . , VL [13] ) . Following the above assumption , let n , be the blood parasitemia of an infected host . The probability that j parasites will be ingested by a vector imbibing volume v of blood per bite form an infected host , r ( j ) , is Poisson distributed ( where nv is its mean ) , ( 1 ) We assume that the infection process within the vector is density-independent ( i . e . , the probability , p , that a single parasite will infect a vector is a fixed value independent of the presence of other parasites ) . Thus , each parasite is equally infectious regardless of the number of parasites ingested . Indeed , the number of Leishmania parasites at the initial stages of the sand fly infection is usually very low ( e . g . , between 1–500 ) , while sand flies with mature infection frequently harbor tens of thousands of parasites [14] , [15] , [16] . Once the infection in the vector gut has been established through this density-independent process , the progression ( that may be density-dependent due to intraspecific competition ) to mature transmissible infection , is thought to be almost definite [14] , [15] , [16] . Thus , the probability that the j parasites in equation ( 1 ) will infect the vector , s ( j ) , is given by: ( 2 ) Since the number of parasites ingested during a single blood meal and the infectiousness of any one of these parasites are independent random variables , we have: ( 3 ) Where q ( n ) in equation ( 3 ) is the infectious probability ( i . e . , the probability that the vector will be infected after biting a host with parasitemia n ) , and λ1 = vp , is the probability that a host with 1 parasite per ml of blood will infect the vector . Empirical studies have demonstrated that a certain proportion of the vectors do not become infected irrespective of the number of parasites ingested during feeding ( i . e . , irrespective of n ) , due to incompatible gut microbiota , for example [17] , [18] , [19] . Let 1-λ2 , be that proportion . By assuming that the infectious probability of the host , q ( n ) , and the probability that a vector is able to become infected , λ2 , are independent , we have: ( 4 ) Equation ( 4 ) represents the infectiousness of a host with parasitemia , n , for a feeding vector . Note that equation ( 4 ) depends on two parameters ( λ1 , λ2 ) , which vary according to vector , host , disease type , and environmental parameters ( e . g . , temperature ) [20] . This equation is our key result since it enables the calculation of the HIP from the distribution of host parasitemias . As part of a cohort study aiming at improving our understanding of the transmission dynamics of VL , blood samples were collected from N = 4 , 695 villagers in the Sheraro district of northern Ethiopia . A highly sensitive quantitative real-time PCR was performed to detect Leishmania kinetoplast DNA ( kDNA ) in dried blood samples [3] . These results were used to obtain the proportion of various parasitemia ranges within the population . The errors in these proportions were calculated according to the variance of a multinomial distribution . Once the data on the distribution of parasitemias in the host population is obtained , it is possible to calculate , with the aid of equation ( 4 ) , the contribution of hosts with different parasitemia levels to the infected vector population , i . e . , the HIP . First , an estimation of the model parameters ( λ1 and λ2 , equation ( 4 ) ) should be performed by fitting the model ( equation ( 4 ) ) to data obtained from artificial infection experiments . Provided that all hosts have an equal probability of being bitten , the calculation of the contribution of hosts with different parasitemia ranges to the infected vector population is straightforward; Let z1 and z2 , z1<z2 , denote two parasite concentrations , and m denote the number of hosts with parasitemias between z1 and z2 . If I1→2 is the proportion of vectors ( from all infected vectors ) that were infected by biting hosts with parasitemias between z1 and z2 , then: ( 5 ) In equation ( 5 ) , the numerator sums the q ( nj ) ( the probability that the vector will be infected by biting a host with parasitemia nj , equation ( 4 ) ) over all m hosts for which parasite concentrations are between z1 and z2 . The denumerator sums the q ( nj ) for the entire cohort , ( i . e . , N = population size = 4 , 695 in our case ) . Previous studies indicate that the vector biting rate is not uniform among the host population but depends on the host's specific volatiles that may cause sick hosts , for example , to be more attractive for the vectors [21] , [22] , [23] , [24] . We assumed uniform biting rates in our model since most people infected with L . donovani remain asymptomatic ( i . e . , disease free ) and are , therefore , presumed not to emit volatiles associated with sickness [8] . Thus , as long as the vectors' preference is not affected by host parasitemia per se , equation ( 5 ) remains valid . A more general version of equation ( 5 ) which takes into account putative vector preferences is discussed below in equation ( 6 ) . Note that although equation ( 4 ) was developed to calculate the probability , q ( n ) , of a vector being infected after a single bite , equation ( 5 ) holds true irrespective of the number of bites the vector delivers through its life . The reason is that the HIP is determined exclusively by naïve vectors , that acquire their infection only once ( infected vectors remain infected till they die ) .
Our infectiousness model ( equation ( 4 ) ) depends on two parameters; λ1 , the infectious probability of a host with one parasite per ml , and λ2 , the probability that a vector will be infected irrespective of the number of parasites ingested during blood feeding . First , the model parameters ( λ1 and λ2 , equation ( 4 ) ) were estimated by performing nonlinear regression analyses of sand fly infection data obtained by artificially feeding P . orientalis sand flies with different concentrations of L . donovani parasites and determining their infection rates ( Figure 1A ) [6] . In order to demonstrate the generality of the model , we also estimated its parameters for previously published data on the infection rates of Aedes aegypti mosquitoes , the vectors of Chikungunya virus , after feeding on bovine blood with different virus concentrations [25] ( Figure 1B ) . The most common way to fit frequency data ( e . g . , infection rate , Figure 1 ) to a set of explanatory variables is through logistic regression [2] . The logistic function ( y = 1/[1+exp ( -x ) ] ) links between a linear combination of the explanatory variables ( x ) and the frequencies that can only take values of positive fractions or zero ( y ) . Yet , the logistic function and the linear combination of the model predictors may not always reflect reality correctly . A mechanistic model that relies on first principles and combines current knowledge of the respective systems may often be advantageous . In Figure 1 we also present the fitted results of the data sets of VL and Chikungunya to a logistic model ( Figure 1 , green lines ) . Clearly , the mechanistic model ( equations ( 4 ) ) has a better fit to the data sets of both diseases . From a cohort of 4 , 695 volunteers , 86% were kDNA-PCR negative ( i . e . without parasites , in their blood , n = 4 , 037 ) while 14% ( n = 658 ) were kDNA PCR positive indicating the presence of Leishmania parasites in their blood , rendering them potentially infectious to biting sand flies . The distribution of parasite concentrations in the infected population ( 14% , n = 658 ) is presented in Figure 2A . Figure 2A indicates that most infected individuals had very low parasitemias ( ∼70% had between 1–10 parasites per ml , n = 458 ) , while very few had high parasitemias ( 3 . 2% had above 1 , 000 parasites per ml , n = 21 ) . The distribution of the host parasitemias obtained from the cohort study ( Figure 2A ) , and the estimation of λ1 and λ2 ( Figure 1A ) , facilitated the calculation of the HIP ( Figure 2B ) via equation ( 5 ) . Figure 2B indicates that persons with 1 , 000 parasites per ml or higher comprised only 3 . 2% of the infected human population , yet according to the model , they were responsible for the infection of between 53% and 79% ( mean of 62% ) of the infected P . orientalis sand flies ( Figure 2B ) . The model can easily be extended to facilitate the calculation of the contribution of a specific host species to the infected vector population in cases where the VBD is hosted by several species . To this end , we now formulate a more general form of equation ( 5 ) ( which calculates the contribution of hosts with certain parasitemia range to the infected vector population , see methods ) , that can be used in cases of multihost VBD . Let N and Nj be the number of host species involved in transmission and the number of individuals of species j ( 1≤j≤N ) in a sample/cohort , respectively . The proportion of vectors that would be infected by feeding on species j from all infected vectors , Ij: , is given by: ( 6 ) In equation ( 6 ) , the numerator sums the contribution of all individuals of species j to the infection of the vector population , and the denominator sums that contribution of all individuals in the community . Note that here , unlike equation ( 4 ) , qj , the probability that host species j would infect the vector as a function of its parasitemia , is species-specific , i . e . , each host species is characterized by different λ1j and λ2j , representing the λ1 and λ2 of species j ( equation ( 4 ) ) . Thus , qj ( nij ) is the probability that an individual i ( 1≤i≤Nj ) of host species j with parasitemia nji would infect a naïve vector . Equation ( 6 ) is in fact a weighted average version of equation ( 5 ) ; the weights , αj , represent the vectors' preference towards species j . Numerous studies indicate that vectors may demonstrate preferences toward certain host species and even to specific individuals within the same species [1] , [26] , [27] , [28] . The preference weights can be determined by attraction experiments , where the number of bites , or the number of vectors attracted to different host species are counted [29] , [30] , [31] . For example , if the numbers of vectors attracted to two host species ( denoted by r and s ) are kr and ks , then: ( 7 ) Where 1≤r≤N &1≤s≤N . By using different combinations of equation ( 7 ) for all 1≤r , s≤N , the various preference weights αj ( equation ( 6 ) , ( 7 ) ) can be calculated . Once these weights , λ1j , λ2j , and the parasitemia of each individual of species j in the sample , nij , ( 1≤i≤Nj ) are measured , Ij can be straightforward calculated via equation ( 6 ) . It should be stressed that the use of equation ( 6 ) is by no means restricted to multihost diseases . It can be used in any case where a host population/community can be divided into groups which vary in their attractiveness to vectors . In general , N , is the total number of host groups displaying distinct levels of attractiveness to vectors , Ni , is the number of individuals in each group ( group i ) , and αi is the vector preference toward group i that can be determined experimentally via equation ( 7 ) .
The model developed in this study facilitated the quantification of the contribution of hosts with different parasitemias to the infected vector population in VBD that are transmitted through blood sucking insects . In the case of Leishmania donovani carriers in northern Ethiopia , our model results indicate that the proportions of infected sand flies that became infected by feeding on persons belonging to different parasitemia categories ( i . e . , the HIP , Figure 2B ) were substantially different from the proportions of these categories within the host population ( Figure 2A ) . This difference can only be quantified by using the model ( equations ( 4 ) and ( 5 ) ) . Our main finding was that only few infected people ( about 3% ) were responsible for the majority ( about 60% ) of the transmissions of VL in the region . Heterogeneity in the contribution of different host groups to the transmission of infectious diseases has been phrased as the “20/80” rule , that is “20% of the hosts contribute at least 80% of the net transmission potential” [1] . However , the “20/80” rule refers to heterogeneity in contact rates between the disease infectious agents ( i . e . , hosts , vectors ) , and not to a difference in the infectiousness of different host groups for the vectors , as the current study does [1] , [32] . This is an important difference , since the basic reproductive number , R0 ( an important measure of disease risk ) in this study , is not necessarily higher compared to a population with homogenous host infectiousness [1] , [32] . As a property of the hosts , the HIP is unaffected by the vectors' biology . For example , temporal or spatial changes in vector abundance such as seasonal fluctuations or spatial heterogeneity in vector density , time delay between biting and infection that cause some vectors to die before infection is acquired [6] , [20] , and low temperature that reduces the success of the infection process within the vector body [20] , may affect the total number of infected vectors , but not the host category they acquired their infection from ( i . e . , the HIP ) . The HIP will preserve its form ( Figure 2B ) even when the parasitemias of individual hosts change with time , as long as the parasitemia categories ( Figure 2A ) remain at dynamic equilibrium ( i . e . , the number of individuals that enter each category per unit time equals the number of individuals who leave it ) . When the parasitemia categories are not at dynamic equilibrium , the parasitemia distributions within the population ( Figure 2A ) and the resulted HIP ( Figure 2B ) will change with time , i . e . , the contribution of each host parasitemia category to the infected sand fly population will be time dependent . The HIP is therefore a property which preservers under wide range of natural circumstances and spatio-temporal heterogeneities , a fact that makes it highly useful for studying the dynamics and control of many VBDs . The presented model outlines a relatively straightforward and fast procedure for calculating the HIP ( Figure 2B ) by facilitating the prediction of the host infectiousness as a function of its parasitemia from the limited data often obtained by xenodiagnostic or artificial feeding experiments ( Figure 1 ) , and combining this prediction with the data on the distribution of host parasitemias within the population ( Figure 2A ) . Furthermore , the model can easily be extended to quantify the contribution of different host groups with different attractiveness to the vectors ( e . g . , other host species ) to the transmission of many multihost VBDs ( equation ( 6 ) ) . According to previous surveys , these VBD types comprise the majority ( >60% ) of VBDs affecting humans ( e . g . , West Nile Fever , other types of Leishmaniasis , [33] , [34] , etc ) . When the vectors show preference toward a specific host group , the basic reproductive rate of the disease , R0 , may increase , and consequently , so is the risk of disease outbreak and morbidity [35] . Our results imply that in the case of VL in northern Ethiopia , intervention focusing on a small fraction of the total population with high parasitemias ( 0 . 45% = 3 . 2% of the 14% infected humans ) , may substantially reduce the number of infected sand flies , and consequently the incidence of VL . Implementation of such targeted interventions is becoming more feasible thanks to the development of rapid mass screening techniques , such as those already recommended by the World Health Organization for prompt and accurate parasitological confirmation of malaria [36] , [37] , [38] . We stress that the mass screening demanded by our model can be both rapid and inexpensive , since it need not be accurately quantitative . It only needs to be sensitive enough to differentiate the hosts with high parasitemias from those with very low ones . It should be stressed that the effectiveness of intervention strategies targeting the 3 . 2% of the hosts with highest parasitemias depends on the absolute size of the infected sand fly population; when the infected vector population decreases , so does the EIR ( the Entomological Inoculation Rate , denoting the number of infectious bites an individual host receives per unit time ) , a well established index for disease persistence and outbreak [32] , [39] . The basic reproductive rate , R0 , of vector-borne diseases is proportional to the EIR , thus both indexes have similar properties; when the EIR is high enough , R0>1 ( i . e . , above the threshold value of R0 = 1 ) and a disease outbreak will occur . The higher is the EIR ( or R0 ) , the higher is the number of predicted infected hosts [32] , [39] . The total number of infected sand flies depends on the HIP as well as the fly population dynamics ( i . e . , birth , death , and migration rates ) that may vary in space and time ( seasonality ) . If the birth and death rates are high , many infected vectors will die and be substituted with new-born susceptible ones . In such a case , although the HIP will be preserved ( Figure 2B ) , the number of infected vectors and their proportion within the total vector population will be lower compared with a case where the population turnover is slow . The EIR also depends on the probability that a bite of an infected vector is infectious to the host . In this study , we discriminated between infected vectors ( vectors with viable parasites in their gut ) and uninfected ones . However , infected vectors may have different infectiousness levels [15] , [40] . Thus , although a reduction in the 3 . 2% most infected hosts will decrease the total number of infected vectors , and consequently , reduce the R0 , EIR , and morbidity , quantifying the success of such a control strategy necessitates the development of a dynamic model that involves data on the vector population dynamics and the distribution of the their levels of infectiousness . Such a model is beyond the scope of the current study . Our study stresses the importance of the HIP due to its robustness to various natural circumstances . In turn , calculating the HIP helps to pinpoint the relevant host groups responsible for the transmission of different VBDs . Further development of dynamic models accounting for host parasitemia profiles ( Figure 2A ) that vary with time and vector population dynamics can broaden our understanding of the role of host groups with different parasitemias in the transmission dynamic of VL , as well as other VBDs . | An important factor influencing the transmission dynamics of vector-borne diseases is the contribution of hosts with different parasitemia ( no . of parasites per ml of blood ) to the infected vector population . In this study we developed a novel mechanistic model that facilitates the quantification of this contribution and applied it to an ample data set of people infected with visceral leishmaniasis in Ethiopia . Among vector borne diseases , visceral leishmaniasis is the second most important killer after malaria . It is caused by infection with Leishmania parasites with most cases ( ∼90% ) occurring in the Indian sub-continent , East Africa , and South America . The disease is transmitted between people and other mammalian hosts by blood-sucking sand flies . Our findings indicate that a 3 . 2% of the most infected people were responsible for the infection of about 65% of the infected sand fly vector population . Identifying the hosts that contribute most towards infection of the vectors is crucial for understanding the transmission dynamics , and planning targeted intervention policy of visceral leishmaniasis , as well as other vector borne infectious diseases ( e . g . , Dengue , West Nile Fever ) . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"plant",
"science",
"emerging",
"infectious",
"diseases",
"infectious",
"disease",
"epidemiology",
"epidemiology",
"vector",
"biology",
"disease",
"dynamics",
"screening",
"guidelines",
"plant",
"pathology",
"biology",
"and",
"life",
"sciences",
"vector-borne",
"diseases",
"infectious",
"disease",
"control",
"sand",
"flies",
"parasitic",
"diseases",
"health",
"care",
"policy",
"disease",
"vectors",
"health",
"care"
] | 2014 | Quantifying the Contribution of Hosts with Different Parasite Concentrations to the Transmission of Visceral Leishmaniasis in Ethiopia |
DNA sequences on X chromosomes often have a faster rate of evolution when compared to similar loci on the autosomes , and well articulated models provide reasons why the X-linked mode of inheritance may be responsible for the faster evolution of X-linked genes . We analyzed microarray and RNA–seq data collected from females and males of six Drosophila species and found that the expression levels of X-linked genes also diverge faster than autosomal gene expression , similar to the “faster-X” effect often observed in DNA sequence evolution . Faster-X evolution of gene expression was recently described in mammals , but it was limited to the evolutionary lineages shortly following the creation of the therian X chromosome . In contrast , we detect a faster-X effect along both deep lineages and those on the tips of the Drosophila phylogeny . In Drosophila males , the dosage compensation complex ( DCC ) binds the X chromosome , creating a unique chromatin environment that promotes the hyper-expression of X-linked genes . We find that DCC binding , chromatin environment , and breadth of expression are all predictive of the rate of gene expression evolution . In addition , estimates of the intraspecific genetic polymorphism underlying gene expression variation suggest that X-linked expression levels are not under relaxed selective constraints . We therefore hypothesize that the faster-X evolution of gene expression is the result of the adaptive fixation of beneficial mutations at X-linked loci that change expression level in cis . This adaptive faster-X evolution of gene expression is limited to genes that are narrowly expressed in a single tissue , suggesting that relaxed pleiotropic constraints permit a faster response to selection . Finally , we present a conceptional framework to explain faster-X expression evolution , and we use this framework to examine differences in the faster-X effect between Drosophila and mammals .
Comparing the evolutionary rates of X-linked ( or Z-linked ) and autosomal genes can be informative of the nature of allelic dominance [1] , the type of variation acted upon by natural selection [2] , [3] , the mutational process [4]–[8] , and the effect of differences in population size on the efficacy of natural selection across taxa [9] , [10] . Notably , DNA sequences on X ( or Z ) chromosomes often evolve faster than autosomal sequences ( i . e . , the “faster-X” effect ) . This may be a result of the adaptive fixation of recessive beneficial mutations in X-linked genes [1] , [11]–[14] , mutational biases associated with dosage compensation [15] , or the smaller effective population size ( ) of sex chromosomes [9] , [10] . The faster-X effect is especially pronounced in the protein coding sequences of genes with male-biased expression ( i . e . , genes expressed higher in males than females ) or genes specifically expressed in male reproductive tissues in male heterogametic ( XY ) taxa [16]–[20] . These results support the theoretical prediction that the adaptive fixation of recessive X-linked male-beneficial mutations in hemizygous males can drive faster-X evolution [1] . Comparisons of expression divergence between X-linked and autosomal genes are not as prevalent as analyses of DNA sequences . Some experiments have suggested that the disproportionate effect of X-linked loci on interspecific hybrid fitness ( the “large X” effect [21] ) is the result of divergence in the regulation of gene expression . For example , gene expression from the X chromosome may be misregulated in the male germline of interspecific hybrids [22]–[24] , and dosage compensation of the X chromosome could also be affected in hybrids [25]–[27] . With the advent of high throughput technologies to measure expression in multiple species we can now directly test whether the rate of expression evolution differs between X-linked and autosomal genes . The first such analysis did indeed find evidence for the faster-X evolution of gene expression shortly following the creation of the therian X chromosome [28] . Gene expression is determined by an interaction of cis regulatory elements and the proteins that bind to them ( e . g . , transcription factors , histones ) to either promote or inhibit transcription . X chromosomes often have a unique chromatin environment because of the need to dosage compensate X-linked genes in males . In mammals , this is hypothesized to be accomplished by the upregulation of X-linked gene expression in both sexes , followed by random silencing of one X chromosome in females [29]–[31] ( although this model is not universally accepted [32] ) . Drosophila compensate for reduced X chromosome dose in males by modifying the chromatin structure of the X in a male-specific manner . The dosage compensation complex ( DCC; or male-specific lethal [MSL] complex ) , a ribonucleoprotein structure , binds the X chromosome in males , acetylating histone H4 at lysine 16 [33]–[35] . This is thought to promote the expression of X-linked genes via some combination of relaxing compacted chromatin [36] , [37] , enhancing recruitment of RNA polymerase II [38] , and/or increasing transcriptional elongation [39] . The DCC only assembles in males because one of the essential proteins , MSL-2 , is not produced in females [40]–[42] . Recently , chromatin immunoprecipitation ( ChIP ) experiments followed by microarrays ( ChIP-chip ) or sequencing ( ChIP-seq ) have revealed regions of the Drosophila melanogaster X chromosome that are enriched with DCC binding and bound by the DCC in the absence of essential DCC components [43] , 44 . These chromatin entry or high affinity sites ( HASs ) contain a DNA sequence motif that is thought to direct the DCC to the Drosophila X chromosome [43] , [44] . After initially binding to the 100–300 HASs , the DCC is hypothesized to spread in cis to promote the upregulation of expression by inducing transcriptionally activating chromatin marks [45]–[49] . To examine how X-linkage , chromatin environment , and breadth of expression affect the evolution of gene expression , we calculated expression differences between Drosophila species using data collected from male and female whole flies and heads using microarrays and high throughput RNA sequencing ( RNA-seq ) [50]–[52] . We detect a robust signal of faster-X evolution of gene expression . This faster-X effect is most pronounced in genes that are located in transcriptionally repressive chromatin in cell culture and genes that are narrowly expressed in a limited number of tissues . In addition , we analyzed measurements of intraspecific variation in gene expression , and we found that the faster-X effect cannot be explained by relaxed selective constraints . Our results suggest that the faster-X evolution of gene expression is the result of the adaptive fixation of X-linked mutations that affect gene expression in cis .
We analyzed expression measurements in six Drosophila species ( D . melanogaster , Drosophila yakuba , Drosophila ananassae , Drosophila pseudoobscura , Drosophila mojavensis , and Drosophila virilis ) collected from whole females and males separately using microarrays . Following the method of Brawand et al . [28] , we calculated the similarity in expression between pairs of species , within each sex , using Spearman's rank correlation coefficient ( ) sampling only genes present as 1∶1∶1∶1∶1∶1 orthologs across all six species [53] . To determine if the expression levels of X-linked genes diverge faster than autosomal genes , we compared across the five major chromosome arms ( also known as Muller elements [54] ) . In every pairwise comparison , the correlation of expression levels of X-linked genes ( ) , in both females and males , is significantly lower than that of autosomal genes ( ) ( Figure 1 ) . We confirmed this pattern using a different pipeline to handle the microarray data ( Figure S1 ) and with RNA-seq data from three species ( D . melanogaster , D . pseudoobscura , and D . mojavensis; Figure S2 ) . These results suggest that X-linked gene expression levels diverge faster than the expression levels of autosomal genes . While there is congruence between the microarray and RNA-seq data with regards to the faster-X evolution of expression , we observe two notable differences between these data sets . First , expression levels estimated using RNA-seq are more highly correlated than those estimated from microarray data ( Figure 1 , Figure S2 ) , possibly because of the increased dynamic range of RNA-seq [55] . Second , the magnitude of the difference between and is greater in the microarray data than the RNA-seq data ( Figure 1 , Figure S2 ) . We reanalyzed the microarray data using only the genes present in the RNA-seq dataset , and these correlations resemble the microarray analysis more than the RNA-seq ( Figure S3 ) . Therefore , the difference in magnitude of in the microarray and RNA-seq analyses is not attributable to differences in the gene sets analyzed . Regardless of the cause of these differences , both methodologies provide evidence in support of the faster-X evolution of gene expression ( Figure 1 , Figure S2 ) . The faster-X evolution of Drosophila protein coding sequences is especially pronounced in genes with male-biased expression that are expressed in male reproductive tissues [18]–[20] , possibly because the hemizygosity of the X chromosome favors the adaptive fixation of recessive male-beneficial mutations in X-linked genes [1] . Additionally , genes with male-biased expression tend to have more divergent expression between species than genes with female-biased or non-sex-biased expression [56] , [57] . The faster-X evolution of gene expression , however , is not limited to genes expressed in male reproductive tissues: we detect the faster-X effect when gene expression is measured in females ( Figure 1 ) or heads ( Figure 2 , Figure S4 ) , although the pattern is not as striking as when whole fly data are used . To further examine the effect of expression in male-reproductive tissues on the faster-X effect , we excluded genes with male-biased expression in D . melanogaster ( 765 genes at a false discovery rate [FDR] of 0 . 05 ) , male-biased expression in all of the species ( 2027 genes at a FDR of 0 . 20 ) , or biased expression in male reproductive tissues in D . melanogaster ( 439 genes based on expression data from FlyAtlas [58] ) . In all cases , we detect the faster-X evolution of gene expression even when genes with male-biased expression are removed ( Figures S5 , S6 , S7 ) . In addition , genes that are narrowly expressed in non-reproductive tissues exhibit a faster-X effect comparable to genes with biased expression in male reproductive tissues ( Figure 3 ) . On the other hand , we fail to detect the faster-X effect when we consider only genes with biased expression in female-limited reproductive tissues ( Figure 3 ) . We therefore conclude that the faster-X evolution of gene expression requires expression in males but not necessarily male-biased expression . Genes that arose recently by duplication tend to have male-biased expression [59] , [60] . Many new Drosophila genes are located on the X chromosome and show evidence of a faster-X effect in their protein coding sequences [61] . We find that new genes ( those that arose after the split between the D . melanogaster and D . virilis lineages ) do not exhibit evidence of faster-X expression evolution , while old genes shared by the entire genus do ( Figure S8 ) . Therefore , the faster-X evolution of male reproductive genes , genes with male-biased expression , or new genes are not entirely responsible for the faster-X evolution of gene expression in Drosophila . Previous work in mammals found evidence for faster-X evolution of gene expression that was limited to evolutionary lineages closely following the creation of the therian X chromosome [28] . To test for a similar lineage-specific faster-X effect in Drosophila , we used our calculations of from whole fly expression measurements between species to estimate branch lengths along the Drosophila phylogeny . This approach assumes that there is a phylogenetic signal in these pairwise correlations . Some pairwise correlations are lower for more closely related species than more distantly related ones ( Figure 1 ) , suggesting that the correlations may not have an underlying phylogenetic signal . To further test for phylogenetic signal , we estimated the divergence in expression between species as [28] . We were indeed able to reconstruct the evolutionary relationships of the six species using this distance matrix ( Figure S9 ) , demonstrating that the expression correlations contain a phylogenetic signal . We next tested whether the faster-X evolution of gene expression is limited to particular branches in the Drosophila phylogeny using matrices of and to estimate branch lengths along the known topology ( Figure 4 ) . These branch lengths represent the contribution that each lineage makes toward , and larger values indicate a lower correlation in expression . Phylogenies constructed using X-linked gene expression from both males and females have longer internal and terminal branch lengths ( Figure 4 ) , unlike in mammals where a faster-X effect is only detected on internal branches [28] . Interestingly , branch length estimates closest to the root do not show evidence for a faster-X effect in Drosophila ( Figure 4 ) . This is not necessarily evidence against the faster-X evolution of gene expression along these internal branches . We instead hypothesize that it is the result of low power to resolve deep branching orders using these correlation matrices , which leads to poor estimates of branch lengths around deep nodes . Supporting this hypothesis , when we use the correlation matrices to estimate the tree topology , some of the deep nodes have the lowest bootstrap support for the correct topology ( Figure 4 , Figure S9 ) . In addition , the bootstrap support for these nodes is lower for X-linked gene expression levels than autosomal expression ( Figure 4 ) . Furthermore , when we exclude genes on the X chromosome , we observe an increase in bootstrap support for the correct branching order between D . pseudoobscura and the melanogaster group ( Figure 4 , bottom number in bootstrap boxes , in italics ) . We therefore hypothesize that the faster-X evolution of gene expression complicates the inference of the correct branching order more for X-linked genes than autosomal genes at these deep nodes , leading to a flawed measurement of the branch lengths . In summary , depending on the branch in question , either longer branch lengths or lower bootstrap values support the hypothesis that X-linked gene expression levels diverge faster than autosomal expression levels across most of the phylogeny . D . pseudoobscura and Drosophila willistoni each have an independently derived neo-X chromosome arm ( Muller element D ) that is autosomal in all other species [62] . If the faster evolution of gene expression closely follows the creation of an X chromosome , we would expect to detect a faster-X effect in genes on these neo-X chromosome arms . Using available RNA-seq data collected from D . pseudoobscura , D . willistoni , D . melanogaster , and D . mojavensis heads [52] , we find some evidence for the faster evolution of gene expression on the neo-X chromosome arms ( Figure S4 ) . However , we fail to detect evidence for faster-X expression evolution in genes on the D . pseudoobscura neo-X chromosome when expression is measured in whole fly ( Figure 1 , Figure 4 ) . The latter result may be because of low power to detect a faster-neo-X effect; the X-autosome fusion giving rise to the D . pseudoobscura neo-X occurred recently relative to the pseudoobscura-melanogaster common ancestor [60] , [63] . The Drosophila X chromosome has a unique chromatin environment because of the need to compensate dosage in hemizygous males [33] , [34] , [36] , [64] , and these histone modifications are correlated with gene expression levels [65] . We therefore considered whether DCC binding is associated with the faster-X effect . To do so , we calculated pairwise expression divergence for each 1∶1∶1 ortholog between D . melanogaster , D . yakuba , and D . ananassae . We selected these three closely related species because DCC binding and chromatin states have only been inferred for D . melanogaster [45] , [46] , and these inferences are less likely to be accurate in more distantly related species . In addition , these gene-wise estimates of expression divergence differ from the correlations in expression levels across entire chromosomes ( see Methods ) . We chose this approach because , as we introduce more parameters into our analysis , gene-wise expression divergence is easier to interpret than correlations of chromosome-wide expression between species . Using the gene-wise estimates of expression divergence between D . melanogaster and either D . yakuba or D . ananassae , we found that X-linked genes that are unbound by the DCC have greater expression divergence than DCC bound X-linked genes ( Figure 5A ) . Additionally , in the comparison between D . melanogaster and D . ananassae , unbound X-linked genes have greater expression divergence than autosomal genes ( Figure 5A ) . Furthermore , the expression levels of DCC bound X-linked genes are more evolutionarily conserved than autosomal genes ( Figure 5A ) . DCC bound genes tend to be in close proximity to HASs [48] , and HASs have the highest concentration of DCC binding [43] , suggesting that proximity to an HAS may also predict expression divergence . Distance from the nearest HAS is indeed positively correlated with expression divergence ( Figure 5B ) . We observe these patterns when expression is measured in either females or males ( Figure 5A–5B ) . Highly expressed genes tend to have more conserved protein coding sequences [19] , [66] , and there may be a positive correlation between the rate of protein coding sequence evolution and divergence in gene expression [67]–[69] . Genes bound by the DCC have higher expression levels than unbound genes [48] , suggesting that the relationship between DCC binding and expression divergence ( Figure 5A–5B ) may be a byproduct of highly expressed genes with less expression divergence . We found a negative correlation between expression level and expression divergence for both X-linked and autosomal genes ( Figure 5C ) , demonstrating that highly expressed genes have more conserved expression levels . To test whether the relationship between DCC binding and expression divergence is merely an artifact of the correlation between expression level and expression divergence , we calculated partial correlations between expression divergence , distance from the nearest HAS , and expression level . If DCC binding and expression divergence are directly related , genes further from an HAS should have elevated expression divergence even when expression level is taken into account . Distance from the nearest HAS is positively correlated with expression divergence in most of our partial correlations ( Figure 5D ) , demonstrating that genes that are not directly regulated by the DCC have faster evolving expression levels . In addition , expression level and expression divergence are negatively correlated ( Figure 5D ) , supporting the hypothesis that highly expressed X-linked genes have more conserved expression levels independent of DCC binding . Lastly , distance from an HAS is negatively correlated with expression level ( Figure 5D ) , providing additional evidence that highly expressed genes are more directly regulated by the DCC [48] . The DCC is both attracted to and promotes chromatin modifications associated with transcriptional activity [35] , [49] , suggesting that genes unbound by the DCC are in transcriptionally repressive chromatin . The faster expression evolution of X-linked genes that are unbound by the DCC could therefore be a general property of genes associated with repressive chromatin . To test this hypothesis , we obtained mapped chromatin states in the D . melanogaster genome from two different cell lines [65] , and we used these data to assign genes to one of two chromatin states: transcriptionally active or repressive . We found that X-linked genes that are bound by the DCC are indeed almost always ( 97 . 8–100% ) associated with active chromatin , while unbound genes tend to be in repressive chromatin ( Table S1 ) . In addition , genes in transcriptionally active chromatin have higher expression levels than genes in repressive chromatin ( Figure S10 ) . Both autosomal and X-linked genes associated with transcriptionally repressive chromatin have more divergent expression levels than genes associated with active chromatin regardless of which cell type is used to infer chromatin states ( Figure 6 ) . However , X-linked genes that are located in repressive chromatin have more divergent expression between D . melanogaster and D . ananassae than autosomal genes in repressive chromatin ( Figure 6 ) . Furthermore , X-linked genes associated with active chromatin tend to have less expression divergence than other genes in the genome ( Figure 6 ) . We observe similar patterns when we use DCC binding as a proxy for transcriptionally active chromatin in X-linked genes ( Figure S11 ) . These results provide further support for the hypothesis that the faster-X evolution of gene expression is driven by genes that are not directly regulated by the DCC . Genes expressed narrowly ( i . e . , in a limited number of tissues ) tend to have rapidly evolving protein coding sequences [19] , [66] , which raises the possibility that expression breadth may also affect expression divergence . We find that narrowly expressed genes do tend to have elevated expression divergence ( Figure 7A , Figure S12 ) . In addition , DCC bound genes tend to be broadly expressed , and genes further from an HAS are more narrowly expressed ( Figure 7B , Figure S13 ) . Furthermore , genes that are in transcriptionally active chromatin tend to be broadly expressed , while genes in repressive chromatin tend to be narrowly expressed ( Figure 7C , Figure S14 ) . This raises the possibility that the association between chromatin environment and the faster-X effect ( Figure 6 ) could be an artifact of the correlation between expression breadth and expression divergence . If the faster-X evolution of gene expression is affected by expression breadth and not chromatin environment , we expect to only detect the faster-X effect in narrowly expressed genes . Consistent with this prediction , we detect the strongest evidence for faster-X evolution in narrowly expressed genes ( Figure 8 , Figure S15 ) . The faster-X effect is , however , limited to narrowly expressed genes in transcriptionally repressive chromatin ( Figure 8 , Figure S15 ) , suggesting that narrow expression and transcriptionally repressive chromatin environment both promote faster-X expression evolution . Narrowly expressed genes in transcriptionally repressive chromatin are more likely to have low expression levels [19] , [66] ( Figure S10 ) , which could increase the error in expression level measurements . However , measurement error is unlikely to explain the association of expression breadth and chromatin environment with the faster-X effect for two reasons . First , experimental and biological variance should not produce the consistent signal of faster-X evolution . Second , we still detect the faster-X effect when genes with low expression levels are excluded ( Figure S16 ) . x The faster-X effect could be a result of differences in gene content between the X chromosome and the autosomes if , for example , X-linked genes were more narrowly expressed or more likely to be in transcriptionally repressive chromatin than autosomal genes . The D . melanogaster X chromosome , however , harbors a deficiency of narrowly expressed genes [52] , [70] , and there is a paucity of X-linked genes in repressive chromatin ( Figure 6; using Fisher's exact test ) . In addition , we fail to detect a significant difference in expression breadth between X-linked and autosomal genes in repressive chromatin ( Figure S17 ) . It is therefore unlikely that the unique gene content of the Drosophila X chromosome is responsible for the faster-X effect . Our power to detect the faster-X effect is also limited by the small sample size of narrowly expressed genes in repressive chromatin on the X chromosome , demonstrating that our results are conservative . The faster-X evolution of protein-coding sequences is most pronounced for genes that are narrowly expressed in male reproductive tissues [19] , [20] . We showed , however , that expression in male reproductive tissues is not solely responsible for the faster-X evolution of gene expression ( Figure 2 , Figure 3; Figures S4 , S5 , S6 , S7 ) . This does not exclude the possibility that X-linked genes expressed in male reproductive tissues have faster evolving expression levels than autosomal genes ( e . g . , Figure 3 ) . We do find some support for the faster-X evolution of male expression levels among genes in repressive chromatin that are expressed narrowly in male reproductive tissues , but the evidence is not exceedingly strong ( Figure S18 ) . Most notably , we fail to detect the faster-X effect when we limit the analysis to genes in repressive chromatin that are narrowly expressed in female reproductive tissues ( Figure S18 ) , consistent with our earlier analysis of chromosome-wide correlations of expression ( Figure 3 ) . Therefore , genes with limited expression in females do not experience faster-X expression evolution . Accelerated evolutionary divergence can be the result of relaxed selective constraints or an elevated rate of adaptive evolution . To distinguish between these two explanations for increased divergence in gene expression one can examine intraspecific variation in expression levels [57] , [71] , [72] . If relaxed selective constraints were responsible for greater divergence , we would expect increased intraspecific variation in genes with rapidly evolving expression levels . Conversely , if the fast evolution of gene expression is driven by positive selection , we expect rapidly evolving genes to have equivalent ( or less ) expression polymorphism when compared to non-rapidly evolving genes . In making these interpretations we assume that expression variation segregating in natural populations has neutral or slightly deleterious fitness effects , an assumption common to the interpretation of DNA sequence polymorphism and divergence data [73] , [74] . One way to estimate intraspecific variation is to compare expression levels between females and males of the same species . Higher correlation between sexes suggests greater constraint on gene expression . We find that X-linked expression levels are often more correlated between the sexes than autosomal expression levels ( Figure 9A ) , suggesting that X-linked expression levels are not under relaxed constraints . We also used available calculations of the broad sense heritability ( ) of gene expression measured in whole flies from 40 inbred D . melanogaster lines [75] as an estimate of the intraspecific variation in gene expression contributed by genetic factors . Higher implies greater genetic variation underlying gene expression levels , which suggests relaxed selective constraints . In the results presented below , all genes with estimates of were included , but we obtain similar results if we limit ourselves to only genes included in our analysis of expression divergence . Consistent with a previous analysis of the same data [75] and independent experiments in Drosophila simulans [76] , we detect significantly reduced among X-linked genes ( Figure 9B ) . This provides further evidence that the expression regulation of X-linked genes is not under relaxed selective constraints and that the faster-X effect is not a result of relaxed constraints . The faster-X evolution of expression is most pronounced for genes that are unbound by the DCC , in transcriptionally repressive chromatin , or narrowly expressed ( Figure 5A , Figure 6 , Figure 8 ) . If the faster-X effect were the result of relaxed selective constraints , we should observe increased values in genes with the most pronounced faster-X effect . Consistent with this prediction , X-linked genes that are unbound by the DCC have higher than bound genes ( Figure 9C ) , suggesting that unbound genes are under relaxed constraints . Comparisons between X-linked and autosomal genes with similar expression breadth or in similar chromatin environments , however , suggest that the faster-X effect is not the result of relaxed constraints . For example , while narrowly expressed genes have elevated values , there is not a significant difference in between X-linked and autosomal narrowly expressed genes ( Figure 9D ) . Additionally , X-linked genes in repressive chromatin tend to have lower than autosomal genes in repressive chromatin ( Figure 9E ) . Faster-X expression evolution is therefore unlikely to be a result of relaxed selective constraints on X-linked expression levels .
Our analysis relies on inferences of DCC binding and HASs based on experiments performed in D . melanogaster cells [43] , [45] . DCC proteins have experienced adaptive evolution along the D . melanogaster lineage [27] , [77] , as have three HASs on the X chromosome [78] . Despite the potential for enhanced expression divergence because of this rapid evolution , we see greater conservation of expression levels associated with DCC bound genes ( Figure 5A ) . This result implies that , despite the accelerated evolution of DCC components and their binding sites , DCC binding is likely to be conserved across species . Furthermore , if DCC binding sites are turning over , this makes our discovery of a relationship between DCC binding and expression divergence conservative . DCC binding and chromatin states were inferred in a limited number of D . melanogaster cell lines . Genes that were never identified as bound by the DCC are either never compensated ( because their dose does not need to be tightly controlled ) or are compensated in cell types other than those studied thus far . Similarly , genes categorized as in regions of transcriptionally repressive chromatin are likely to be transcriptionally activated in a tissue-specific manner that differs from their state in S2 or BG3 cells . In this way , chromatin state can be used as a second , independent measurement of expression breadth: genes inferred to be in repressive chromatin can be assumed to be narrowly expressed , while genes in active chromatin are likely to be broadly expressed ( Figure 7C ) . Narrowly expressed genes tend to have rapidly evolving protein coding sequences , possibly because they are under fewer evolutionary constraints [19] , [66] . Not only can these relaxed constraints permit faster evolution by purely neutral processes , but genes that are under fewer constraints are also expected to have a higher likelihood of adaptive fixations because they have less pleiotropic restrictions on their evolution [79] . Similarly , genes that are unbound by the DCC , genes that are in transcriptionally repressive chromatin , and narrowly expressed genes have rapidly evolving expression levels ( Figure 5 , Figure 6 , Figure 7 ) . In addition , these genes also have more intraspecific variation in expression ( Figure 9 ) , as do genes with fewer annotated functions [80] , suggesting that the regulation of their expression is under relaxed constraints . If the faster-X evolution of gene expression is driven by positive selection , the faster-X effect should be most pronounced in genes that are most likely to experience adaptive substitutions . Genes in transcriptionally repressive chromatin and narrowly expressed genes do indeed have the most robust evidence for a faster-X effect ( Figure 6 , Figure 8 ) , supporting an adaptive model of faster-X expression evolution . The canonical model of faster-X evolution driven by positive selection posits that X-linked recessive beneficial mutations will be exposed to selection in hemizygous males , this will lead to an increased probability of invasion for X-linked beneficial alleles , and there will be a higher rate of adaptive evolution in X-linked genes [1] , [3] . Before we can apply this model to the faster-X evolution of gene expression , we must determine if two assumptions are met: 1 ) mutations that affect the expression of X-linked genes are themselves X-linked; 2 ) mutations that affect expression level are recessive . We consider each of these assumptions below , and then we develop a conceptual framework to explain the faster-X evolution of expression . Gene expression is inherited in a polygenic manner [81] , and both cis and trans acting factors are responsible for expression differences between Drosophila species [82] , [83] . The expression divergence of X-linked genes is therefore determined by substitutions in X-linked cis regulatory sequences and the trans acting proteins that bind to them . While the cis regulatory elements are all X-linked , the trans factors can be encoded by X-linked or autosomal genes . There are trans factors that preferentially affect the expression of X-linked genes ( e . g . , some nuclear pore proteins [84] , [85] , the DCC [35] , and chromatin modifications that are enriched on the X chromosome [64] ) , but these are unlikely to be the norm [65] . In addition , the preferential targeting of certain trans factors to X-linked loci is ultimately attributable to sequences that are enriched on the X chromosome—either X-linked motifs direct trans factors to the X chromosome [43] , [44] or trans factors are attracted by other proteins that are enriched on the X chromosome because they themselves were directed there by X-linked sequences [85] . Similarly , the expression of autosomal genes is determined by X-linked and autosomal trans acting factors , but the cis regulatory sequences are all autosomal . It is well documented that cis regulatory changes play an important role in gene expression divergence [82] , [83] , [86]–[88] . Therefore , expression changes in X-linked genes are more likely to be the result of mutations in X-linked loci when compared to similar expression changes in autosomal genes . This supports the hypothesis that the faster-X evolution of gene expression is the result of an increased rate of X-linked substitutions affecting expression levels . If the faster-X evolution of expression is driven by adaptive substitutions in the cis regulatory sequences of X-linked genes , we would expect to detect signatures of positive selection near genes on the X chromosome . X-linked loci in D . melanogaster do tend to have reduced genetic variation , and this can be attributed to genetic hitchhiking because of selection at loci within or near X-linked genes [13] , [89] , [90] . In addition , DNA sequence variation is positively correlated with intraspecific expression variation in D . simulans [76] , and sequence divergence upstream of coding sequences is correlated with expression divergence between D . melanogaster and a close relative [83] . These patterns further support the hypothesis that the faster-X evolution of gene expression is driven by X-linked substitutions that affect expression level in cis . While it is clear how X-linked mutations can affect the expression of X-linked genes , it is not obvious why those mutations would be recessive . Non-additive inheritance of gene expression levels is common [91] , [92] , but cis regulatory differences between species are more likely to be inherited in an additive manner [83] , [93] . These results suggest that the phenotypic effects of mutations that affect expression in cis are not likely to be recessive , but what is more important is whether the fitness effects of the mutations are recessive . It is reasonable to assume that the fitness landscape near an optimum is concave , which implies that mutations that push the expression level of a gene toward the optimum will be dominant [94] , [95] . Therefore , empirical results and theoretical predictions appear to challenge the assumption that beneficial mutations that affect expression in cis will be recessive . Recently , however , it has been demonstrated that beneficial mutations with additive phenotypic effects can increase fitness in heterozygotes while at the same time being less fit when homozygous because they overshoot the adaptive peak [96] , [97] ( Figure 10A ) . Therefore , mutations with additive phenotypic effects can have over-dominant fitness effects as a consequence of diploidy [97] . While this could impede adaptation at autosomal loci , the dynamics of this process are likely to differ at X-linked genes because they are effectively haploid in males in the absence of dosage compensation ( Figure 10B ) . Notably , we detect the faster-X effect in genes that appear not be dosage compensation ( Figure 5 ) . Beneficial mutations with additive phenotypic effects on the expression of these uncompensated X-linked genes may therefore be more likely to fix because selection in males does not run the risk of overshooting the fitness optimum as a consequence of diploidy ( Figure 10 ) . Further theoretical work is needed to determine whether this intuitive prediction is a feasible adaptive explanation for the faster-X evolution of gene expression . If the aforementioned model could explain the faster-X evolution of gene expression , we would expect the faster-X effect to be limited to genes expressed in males because they would be present in the hemizygous ( i . e . , haploid ) state . Consistent with this hypothesis , the expression levels of X-linked genes transcribed primarily in female-limited tissues do not evolve faster than autosomal genes with equivalent expression profiles ( Figure 3 ) . We do , however , detect the faster-X effect when expression is measured in either males or females ( Figure 1 , Figure 2 , Figure 4 , Figure 6 , Figure 8 ) . This is because male and female phenotypes are correlated so that selection on expression levels in males will affect the expression levels of genes that are also expressed in females [98] , [99] . Within the Drosophila genus , we observe the most pronounced faster-X effect along the lineage leading to D . ananassae ( Figure 4 , Figure 6 , Figure 8 ) . The Painting of fourth ( POF ) protein localizes specifically to the diminutive D . melanogaster fourth chromosome ( Muller element F , or dot chromosome ) [100] . Numerous lines of evidence suggest that POF promotes the transcriptional output of genes on the fourth chromosome by an unknown RNA-binding mechanism [101]–[103] . Intriguingly , while POF is dot-chromosome-specific in most Drosophila species , it also co-localizes with the DCC on the X chromosome in D . ananassae males [104] . POF localization to the D . ananassae X chromosome in males suggests that X-linked gene expression is uniquely affected in D . ananassae . This could contribute to the increased expression divergence of the X chromosome along the D . ananassae lineage either by directly affecting the expression levels of D . ananassae genes or by creating unique selection pressures on X-linked gene expression . We detect a substantial faster-X effect in female expression along the D . ananassae lineage ( Figure 4A ) , despite the fact that POF does not localize to the X chromosome in D . ananassae females [104] . Therefore , the accentuated faster-X effect along the lineage leading to D . ananassae is unlikely to be a direct result of POF modifying expression levels . It is instead more likely that POF localization to the X chromosome in D . ananassae creates novel selection pressures on X-linked expression levels , which leads to a more pronounced faster-X effect . Expression levels of X-linked genes diverge faster than those of autosomal genes along both internal and terminal branches of the Drosophila phylogeny ( Figure 4 ) . The faster-X effect in mammals , on the other hand , is limited to only some deep lineages [28] . If our conceptual framework for understanding the faster-X evolution of gene expression is correct , we should be able to use it to explain differences in the faster-X effect between Drosophila and mammals . We consider four hypotheses that could explain the extent of faster-X expression evolution in the two taxa , and we examine how each could contribute to the observed incongruities . First , X chromosome gene content differs between Drosophila and mammals [52] , [105] . These differences , however , are unlikely to be responsible for the differences in the faster-X effect between mammals and Drosophila . The mammalian X chromosome harbors an excess of narrowly expressed genes [52] , [106] , i . e . , the same type of genes with the most pronounced faster-X effect in Drosophila ( Figure 8 ) . Therefore , we would expect an even more substantial faster-X effect in mammals if differences in X chromosome gene content were an important contributor to taxon-specific faster-X expression evolution . Second , Drosophila and mammals deal with the haploid dose of the male X chromosome in different ways [29] , [37] . Faster-X gene expression evolution in Drosophila is most pronounced for genes that are unbound by the DCC ( Figure 5 ) , and we hypothesize that the effective haploidy of X-linked alleles in uncompensated Drosophila genes promotes the faster-X effect ( Figure 10 ) . Mammalian dosage compensation , on the other hand , is thought to be a two step process: X chromosome gene expression is upregulated in both sexes , followed by random silencing of one X chromosome in females [29]–[31] . The specific mechanisms of mammalian X chromosome upregulation are not understood , and the phenomenon itself remains controversial [32] . Regardless of the details of mammalian dosage compensation , if the allelic dominance of fitness effects for mutations that change gene expression are affected by the mechanism of dosage compensation , the differences in dosage compensation between Drosophila and mammals could be responsible for the taxon-specific patterns of faster-X expression evolution [1] , [3] . Third , the rate of evolution depends on mutational input and the fixation rate of those mutations . A higher autosomal mutation rate could therefore counteract a higher fixation rate on the X chromosome [107] . The mutation rate in the germline of many mammals is higher in males than females ( “male mutation bias” ) [4]–[8] . Because the X chromosome is transmitted through females 2/3 of the time , the population mutation rate is lower for the X chromosome than the autosomes in species with male mutation bias . This downward biased mutation rate of the X chromosome in some mammalian lineages could therefore be responsible for the lineage-specificity of the faster-X effect in mammals [107] , [108] . Fourth , if the faster-X evolution of gene expression is driven by adaptive substitutions , as we propose , it is likely to be sensitive to [9] , [10] . In small populations a larger fraction of mutations will be effectively neutral [109] , which will decrease the number of beneficial mutations fixed by positive selection . The higher of Drosophila relative to mammals may therefore be more permissive of adaptive faster-X evolution [10] . In summary , we conclude that the difference in the extent of the faster-X evolution of gene expression between Drosophila and mammals could be a result of the unique mechanism of dosage compensation in Drosophila , the pervasiveness of male mutation bias , and/or the differences in between taxa . Determining which factor is most important will require additional theoretical and empirical work to identify the key determinants of gene expression evolution , the nature of selection on expression , and the effects of gene dosage on the dominance of fitness effects .
We obtained microarray measurements of expression from whole fly or head from previously published results [50]–[52] . We calculated the expression level of each gene by first taking the median signal across all probes for each gene within each replicate , and then calculating the median for each gene across all replicates . As an alternative approach , we used expression levels estimated in the LIMMA package of Bioconductor [110] , as described previously [52] . We tested for significant differences in expression between males and females ( i . e . , sex-biased expression ) using moderated t-tests implemented in the LIMMA package with the empirical Bayes function to pool sample variances toward a common value [110] , as described previously [52] . We corrected for multiple tests using a FDR [111] of 0 . 05 when only sex-biased expression in D . melanogaster was considered , and with a FDR of 0 . 20 when sex-biased expression in all species was considered . Genes with significantly higher expression in males were classified as having male-biased expression , and those with higher expression in females as female-biased . RNA-seq data collected from whole D . melanogaster , D . pseudoobscura , and D . mojavensis males and female or D . melanogaster , D . pseudoobscura , D . willistoni , and D . mojavensis heads were obtained from previously published results [52] . Reads longer than 36 bases ( bp ) were trimmed to 36 bp so that all reads were the same length , and reads were then mapped to the transcriptome using BWA [112] . Any read mapping to multiple locations in the genome was discarded , and genes with fewer than 50 mapped reads were excluded from the subsequent analysis . The expression level of each gene was estimated as the number of unique reads mapping to the gene standardized by the total number of mapped reads and the transcript length . We extracted only those genes present as 1∶1 orthologs on the same chromosome arm in all species under consideration , and we quantile normalized the expression levels so that they are identically distributed across all species . We next calculated SpearmanÕs between all pairwise comparisons of expression levels from the species under consideration . This was repeated for each chromosome arm . Confidence intervals ( CIs ) of were estimated by bootstrapping the data 1000 times in the R statistical computing environment [113] . We also calculated correlations of expression between sexes within each species . Microarray measurements of expression were obtained for 14 adult D . melanogaster tissues from FlyAtlas [58] , and the expression breadth was determined for each gene as described previously [19] . Briefly , we calculated [114] , a metric that ranges between 0 ( for broadly expressed genes ) to 1 ( for narrowly expressed genes ) . Genes were said to be narrowly expressed in a tissue if , and genes with were classified as broadly expressed . We used as an estimate of the pairwise divergence in expression between D . melanogaster , D . yakuba , D . ananassae , D . pseudoobscura , D . mojavensis , and D . virilis . We then reconstructed the phylogenetic relationships using the method of Fitch and Margoliash [115] implemented in the PHYLIP software package [116] . Bootstrap support for phylogenetic nodes was estimated by resampling the 1∶1∶1∶1∶1∶1 orthologs 1000 times . Branch lengths were estimated using the method of Fitch and Margoliash [115] with a fixed tree topology implemented in the PHYLIP software package [116] . CIs of the branch lengths were calculated by bootstrapping the data 1000 times . Bootstrap support and branch lengths were estimated for all 1∶1∶1∶1∶1∶1 orthologs , and this was repeated for genes on each chromosome arm separately . All bootstrapping was performed using the R statistical computing environment [113] . We obtained a list of genes bound by the DCC identified using ChIP-chip in three cell types: SL2 embryonic cell culture , larval wing imaginal disc cell culture , and late embryo [45] . A gene was said to be bound by the DCC if it is bound in at least one cell type . In addition , a second list of DCC bound genes was kindly provided by D . Bachtrog [48] . Our results are robust to the gene list used in our analysis . X-linked regions identified as HASs were obtained from previously published results [43] . We calculated the gene-wise expression divergence between 1∶1∶1 orthologs in D . melanogaster , D . yakuba , and D . ananassae as:where and are the expression levels of ortholog in species and , respectively . We then calculated Spearman's between , , and distance to the nearest HAS ( where = D . melanogaster ) . From these pairwise correlations , we calculated partial correlations to determine the direct relationship of each pair of values [117] . The CIs of the pairwise and partial correlations were estimated by bootstrapping the 1∶1∶1 orthologs 1000 times in the R statistical programming environment [113] . Kharchenko et al . [65] analyzed ChIP-chip results for multiple histone modifications and DNA binding proteins in two D . melanogaster cell lines ( S2 and BG3 ) , and they used a hidden Markov model to assign each region of the genome to one of nine chromatin states . States 1–5 are associated with transcriptionally active chromatin marks and states 6–9 with repressive marks . We used the overlap of these regions with annotated protein coding genes to determine whether each D . melanogaster gene is associated with a region of active or repressive chromatin marks . A gene was considered to be in active chromatin if of the gene body overlaps with regions identified as containing active marks , and , conversely , a gene was considered to be in repressive chromatin in of the gene body overlaps with regions identified as harboring repressive marks . We obtain similar results when use overlap cutoffs of and . We obtained estimates of broad sense heritability ( ) for D . melanogaster genes from a published analysis of microarray expression measurements in females and males from 40 inbred lines [75] . These estimates were calculated using an analysis of variance ( ANOVA ) , and we considered only genes in which the line term ( the estimate of ) was significant at a FDR of 0 . 05 . | As species diverge over evolutionary time , they accumulate differences in the sequences of their genes and how those genes are expressed . We show that gene expression changes accumulate faster for genes on the X chromosome than for genes on the other chromosomes ( autosomes ) in Drosophila ( the “faster-X” effect ) . The X chromosome is only found in a single copy in males , whereas the autosomes are found in two copies in both sexes . To compensate for the reduced dosage of X-linked genes in males , a molecular complex binds the Drosophila X chromosome to upregulate gene expression in males . We demonstrate that genes that escape this dosage compensation process have faster evolving expression levels . X-linked genes are inherited in a unique manner , and we hypothesize that this permits a faster rate of adaptive evolution , thereby driving the faster-X evolution of gene expression . We compare these observations with the recently described faster-X evolution of gene expression in mammals , and we explain how differences in dosage compensation , mutation rate , and population size could affect the extent of the faster-X effect . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genome",
"expression",
"analysis",
"genome",
"evolution",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"molecular",
"cell",
"biology",
"chromatin",
"chromosome",
"biology",
"gene",
"expression",
"comparative",
"genomics",
"biology",
"evolutionary",
"genetics",
"microarrays",
"cell",
"biology",
"genetics",
"genomics",
"evolutionary",
"biology",
"genomic",
"evolution",
"computational",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | Faster-X Evolution of Gene Expression in Drosophila |
Natural populations exhibit a great deal of interindividual genetic variation in the response to toxins , exemplified by the variable clinical efficacy of pharmaceutical drugs in humans , and the evolution of pesticide resistant insects . Such variation can result from several phenomena , including variable metabolic detoxification of the xenobiotic , and differential sensitivity of the molecular target of the toxin . Our goal is to genetically dissect variation in the response to xenobiotics , and characterize naturally-segregating polymorphisms that modulate toxicity . Here , we use the Drosophila Synthetic Population Resource ( DSPR ) , a multiparent advanced intercross panel of recombinant inbred lines , to identify QTL ( Quantitative Trait Loci ) underlying xenobiotic resistance , and employ caffeine as a model toxic compound . Phenotyping over 1 , 700 genotypes led to the identification of ten QTL , each explaining 4 . 5–14 . 4% of the broad-sense heritability for caffeine resistance . Four QTL harbor members of the cytochrome P450 family of detoxification enzymes , which represent strong a priori candidate genes . The case is especially strong for Cyp12d1 , with multiple lines of evidence indicating the gene causally impacts caffeine resistance . Cyp12d1 is implicated by QTL mapped in both panels of DSPR RILs , is significantly upregulated in the presence of caffeine , and RNAi knockdown robustly decreases caffeine tolerance . Furthermore , copy number variation at Cyp12d1 is strongly associated with phenotype in the DSPR , with a trend in the same direction observed in the DGRP ( Drosophila Genetic Reference Panel ) . No additional plausible causative polymorphisms were observed in a full genomewide association study in the DGRP , or in analyses restricted to QTL regions mapped in the DSPR . Just as in human populations , replicating modest-effect , naturally-segregating causative variants in an association study framework in flies will likely require very large sample sizes .
Living organisms are subjected to a barrage of toxic compounds , or xenobiotics , in their environment and their diet . Animals are frequently exposed to toxins produced by potential prey and/or plant hosts as chemical defenses [1 , 2] , and are increasingly subject to pressures from human activity , such as pollution and the application of pesticides . Humans themselves are additionally exposed to an array of xenobiotics throughout their lives in the form of pharmaceuticals [3] . Given the evolutionary , agricultural , and medical relevance of the response and resistance to toxins , dissecting the genetic factors responsible for xenobiotic metabolism is essential . Some of the best understood cases of xenobiotic response mechanisms come from insect populations or species that are able to withstand pesticides or toxic host chemical defense compounds . In some cases , certain individuals are simply insensitive to the xenobiotic due to a change in the structure , and therefore function , of the target of the toxin . For example , the Monarch butterfly ( Danaus plexippus ) is resistant to cardenolides , a class of secondary metabolites toxic to most animals , produced by their milkweed host plant . This resistance is due to at least one amino acid change in the Monarch Na+ , K+-ATPase gene that prevents dietary cardenolides binding to the protein , making the protein insensitive to cardenolide inhibition [4 , 5] . In many other cases , prior to reaching its target , the toxin is metabolized into less harmful substances by the action of a sophisticated three step detoxification system [6 , 7] . In the first step , cytochrome P450 monooxygenases ( P450s ) act on the toxic compounds to decrease their toxicity . The products of these reactions subsequently become substrates for phase two enzymes , such as glutathione-S-transferases ( GSTs ) and UDP-glucuronosyltransferases ( UGTs ) , which add large , charged side groups onto substrate molecules making them easier to excrete . Finally , membrane-bound ATP-binding cassette ( ABC ) transporters remove conjugated products from the cell . The Tobacco hornworm ( Manduca sexta ) is a facultative tobacco specialist , and detoxifies ingested nicotine by inducing P450 enzymes [8 , 9] . In addition , as a result of a series of naturally-occurring gene duplication events and transposable element ( TE ) insertions , overexpression of the P450 Cyp6g1 is primarily responsible for resistance to the insecticide DDT ( dichlorodiphenyltrichloroethane ) in Drosophila [10–12] . There are tens to hundreds of P450s , GSTs , UGTs , and ABC-transporters in eukaryotic genomes , and for most xenobiotics the precise series of enzymes responsible for their in vivo metabolism is unknown . We know there is substantial interindividual variation in the response to toxic compounds in animals such as Drosophila ( e . g . , [13–15] ) and in the response to drugs in humans ( e . g . , [16–18] ) , but we rarely know which genes in the detoxification cascade segregate for functional variation , or understand the mechanisms by which genetic variation at these loci influences phenotype . Here we characterize naturally-occurring alleles that contribute to resistance to caffeine in Drosophila . Caffeine is one of the most widely used psychoactive drugs in humans , acting as a stimulant , and has been employed as a convenient model xenobiotic in Drosophila research , where it is known to induce the expression of a number of P450 and GST genes in Drosophila S2 cells , larvae and adults [19–22] . However , simply because a gene responds to caffeine challenge by increasing its expression does not necessarily imply the gene segregates for alleles that give rise to variable drug response . A powerful method to identify natural alleles that contribute to variable xenobiotic response is to employ a large , stable set of highly-recombinant genotypes derived from a multiparental mapping population [23–29] . The Drosophila Synthetic Population Resource ( DSPR [30 , 31] ) consists of a set of >1 , 700 genotyped RILs ( Recombinant Inbred Lines ) derived from a pair of highly-recombinant synthetic populations each founded by eight strains . Our previous work has demonstrated the high statistical power and excellent mapping resolution of this community resource [30] , and we have succeeded in resolving strong candidate genes contributing to nicotine resistance [15] , and the response to chemotherapeutic drugs [14 , 32] . In this study we identify several QTL contributing to variation in Drosophila adult female lifespan during continuous exposure to 1% caffeine , a measure of caffeine resistance . Of the ten mapped QTL , five harbor strong a priori candidate detoxification genes ( e . g . , P450s ) . One of these genes—Cyp12d1—is implicated by our largest-effect QTL , and segregates for copy number variation ( CNV ) that strongly correlates with phenotype in the DSPR . Statistically accounting for the effect of this variant in a genomewide scan eliminates the QTL harboring the locus . In addition , Cyp12d1 shows expression induction in response to caffeine , and RNAi knockdown of the gene significantly reduces resistance . A follow-up association scan using the Drosophila Genetic Reference Panel ( DGRP [33 , 34] ) reveals no significant associations after genomewide or QTL-specific multiple testing correction . Although not significant , the intermediate-frequency CNV at Cyp12d1 shows a trend in the expected direction in the DGRP; lines harboring the duplication have marginally higher resistance on average ( p = 0 . 065 ) . If the effect of the CNV is modest , and particularly if additional functional alleles contribute to the QTL implicating Cyp12d1 , the functional loci may be difficult to individually identify by association in a small sample of natural chromosomes .
We measured the lifespan of female flies on 1% caffeine-supplemented media for 1 , 714 RILs from the DSPR , and 165 inbred lines from the DGRP , testing 16 . 5 females per genotype on average . There is substantial variation in phenotype in both mapping populations ( S1 and S2 Figs ) . The average RIL mean phenotype in the pA DSPR panel is 37 . 3 hours ( SD = 17 . 82 ) , and in the pB panel is 35 . 5 hours ( SD = 16 . 87 ) . The average line mean phenotype in the DGRP is higher at 59 . 8 hours ( SD = 20 . 80 ) , although the range of the genotype means in the two mapping populations is similar ( DSPR range = 6 . 8–119 . 9 hours , DGRP range = 10 . 9–112 . 7 hours ) . The broad-sense heritability of our measure of caffeine resistance is just over 50% in all populations ( DSPR pA and pB = 0 . 53 , DGRP = 0 . 51 ) . Because we employed line means for mapping , thus reducing environmental variance , the broad-sense heritability of the mean measure of caffeine resistance is around 94% ( DSPR pA and pB = 0 . 95 , DGRP = 0 . 94 ) . We note that the contribution of variation in lifespan under control , drug-free conditions likely plays a minimal role in our measure of caffeine resistance . Ivanov et al . [35] measured the lifespan of virgin females from the DGRP , and for the 155 lines common to both studies there is no statistically-significant correlation between lifespan and caffeine resistance line means ( Pearson's r = 0 . 09 , p = 0 . 27 ) . Using the same approach we have used previously [15 , 31] , we mapped QTL for caffeine resistance separately in both pA and pB DSPR mapping populations , identifying ten QTL contributing to resistance ( Fig 1 and Tables 1 and S1 ) . Three QTL are common to both populations ( i . e . , the 2-LOD support intervals overlap ) , one is unique to pA , and six are unique to pB . Given that the seven unique QTL generally explain smaller fractions of the heritability than the three common QTL ( Table 1 ) , it is possible we simply had insufficient power to detect them in the other panel of RILs . However , we cannot discount the possibility that alleles underlying these panel-specific QTL are private to a given panel . Under the assumption the QTL mapped are independent , and act additively , the total variance explained by all mapped QTL is simply the sum of the individual estimated heritability values . On this basis , in the pA population the four mapped QTL explain 31 . 9% of the broad-sense heritability for the trait , while in pB the nine mapped QTL explain 50 . 2% of the heritability . Eight of the ten QTL map away from centromeres ( Fig 1 ) , and are mapped to relatively short intervals of 260–750 Kb that include modest numbers of protein-coding genes ( mean = 64 . 8 , range = 26–122; S2 Table ) . Q1 and Q8 are close to the chromosome 2 and chromosome 3 centromeres , respectively , and map to much broader genomic intervals , suggesting it may be more difficult to resolve the underlying causative loci . In terms of a priori candidate genes—genes one might anticipate being involved in the differential metabolism of xenobiotics—four QTL intervals contain one or more P450 genes ( Q1 , Q2 , Q3 , and Q9 ) , one contains a carboxylesterase ( Q8 ) , and one contains two ABC transporters ( Q1 , although these two genes are only within the Q1 interval implicated in pA , not also the interval implicated in pB ) . None contain GSTs or UGTs . If the locus responsible for each of the three shared QTL ( Q1 , Q2 , and Q3 ) is the same in both panels , the causative gene should be present at a position implicated by both the pA and pB QTL LOD support intervals . For Q1 the area implicated by both mapped intervals is 500 kb , about half the width of the panel-specific intervals ( Table 1 ) , and harbors just 48 protein-coding genes . For Q2 the area implicated by both QTL is the same as that implicated by pA alone , containing 36 genes ( Table 1 ) , and for Q3 the area implicated by both QTL is just 60 Kb and contains ten genes . Each of these three small sets of genes suggested by the intersection of the pA and pB QTL support intervals contains a P450 gene ( Q1 harbors Cyp310a1 , Q2 harbors Cyp12d1 , and Q3 harbors Cyp301a1 ) . These are strong a priori candidates to contribute to variation in resistance to xenobiotics . For each QTL we estimated the phenotypic effects associated with each founder haplotype ( Figs 2 and S3 ) . In many cases it is difficult to clearly discriminate a pattern indicating that just two functional alleles , i . e . , "high" and "low" resistance alleles , segregate at mapped loci . It is possible these QTL represent loci segregating for a series of alleles , as we have suggested previously for toxicity and expression QTL mapped in the DSPR [32 , 36] , and others have observed for QTL mapped in other populations derived from multiple founders [26 , 37] . Nevertheless , given we are mapping QTL to intervals ~500-kb , and because multiple very-closely linked QTL would also be expected to yield non-biallelic patterns of founder effects , we cannot be confident that QTL are truly multi-allelic . Copy number variants ( CNVs ) are strong candidates to contribute to phenotypic variation [38–40] . One of the best described examples of a functional CNV is at the P450 gene Cyp6g1 in D . melanogaster , where a series of structural alleles confer varying levels of resistance to the pesticide DDT [10–12] . Previous work has observed variation in copy number at the Cyp12d1 locus—under our QTL Q2—in natural populations [41–43] , with the D . melanogaster reference strain possessing two copies with nearly identical gene sequences , Cyp12d1-d and Cyp12d1-p [44 , 45] . Using PCR assays we found that DSPR founders A7 , B1 , B5 , and B7 also possess these two gene copies , while the other 11 founders are single copy . Founder haplotypes at Q2 that harbor two copies of Cyp12d1 often have higher caffeine resistance than single-copy founders ( Fig 2 ) , although the founder with the highest mean resistance ( B4 ) only harbors one copy of Cyp12d1 . Considering individual RIL phenotypes , in both populations RILs with two copies of Cyp12d1 survive significantly longer on caffeine than RILs with only one copy of this gene ( Welch's t-test , p < 0 . 0001 in each population; S4 Fig ) . To explore whether the CNV can account for the mapped QTL Q2 we rescanned the genome after statistically controlling for the CNV status of each RIL . While most other QTL are still present , after adjusting for the effect of the Cyp12d1 CNV the signal at Q2 is entirely removed from both DSPR panels ( S1 Table and S5 Fig ) . Thus , it is likely that the CNV at Cyp12d1 , and/or a nearby variant or variants in strong LD with this event , plays a role in variable caffeine resistance in the DSPR . Previous work on the genetic basis of xenobiotic resistance has made extensive use of genomewide expression studies to investigate both the innate differences between strains susceptible and resistant to a given compound , and any changes in gene expression that occur in an organism following exposure to the toxic substance . We took pA RILs with either very high or very low caffeine resistance , generated adult females heterozygous between lines of similar phenotype , and exposed groups of these flies both caffeine-supplemented and control media for short periods . Using heterozygous animals minimizes inbreeding effects that may contribute to gene expression , but does make the assumption that heterozygous progeny are phenotypically similar to their parents . Following RNA extraction we pooled samples within phenotypic class ( high or low resistance ) and treatment , and carried out RNAseq , generating genomewide expression measures for four conditions , and leading to four contrasts of interest ( Low-control versus High-control , Low-caffeine versus High-caffeine , Low-control versus Low-caffeine , High-control versus High-caffeine ) . A total of 242 genes were differentially-expressed , i . e . , survived a genomewide per-contrast FDR threshold of 5% in at least one contrast ( S3 Table ) . At least 10% of these genes are named members of recognized detoxification families , including P450s ( 16 genes ) , GSTs ( 5 genes ) , UGT genes ( 3 genes ) , and ABC transporters ( 5 genes ) . The first two contrasts allow identification of genes having expression differences between genotypes with different phenotypes; 34 and 37 genes significantly change in expression between Low and High resistance genotypes on control food and caffeine-supplemented food , respectively . Eleven of these genes are shared between contrasts , and for all eleven the direction of the expression change is preserved over contrasts . The second two contrasts allow us to identify genes that change in expression on exposure to caffeine; in the Low ( High ) resistance genotype 178 ( 125 ) genes are differentially-expressed between control and caffeine food . One-hundred and one of these genes are shared between contrasts , and again the direction of the expression changes are consistent between genotypes; caffeine either induces expression in both genotypes or represses expression in both . Willoughby et al . [22] previously used a detoxification gene-specific microarray to identify 16 genes ( 11 P450 genes , 5 GSTs ) upregulated in D . melanogaster third-instar larvae on exposure to caffeine . Ten of these 16 are induced in response to caffeine in both High and Low DSPR resistance genotypes , validating our RNAseq , and suggesting adults and larvae may respond to caffeine challenge via similar mechanisms . Since our RNAseq study employed pA-derived genotypes , expression changes at genes within QTL intervals mapped in the pA population ( Q1 , Q2 , Q3 , and Q9 ) are of particular interest . Collectively , seven genes under these QTL show a change in expression that survives a genomewide multiple-testing threshold ( 5% FDR ) in at least one experimental contrast ( Table 2 ) . Four of these genes are annotated , but un-named genes with limited functional information in FlyBase [45] . One additional gene , Cpr49Ae , under Q3 produces a constituent of chitin , and is expressed at higher levels in genotypes that are more resistant to caffeine ( Table 2 ) . As a class , cuticle genes were found to be enriched among the set of genes overexpressed in a D . melanogaster strain resistant to α-amanitin , a toxin produced by a number of species of mushroom [46] . Higher expression of Cpr48Ae in caffeine resistant genotypes could plausibly be responsible for providing additional surface protection against xenobiotic exposure . Nonetheless , it is worth noting that Cpr48Ae is implicated only by Q3 mapped in pA , and is not present within the interval mapped in pB . The strongest expression candidates from this RNAseq screen are P450 genes , Cyp12d1 under Q2 , and Cyp6d5 , one of the two P450s under Q9 ( Table 2 ) . As mentioned previously Cyp12d1 exists in the D . melanogaster reference strain as a tandemly-duplicated pair of genes ( Cyp12d1-d and Cyp12d1-p ) , differing by just 3/1 , 563 coding sequence nucleotides . Recognizing the difficulty distinguishing reads from each gene , we eliminated the annotation for Cyp12d1-d during RNAseq analysis , allowing reads from both gene copies to pile up on Cyp12d1-p , and report these values in Table 2 . Results are essentially identical regardless of which gene copy is masked for analysis . Cyp12d1 shows strong , significant expression induction on caffeine exposure in both High and Low resistance genotypes , as has been shown previously [22 , 47] . In addition , Cyp12d1 gene expression is significantly higher for High resistance genotypes under both control and caffeine treatments . We note that at least some of the difference between Low and High genotypes is plausibly explained by the known CNV at this gene; Of the 18 RILs comprising each genotype pool , 17/18 Low genotypes are single copy for Cyp12d1 , while 14/18 High genotypes have two copies . Array-based female head-specific expression data derived from the progeny of crosses between DSPR RILs [36] confirms a significant effect of the Cyp12d1 CNV on expression of this gene ( S6 Fig ) . Cyp12d1 is a clear candidate for the causal locus responsible for Q2 . Cyp6d5 ( Q9 ) shows highly-significant expression induction following caffeine exposure , but there is no significant difference in expression between Low and High genotypes . The second P450 under Q9 ( Cyp313a1 ) shows a reduction in expression in response to caffeine in the Low resistance genotypes at a nominal level ( p = 0 . 005 ) , which is not the pattern expected under the assumption that higher expression leads to greater resistance to the drug . Thus , Cyp6d5 is the stronger candidate to causally underlie Q9 . Neither of the P450s under Q1 or Q3 mapped in pA ( Cyp310a1 and Cyp301a1 , respectively ) show significant expression variation across samples even at a nominal , p < 0 . 05 level ( S1 Dataset ) . This result doesn't rule out the possibility the genes contribute to variable resistance to caffeine , but suggests any such role may not involve regulatory variation . To confirm effects of a subset of the P450 genes uncovered by our mapping and expression studies ( Cyp12d1-d and Cyp12d1-p , Cyp301a1 , and Cyp6d5 under Q2 , Q3 , and Q9 , respectively ) we carried out knock-down experiments using the Gal4-UAS RNAi system . When knocked down ubiquitously ( RNAi in all cells at all timepoints ) , all but one of the Gal4-UAS showed a significant reduction in caffeine resistance compared to controls ( Fig 3A and 3B ) , with the remaining genotype showing a minor reduction in resistance ( Fig 3B; transformant ID 21235 , Cyp12d1-p , Welch's t-test , p = 0 . 066 ) . In two cases the RNAi appears to interfere with some essential developmental process , since we struggled to generate individuals of the appropriate genotypes for assay ( Fig 3A; genes Cyp12d1-d , transformant ID 109248 , and Cyp301a1 ) . Since Cyp301a1 appears to be involved in cuticle development , and reduced expression of the gene results in cuticle malformation [48] , this may have played a role in our failure to generate large numbers of progeny . We also cannot rule out any minor developmental abnormalities in the other RNAi genotypes that may have led to reduced adult fitness , and a nonspecific reduction in caffeine resistance . To control for any such pleiotropic effects of the test genes during development , we subsequently used RU486-inducible Gal4 to drive RNAi in all cells in young adults ( Fig 3C and 3D ) . Over two trials ( differing only in the amount of time the test animals were fed RU486 prior to caffeine+RU486 exposure ) , the only RNAi knockdown to show a consistent , highly-significant reduction in caffeine resistance was for Cyp12d1-d ( Fig 3C and 3D ) . Since the effect of RU486 on expression knockdown is dose-dependent [49] , it is possible higher concentrations of RU486 could have replicated our findings using the ubiquitous Gal4 driver ( Fig 3A ) , and Cyp301a1 and Cyp6d5 remain excellent candidates to underlie QTL Q3 and Q9 . Critical to interpreting RNAi results with Cyp12d1-d and Cyp12d1-p is the fact that these genes appear to be the result of recent duplication [42] , and have nearly identical sequences [44 , 45]; The genes differ by 3/1 , 563 coding sequence nucleotides , leading to 3/521 amino acid differences . The hairpin sequences of the UAS-RNAi constructs targeting one member of the gene pair exactly , exhibit zero ( transformant ID 50507 ) , one ( transformant IDs 21235 , 109248 , 109256 ) , or two ( transformant ID 49269 ) nucleotide differences from the other member of the gene pair ( S2 Dataset ) . So it is highly likely that in each Cyp12d1 knockdown experiment we are reducing the expression of both genes , albeit perhaps to different degrees . The DSPR has excellent power to identify QTL contributing to trait variation [30] , but given the haplotype structure of the DSPR RILs the approach suffers from a lack of resolution in comparison with population-based association mapping . It is possible to carry out local , QTL-centric association scans in the DSPR by imputing SNP genotypes using the estimated mosaic founder haplotype structure of each RIL , but again LD ( linkage disequilibrium ) is such that large numbers of SNPs tag the exact same haplotype combinations , precluding localization of causative variants [15 , 32] . To attempt to identify candidate causative polymorphisms we used the DGRP [33 , 34] , a sample of fully-resequenced inbred lines derived from a single collection , to carry out association mapping . Following a GWAS ( Genomewide Association Study ) , testing 1 . 9 million polymorphic events for association with caffeine resistance , the site with the highest association statistic ( p = 1 . 7 × 10−6 ) was far from a genomewide Bonferroni significance threshold of p ~ 10−8 ( S4 Table ) . The lack of a strong association is unsurprising given the likelihood that causative variants explain relatively small fractions of the variance ( Table 1 ) , and the low power in a GWAS with < 200 lines to find sites with such effect sizes [50 , 51] . To avoid a harsh , genomewide significance threshold we instead restricted our focus to those variants present within QTL intervals mapped in the DSPR . This procedure is more appropriate given we are attempting to validate mapped QTL , rather than attempting de novo discovery in the DGRP . Nevertheless , no site reached a QTL-specific Bonferroni threshold , and only one site beneath a mapped QTL reached p < 10−5 ( one of the 24 , 782 tested variants within the large Q8 interval ) . Tens to hundreds of variants within QTL intervals have p-values that reach a nominal 5% threshold , but in each case the fraction of such variants is essentially what would be expected by chance . We additionally examined associations within 5-kb of the start and end of each of the six annotated P450 genes we have discussed previously ( Cyp12d1-d and Cyp12d1-p , Cyp301a1 , Cyp310a1 , Cyp313a1 , Cyp6d5 ) , but again no site in these regions survived a per-gene Bonferroni correction for multiple tests . In the DSPR we found evidence supporting the role of copy number variation at Cyp12d1 in conferring resistance to caffeine . Just as for the DSPR we genotyped lines from DGRP using PCR based assays [12 , 42] , and confirmed our data with bioinformatic calls based sequencing read-depth [41] . Using only those 151 DGRP lines for which we had caffeine phenotypes , where we were confident the lines were homozygous at the Cyp12d1 locus , and where PCR-based calls and bioinformatics calls were identical , the mean resistance of DGRP lines carrying one Cyp12d1 copy was 57 . 6 hours ( SD = 20 . 88 , N = 119 ) , and the mean resistance of lines with two copies was 65 . 4 hours ( SD = 20 . 67 , N = 32 ) . While there was a trend in the anticipated direction , the effect of the Cyp12d1 duplication on caffeine resistance is not significant at the 5% level in the DGRP ( Welch's t-test , p = 0 . 065; S4 Fig ) . Given power constraints in the DGRP [50] we may be unable to detect the CNV in this population , particularly if we have overestimated the heritability explained by Q2 ( Beavis effects [52 , 53] ) , or if the effect of Q2 in the DSPR is not entirely due to the CNV , and other variants at Cyp12d1 are additionally causative . However , we cannot rule out a scenario in which the Cyp12d1 CNV has no effect on caffeine resistance in the DGRP .
In this study we employed a large mapping population , the DSPR , to identify loci segregating for naturally-occurring alleles with variable effects on xenobiotic toxicity in D . melanogaster . Heritability of resistance to caffeine is around 50% in all mapping panels queried in this study , and we succeeded in mapping a number of QTL that collectively explain between a third and half of the broad-sense heritability . Aside from those QTL mapping to centromeric regions , we map QTL to genomic intervals containing fairly modest numbers of protein-coding genes ( 26–122 ) , and in those cases where a QTL is mapped to the same location in the pA and pB RIL panels , we can further refine the list of the most likely candidates to those genes implicated by both intervals ( i . e . , 48 , 36 and 10 genes for Q1 , Q2 , and Q3 , respectively ) . Dissection with this level of resolution is difficult with standard , 2-parental QTL mapping using a single generation of meiotic recombination ( i . e . , using an F2 or backcross mapping design ) , and can only reasonably be achieved for linkage mapping designs employing populations subjected to multiple rounds of meiotic recombination ( e . g . , [54] ) . Short QTL mapping windows facilitate the identification of true causative gene ( s ) , and in this study our ability to suggest plausible functional loci was additionally enhanced by the nature of our target phenotype . Genes involved in metabolic detoxification pathways represent strong a priori candidates to harbor segregating variation contributing to caffeine resistance . Indeed , four caffeine resistance QTL harbor P450 genes ( Q1—Cyp310a1 , Q2—Cyp12d1 , Q3—Cyp301a1 , Q9—Cyp6d5 and Cyp313a1 ) , many of which appear to be involved in detoxification [55] , and these present excellent targets for future research to identify the precise sequence variants impacting resistance to caffeine . Our mapping results , in combination with RNAseq and RNAi data , imply that a significant fraction of natural variation in resistance to caffeine is due to metabolic detoxification driven by changes in P450 genes ( see also [47] ) . The strongest candidates from our study are Cyp12d1 and Cyp6d5 . These loci are present within QTL intervals ( Cyp12d1 is additionally present within QTL mapped in both pA and pB mapping panels ) , show expression induction in the presence of caffeine ( Table 2 ) , and a reduction in resistance on RNAi gene expression knockdown ( Fig 3 ) . In addition , a number of previous studies have shown effects of these loci consistent with a role in xenobiotic metabolism: Both genes are expressed in the larval midgut [56] , a likely site of xenobiotic metabolism , both have been shown to be induced by caffeine and phenobarbital [22] , and silencing both genes alters downstream metabolism of caffeine [47] . We also show that the copy number variation segregating at Cyp12d1 [12 , 41 , 42] associates with our caffeine resistance phenotype in the DSPR RILs . Gene duplications are known to be important contributors to complex variation ( e . g . , [57] ) , and evidence from other studies demonstrates that structural variation at P450 genes can be associated with toxin resistance . For instance , an allelic series of CNV events and TE insertions at Cyp6g1 impacts DDT resistance in D . melanogaster [12] , although in this same study no association was seen between Cyp12d1 copy number and DDT resistance . In the DGRP , the Cyp12d1 CNV shows an effect on caffeine resistance in the same direction as in the DSPR , though this trend is not significant at the 5% level ( p = 0 . 065 ) . Heterogeneity in the genetic architecture of caffeine resistance across mapping panels could explain the disconnect among studies , with copy number at Cyp12d1 positively , and causally , associating with resistance in the DSPR but not in the DGRP . Alternatively , the CNV could be in strong LD with the true causative variant in the DSPR , whereas these events may not be in LD in the DGRP . Notably however we did not identify any other associations around Cyp12d1 in the DGRP that could represent this indirect association . Given the multiple lines of evidence supporting an effect of the Cyp12d1 locus , it remains highly likely that this gene is involved in caffeine resistance . Future targeted fine-mapping and/or functional genomics work will be required to localize the causative variant ( s ) in the Cyp12d1 region that lead to natural variation in caffeine resistance . Other P450s implicated ( Cyp301a1 , Cyp310a1 , and Cyp313a1 ) in our mapping did not show changes in expression , either between susceptible or resistant genotypes , or between control and caffeine-supplemented media ( S1 Dataset ) . This may imply that the effect of these genes on variable caffeine resistance is not regulatory in origin , and thus cannot be captured by RNAseq . Alternatively , overall expression levels may be too low—which is likely true for Cyp301a1 ( FPKM = 1 . 88–2 . 24 ) and Cyp310a1 ( FPKM = 0 . 52–0 . 67 ) —and any expression differences may be subtle , precluding the identification of significant expression changes without high levels of biological replication , or by carrying out RNAseq on the likely tissues in which the detoxification process takes place ( e . g . , [56] ) . However , without further validation it is not possible to exclude the possibility that these P450 genes are not the causative loci , and instead are merely closely linked to the true caffeine resistance genes . For the six QTL not containing P450s , all of which have modest effects and were mapped only in the pB panel of DSPR RILs , it is difficult to suggest likely causative genes solely from the lists of implicated loci . QTL Q8 does harbor a carboxylesterase , Est-Q , that is known to show expression in the larval midgut , a site of xenobiotic metabolism [56] . However , while this is a plausible candidate gene , because Q8 implicates a total of 398 genes , other evidence for a role in caffeine resistance would be desirable prior to any attempt to functionally validate this gene . Similarly , the Smc5 gene , shown to be required for resistance to caffeine exposure during development [58] is also present within Q8 , and remains a plausible candidate . The genes cnc , Hr96 , and Keap1 are known to regulate the transcriptional response to xenobiotic challenge in Drosophila [20 , 59] , and ERR is associated with the regulation of a number of P450 genes [60] . However , none of these genes are located under the caffeine resistance QTL we map here , suggesting that natural variation in the regulation of the overall transcriptional response to xenobiotics is not strongly involved in caffeine resistance , at least in our assay . Gustatory receptors have been shown to be involved in the avoidance of noxious substances , and Gr66a and Gr93a are required to avoid caffeine ingestion [61 , 62] . Neither of these genes are implicated by our mapped QTL , although other members of the family are present in mapped intervals ( Gr47b is present within the Q2 interval mapped in pB , and Gr77a is present under Q8 ) . The absence of clear candidate genes underlying mapped QTL is emblematic of the challenge facing most linkage mapping studies . Here , we used a combination of RNAseq and RNAi to validate the effects of certain P450 genes on caffeine resistance , and to refine our list of potential genes for those QTL without obvious a priori candidates . Moving from causative genomic regions to the true causative sites is an ongoing challenge for mapping studies . Combining multiple sources of data to provide several lines of evidence supporting candidate loci may prove generally valuable to resolve functional loci within mapped genomic windows , and data on the positions of QTL from this study represent the starting point for such exploration ( Tables 1 and S2 ) . A goal of our research is not only to identify the causative genes , but also to identify the causative sequence variants , and we employed the DGRP association mapping panel to this end [33 , 34] . Using a similar phenotyping strategy to that employed in the DSPR , and a straightforward GWAS analysis [33] , we found no genomewide significant associations after controlling for multiple tests . This result is likely explained by the low power of the DGRP on a genomewide scale: With 158 lines and a genomewide significance threshold of p < 10−8 , the power to find an intermediate-frequency causative polymorphism ( minor allele frequency , MAF = 0 . 4 ) that explains 10% of the broad-sense heritability is just 3% ( see [50] for description of power calculation ) . Due to the lack of power for de novo identification of associations in a small GWAS panel , we instead attempted to validate associations within intervals implicated by DSPR QTL , and find associations directly at the six P450 genes suggested by our study . However , even with a lower significance threshold ( approaching p < 2 × 10−4 for the gene-centric association scans ) and a concomitant increase in power ( 62% for a site at a frequency of 0 . 4 that explains 10% of the variation among lines ) , we found no QTL-specific variants , and no variants at the set of P450 genes that survive multiple test correction . Thus , power deficits in the DGRP may not entirely explain the lack of overlap between findings in the DSPR and DGRP . A lack of overlap between the DSPR and DGRP might be expected under a situation in which we have overestimated the heritability explained by QTL mapped in the DSPR . If true QTL effects are much smaller than we estimate ( Table 1 ) , power to validate such effects is in turn reduced in the DGRP . Beavis [52] showed that the effects of mapped QTL tended to be overestimated , with the greatest overestimation associated with experiments employing small numbers of samples . Given that the sample size for each DSPR panel was >800 genotypes , and our power to map QTL contributing 5% to variation is over 80% [30] , we anticipate that Beavis effects will not be large ( see [52] and Figure 15 . 8 in [63] ) . Nonetheless , work on the Beavis effect [52 , 53] has focused on two-way crosses involving a single-generation of meiotic recombination , and there has been no exploration of the expected magnitude of the Beavis effect in highly-recombinant , multiparental mapping populations . One possible reason for the failure to find associations in the DGRP within mapped QTL intervals could be that causative variants present in the DSPR are absent in the DGRP . Variants at intermediate frequency will typically be shared by both panels , and given sufficient power , should replicate . Rare variants will be shared less often , and will only replicate some of the time . Those SNPs present in two or more DSPR founders are found in the DGRP >91% of the time , while those SNPs private to a single DSPR founder are seen in the DGRP ~70% of the time ( see DSPR founder SNP data from FlyRILs . org and DGRP variant data from dgrp2 . gnets . ncsu . edu ) . In addition , founder means at QTL do not show a pattern consistent with a single founder having a private allele ( Figs 2 and S2 ) . Thus , there is limited evidence that the variants responsible for our QTL are unique to the DSPR . A related possibility is that causative variants have very different frequencies in the DSPR and DGRP . Population-based association mapping approaches struggle to identify low frequency causative variants [50 , 64 , 65] . Whereas provided the minor allele is captured among one of the founders , a multiparental mapping population has good power to find a low frequency causative variant [30] . Thus , DSPR QTL generated by rare causative loci may not replicate in the DGRP . One piece of evidence against this possibility is the observation that Q1 , Q2 , and Q3 were all mapped in both DSPR panels . This implies these QTL are not due to rare causative variants present solely in one panel of the DSPR , and suggests the causative variants underlying these QTL are likely present at intermediate-frequency in the DGRP . For instance , the frequency of the duplication allele at the plausible causative CNV at the Cyp12d1 gene within Q2 is 12 . 5% among DSPR pA founders , 37 . 5% among pB founders , and 21 . 2% among the DGRP lines . Since we do not implicate a causative polymorphism for all other QTL we map in the DSPR it is not possible to determine whether the ( unknown ) polymorphisms responsible have similar frequencies in the DGRP . A final possibility for our failure to replicate our QTL mapping results in the DGRP is suggested by the observation from a number of different multiparental mapping studies that QTL frequently segregate for multiple alleles [26 , 36 , 37] . Similarly , the strain effects for some of the QTL we map here do not show a clear pattern indicative of biallelic QTL simply segregating for a single "high" and "low" allele ( Figs 2 and S3 ) . A multiparental panel tests the effect of a local haplotype on phenotype , integrating over any causative variants in a short genomic window . So if QTL mapped in the DSPR are routinely generated by the combined action of multiple variants , each with very small effects , the power to identify any of these individually in the DGRP will be extremely limited . Detection of very subtle effect , intermediate-frequency causative variants , contributing <5% to the natural variation will require massive sample sizes as is common in the human genetics community ( e . g . , [66] ) . If instead the multiple causative variants present at genes contributing to polygenic trait variation are all rare , as is the case for genes leading to single-gene , Mendelian disorders , typical GWAS analytical approaches are very poorly powered to detect such genes [65] . In summary , we have used the DSPR to map QTL that collectively explain a large fraction of the natural variation for caffeine resistance in flies . Four of these QTL map to genomic regions containing members of the large family of cytochrome P450 detoxification enzymes . These loci present excellent candidates to harbor variant ( s ) contributing to phenotypic variation , and for three of these loci—Cyp6d5 , Cyp12d1 , and Cyp301a1—we validate their effects on caffeine resistance via RNAseq and/or RNAi , suggesting an important role for variation in metabolic detoxification in the control of caffeine resistance . In addition , our data suggests that the number of copies of Cyp12d1 is a strong determinant of caffeine resistance . We were unable to resolve any additional candidate causative sequence changes by association in a small panel of natural chromosomes , likely due in part to concerns related to power , and potentially as a result of our QTL being generated by the combined action of multiple , small-effect causative alleles . Nonetheless , within each interval implicated by a QTL , tens to hundreds of variants show associations at nominally-significant levels ( e . g . , p < 0 . 05 ) . If some of these modest-effect associations could be validated in an independent , large association mapping panel this would provide strong evidence for their causative role in affecting phenotypic variation . Ultimately , for strong candidate regions , in the future it will be possible to employ CRISPR-Cas9 genome editing to directly compare alleles at candidate loci in an otherwise identical genetic background to localize regions , and ultimately nucleotide variants directly contributing to phenotypic variance .
To map QTL for caffeine resistance we used RILs from the DSPR ( FlyRILs . org ) . Full details of the mapping panel , its development , the analytical methods employed to map QTL , and simulations supporting the power and resolution of the approach have been described previously [30 , 31] . Briefly , the DSPR consists of two panels of >800 RILs ( pA and pB ) . Each set is descended from an advanced generation intercross among eight highly-inbred founder lines , seven of which are specific to a single panel ( A1-A7 or B1-B7 ) , with one is common to both panels ( AB8 ) . Each intercross population was maintained as a pair of independent replicate subpopulations ( pA1 , pA2 , pB1 , pB2 ) at large population size for 50 generations , after which RILs were generated via 25 generations of full sibling mating . The 15 founder lines have been resequenced , and the RILs genotyped via Restriction site Associated DNA ( RAD ) tags . These data , along with a hidden Markov model ( HMM ) , allowed the underlying mosaic haplotype structure of each RIL to be determined at a fine scale . To carry out the GWAS we used strains from the DGRP , a set of inbred lines descended from gravid females caught at a single collection location ( Raleigh , NC , USA ) , and inbred via 20 generations of full sibling mating . DGRP lines were purchased from the Bloomington Drosophila Stock Center ( flystocks . bio . indiana . edu/ ) . The lines have been fully resequenced , and molecular variation in the panel has been described [33 , 34] , along with the computational pipeline used to make variant calls [33 , 67] . Flies from each genotype were allowed to lay eggs for up to 48 hours on standard cornmeal-yeast-molasses food in regular narrow , polystyrene fly vials , and adults were cleared to maintain a roughly equal density of eggs across experimental vials . Eight days after the start of egg laying , any newly emerged adults were cleared from the experimental vials , and two days later 20–25 , likely mated female flies were harvested from each vial under CO2 anesthesia . These flies were kept as groups on standard media and allowed to recover for 24 hours prior to caffeine challenge . Our measure of caffeine resistance for a genotype is taken as mean lifespan on media supplemented with 1% caffeine ( C0750 , Sigma ) , the same concentration as employed in a previous study of caffeine resistance variation in D . melanogaster [13] . To increase throughput , and provide for relatively automated data collection , we made use of the Drosophila Activity Monitoring System ( DAM2 , TriKinetics , Inc . ) . Twenty-four hours before the start of the assay we made cornmeal-yeast-dextrose media , adding caffeine at ~50°C to minimize any heat-induced degradation . We replaced the molasses from our typical culture medium with dextrose to achieve better batch-to-batch reproducibility since small volumes of molasses proved difficult to dispense accurately . The molten media was poured into 100mm diameter petri dishes and allowed to set for 2–3 hours . We pushed the ends of large bundles of polycarbonate activity monitor tubes ( 5mm diameter × 65mm length ) into the media , filling each tube to a height of ~10mm , cleaned the tubes to remove excess media , and sealed the food plug in each tube with molten paraffin wax . On the day of the assay we aspirated individual , 1–3 day old experimental females into monitor tubes without anesthesia , capped the tubes with small foam plugs cut from Droso-Plugs ( Genesee Scientific Corporation ) , and inserted tubes into monitors . All experimental individuals were reared and assayed at 25°C and 50% relative humidity on a 12 hour light/12 hour dark cycle . The time when experimental flies were introduced to caffeine food in monitor tubes was recorded using a custom Excel macro ( Microsoft Corporation ) , facilitated by arbitrary barcodes on experimental vials and monitors . Activity data was recorded every minute for six days for each experimental block using the DAMSystem3 data collection software ( TriKinetics , Inc . ) , and any permanent cessation of activity was interpreted to be a result of fly death . The data were filtered to remove animals with very low activity levels from the start of the assay , presumably due to damage during aspiration into monitor tubes , and the time until death on caffeine food for each fly was extracted using a series of custom scripts written in R ( www . R-project . org ) . The above assay was carried out on 1 , 714 DSPR RILs ( 853 pA and 861 pB lines ) and 165 DGRP lines across 37 experimental blocks . For 1 , 623 DSPR and 159 DGRP lines we generated a single experimental vial of flies , and tested 16 females from that vial in a single experimental assay block . For a small minority of lines we independently tested 16 flies in each of two blocks , and the correlation between means calculated for each block is high ( Pearson's r = 0 . 89 , p < 10−15 ) , giving us confidence in our phenotypic measurements . Over all mapping panel genotypes we scored an average of 16 . 5 females per genotype ( SD = 3 . 62 , range = 9–32 ) . Raw phenotype data is presented in S5 Table . Broad-sense heritability of our measure of caffeine resistance ( the time until death of an adult female fly on caffeine-supplemented media ) was estimated by calculating the genetic and phenotypic variance components from a linear mixed model using the lme and VarCorr functions in the nlme package in R [15 , 68] . The estimated genetic variance component divided by the total variance of line means yielded the heritability of the line means that were used for mapping . Heritabilities were separately calculated for both panels of DSPR RILs and the DGRP . Identification of QTL in the DSPR has been described previously [30 , 31] . In essence , for each region in each RIL an HMM is used to assign a probability that the genotype is one of 36 possible homo- or heterozygous founder genotype combinations . Subsequently we assume that the very small number of heterozygous states we find are intermediate between the respective pair of homozygous states , generating eight additive probabilities per position . We then regress the mean line phenotype on these eight probabilities , analyzing the pA and pB panels separately . We did not include a subpopulation covariate since there was no significant difference between the two pA or pB subpopulations in mean RIL caffeine resistance ( pA , Welch's t = 1 . 8 , p = 0 . 06; pB , Welch's t = 1 . 9 , p = 0 . 06 ) . Following Churchill and Doerge [69] we use 1 , 000 permutations of the data to find appropriate genomewide significance thresholds for QTL identification . We used 2-LOD support intervals to estimate the true positions of causative loci , which simulations suggest correspond to 93–94% confidence intervals for our experimental design , sample size , and mapped QTL effect sizes [30] . We note that there is a clear visual , positive correlation between line mean phenotypes and within-line variance ( S1 Fig ) . This pattern is eliminated following a square root transformation of the line means . However , such a transformation does not change the genomewide likelihood profiles , or alter the QTL we map , and we only report analysis based on the raw , untransformed dataset here . To account for any effect of the CNV at Cyp12d1 we performed a similar genomewide scan for QTL as described above , but additionally included a covariate describing the CNV status of each RIL . For this analysis we only used those RILs for which we could be confident of the CNV status , i . e . , where the founder allele at the gene was estimated with at least 95% certainty . This resulted in a reduction in the sample size for mapping from 853 to 803 RILs in pA , and from 861 to 812 RILs in pB . Phenotype means for 158 DGRP strains assayed were provided to the Mackay lab's web-based analytical engine ( dgrp2 . gnets . ncsu . edu ) to carry out a GWAS using nearly 2 million common ( minor allele frequency > 0 . 05 ) SNP and non-SNP variants identified in the panel . Prior to the association tests the line means are adjusted to account for any effect of the Wolbachia bacteria that infects around half of the lines , and to account for any effects of five major chromosomal inversions: In ( 2L ) t , In ( 2R ) NS , In ( 3R ) P , In ( 3R ) K , and In ( 3R ) Mo . After which association tests are carried out for each variant using a linear mixed model accounting for variation in relatedness across the lines . See Huang et al . [33] for full details . We selected two sets of pA RILs that had either very low ( in the bottom 2 . 3% of lines ) or very high ( in the top 4 . 2% of lines ) caffeine resistance , and crossed RILs in pairs within each phenotypic class . From each cross , we collected 2–3 day old adult female trans-heterozygous progeny under CO2 anesthesia , allowed the flies to recover for 24 hours , then exposed flies from each genotype either to control food or to 1% caffeine food for 4 hours . No flies died during this short exposure . Flies from each genotype and treatment were then snap-frozen in liquid nitrogen , and total RNA was isolated from each sample using TRIzol reagent ( 15596–018 , Life Technologies ) . Equal amounts of RNA from each sample were then pooled within each phenotype/treatment combination: susceptible genotypes/control food , susceptible genotypes/caffeine food , resistant genotypes/control food , and resistant genotypes/caffeine food . Ultimately , each of these four pooled samples contained RNA from ~100 experimental females , and haplotypes from 18 pA RILs . The four pooled samples were cleaned through RNeasy Mini columns ( 74104 , Qiagen ) following the manufacturer's protocols , used to generate Illumina TruSeq RNAseq libraries , and sequenced over four lanes of an Illumina HiSeq 2500 instrument to generate single-end 50bp reads ( Genome Sequencing Facility , KU Medical Center ) . Raw sequencing reads were deposited in the Sequence Read Archive ( SRA ) under project accession number SRP051835 . We trimmed raw sequencing reads with sickle ( version 1 . 200 , github . com/najoshi/sickle ) , and used TopHat ( version 2 . 0 . 9 , tophat . cbcb . umd . edu; [70 , 71] ) to assemble reads from each sample to the D . melanogaster reference genome ( NCBI build 5 . 3 , tophat . cbcb . umd . edu/igenomes . shtml ) . In order to get an accurate expression measure for the Cyp12d1 gene , a gene that exists as two , nearly identical copies in the reference ( Cyp12d1-d and Cyp12d1-p ) and is subject to copy number variation in the DSPR founders , we modified the reference annotation to eliminate the Cyp12d1-d gene . Following TopHat we had 135–169 million quality-trimmed reads per sample , and 89 . 1–90 . 4% of these aligned to the reference genome . Subsequently we used Cuffdiff ( version 2 . 1 . 1 , cufflinks . cbcb . umd . edu; [72 , 73] ) to identify differentially expressed genes ( see S1 Text for details of the code and settings ) . For each comparison between samples we considered only those tests that were successfully executed ( 'status' column in "gene_exp . diff" Cuffdiff output file has 'OK' flag ) , and unless otherwise stated only considered genes to be significantly differentially expressed if they survived a Benjamini-Hochberg correction for multiple-testing ( q < 0 . 05 ) . To functionally test four P450 genes implicated in our mapping and RNAseq experiments we employed RNAi using the bipartite Gal4-UAS system , and UAS strains from the Vienna Drosophila Resource Center ( VDRC [74] ) . We used strains harboring phiC31-based UAS transgenes , or "KK" lines ( 107641 , Cyp6d5; 109248 , Cyp12d1-d; 109256 , Cyp12d1-p; 109771 , Cyp301a1 ) , and strains harboring P-element transgenes , or "GD" lines ( 50507 , Cyp12d1-d; 21235 and 49269 , Cyp12d1-p ) , along with the respective background control lines ( 60100 , KK landing site control strain; 60000 , w1118 host strain for GD library ) . It has been reported that the 60100 control strain for the KK library has two landing sites [75] . We used the primers described in Green et al . [75] to determine which of the landing sites are occupied in the KK-UAS strains we employed: Strain 109248 has the vector integrated into the annotated target site only , strains 107641 and 109256 have vectors in the newly-identified , previously unannotated site , and 109771 has vectors in both landing sites . These differences in transgene position and number make it more difficult to directly compare effects across genotypes . We tested the phenotypic effects of RNAi-based expression knockdown using two Gal4 drivers . First , we crossed males from a ubiquitous Gal4 driver under the control of the Actin 5C promoter ( Bloomington Drosophila Stock Center , BDSC number 25374 ) to females of each UAS or control strain , testing female progeny from these crosses in our standard caffeine resistance assay . Second , we used an RU486-inducible "GeneSwitch" Gal4 driver [49] also under the control of the Act5C promoter ( BDSC number 9431 ) . Using this strain we again ubiquitously expressed Gal4 in all cells , but exerted some temporal control by driving Gal4 only in young adults . Males from the RU486-inducible Gal4 strain were crossed to females of the KK-UAS and 60100 control strains , and the resulting female progeny fed on media containing RU486 ( M8046 , Sigma ) at 160μg/ml for 24 hours ( trial 1 ) or 48 hours ( trial 2 ) prior to the assay . The resistance assay was conducted as described , except that in addition to 1% caffeine the test media contained 160μg/ml RU486 . We used primer sets provided in Schmidt et al . [12] and McDonnell et al . [42] to call copy number variation at the Cyp12d1-d/Cyp12d1-p locus in lines used in our experiments , and for the DGRP validated our genotype calls against bioinformatic , sequencing read-depth calls from Good et al . [41] . CNV genotype calls for the DSPR founders and the DGRP are presented in S3 Dataset , while inferred CNV calls in the DSPR RILs—based on the mosaic haplotype structure of each RIL—are presented in S6 Table . All statistics were carried out using core , or custom written , routines using the R statistical programming language ( www . R-project . org ) . | A striking feature of biology is that within populations there is substantial interindividual phenotypic variation for traits of biomedical significance . Some fraction of this variation is due to environmental factors , but for many traits segregating genetic differences contribute significantly to phenotypic variation . Elucidating those causative sequence changes that result in complex trait variation is of central importance to biology , and requires the coordinated use of multiple approaches . Here we employ a multi-level strategy to dissect genetic variation in caffeine resistance in Drosophila melanogaster , leveraging powerful genetic screening in a multiparental mapping panel , a genomewide association study , high-throughput RNA sequencing , and gene knockdowns using RNA interference . We identify several short genomic regions that collectively explain a substantial portion of the heritable variation for caffeine resistance , and find that several of these regions harbor members of known detoxification enzyme families . One such gene—Cyp12d1—shows increased expression on exposure to caffeine , and experimentally reducing gene expression leads to a reduction in caffeine resistance . We additionally show that variation in the number of copies of Cyp12d1 is positively associated with resistance . These compelling lines of evidence imply that structural variation at this gene causally contributes to xenobiotic resistance in Drosophila . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Identifying Loci Contributing to Natural Variation in Xenobiotic Resistance in Drosophila |
For directional movement , eukaryotic cells depend on the proper organization of their actin cytoskeleton . This engine of motility is made up of highly dynamic nonequilibrium actin structures such as flashes , oscillations , and traveling waves . In Dictyostelium , oscillatory actin foci interact with signals such as Ras and phosphatidylinositol 3 , 4 , 5-trisphosphate ( PIP3 ) to form protrusions . However , how signaling cues tame actin dynamics to produce a pseudopod and guide cellular motility is a critical open question in eukaryotic chemotaxis . Here , we demonstrate that the strength of coupling between individual actin oscillators controls cell polarization and directional movement . We implement an inducible sequestration system to inactivate the heterotrimeric G protein subunit Gβ and find that this acute perturbation triggers persistent , high-amplitude cortical oscillations of F-actin . Actin oscillators that are normally weakly coupled to one another in wild-type cells become strongly synchronized following acute inactivation of Gβ . This global coupling impairs sensing of internal cues during spontaneous polarization and sensing of external cues during directional motility . A simple mathematical model of coupled actin oscillators reveals the importance of appropriate coupling strength for chemotaxis: moderate coupling can increase sensitivity to noisy inputs . Taken together , our data suggest that Gβ regulates the strength of coupling between actin oscillators for efficient polarity and directional migration . As these observations are only possible following acute inhibition of Gβ and are masked by slow compensation in genetic knockouts , our work also shows that acute loss-of-function approaches can complement and extend the reach of classical genetics in Dictyostelium and likely other systems as well .
For cells to move , their cytoskeletal structures become spatially organized by internal polarity signals [1–3] as well as external chemoattractant [4–6] . How such signaling cues tame actin dynamics to produce a pseudopod and guide cellular motility remains a key question in eukaryotic chemotaxis . By now , several key regulators of the actin cytoskeleton have been identified: in most cells , nucleation promoting factors ( NPFs ) of the Wiskott-Aldrich Syndrome Protein ( WASP ) and SCAR/WAVE family stimulate actin nucleation through the Arp2/3 complex and are essential for regulating polarity and motility for cells ranging from Dictyostelium [6 , 7] to metazoans [8–10] . NPFs themselves are regulated by self-association on the plasma membrane [1 , 11] and actin polymerization-based autoinhibition [1 , 12 , 13]; the actin polymer that they generate facilitates the removal of these NPFs from the plasma membrane . These positive and negative feedback interactions of the NPFs [1 , 14] and other actin regulators give rise to a range of highly dynamic , free-roaming , nonequilibrium actin structures such as flashes and traveling waves [1 , 2 , 5 , 6 , 15–21] , but how the actin machinery is coaxed to form these very different activity patterns is not well understood . Particularly striking displays of NPF and actin dynamics are actin oscillations , which can be observed in many cell types and contexts [1 , 2 , 5 , 22 , 23] . Biological oscillations are typically generated through a combination of ( 1 ) fast positive feedback , which amplifies small signals into an all-or-none output; and ( 2 ) delayed inhibition , which turns the output off and resets the system for the next pulse . By spatially coupling oscillators , spreading or synchronization over long distances can be achieved [24–26] . Recently , small oscillating SCAR/WAVE foci have been discovered at the periphery of Dictyostelium cells [2] . These foci may constitute the basic cytoskeletal units from which pseudopods are formed . In the absence of signaling cues , these oscillators are present but lead to only small undulations of the cell boundary . In response to upstream signals , however , full-blown protrusions emerge [2 , 27–31] , likely from the coordination of these foci . Some intracellular signals ( such as Ras and phosphatidylinositol 3 , 4 , 5-trisphosphate [PIP3] ) have been identified that affect this transition , but whether other signals link receptor activation with the SCAR/WAVE foci , and , more generally , which properties of the foci are modulated to enable large-scale coordination , are not known . Here , we find that the heterotrimeric G-protein subunit Gβ sets the coupling range of an actin-based activator—inhibitor system . Specifically , acute sequestration of Gβ leads to strong global synchronization of normally weakly coupled cytoskeletal oscillators , and these effects are independent of known upstream regulators of these oscillators , such as Ras and PIP3 . We show that this extended range of spatial coupling is detrimental for cell polarity , cell motility , and directional migration . To guide our intuition for how coupling between oscillators could affect the cell’s ability to sense directional cues , we developed a simple mathematical model that represents its minimal features . Simulations show that the ability to sense a noisy input signal is facilitated by an intermediate strength of oscillator coupling , allowing different membrane regions to share information about the stimulus . We propose that in wild-type cells , Gβ sets the coupling strength of actin oscillators to an appropriate level to sense directional upstream cues .
Strong loss-of-function phenotypes in cell motility are rare [6 , 32–38] . One reason may be that genetic perturbations are slow to act and may give cells time to compensate for gene loss [39–42] . Redundantly controlled processes like actin rearrangements during motility may be particularly susceptible to such compensation . To overcome this limitation , we developed a system that enables fast loss-of-function perturbations to cell signaling events involved in Dictyostelium cell motility . Here , we focus on its application to Gβ . Heterotrimeric G-proteins consist of one α , β , and γ subunit and link receptor-mediated signals to directed migration and polarization in eukaryotic cells ranging from yeast to neutrophils to Dictyostelium [43–46] . Both intra- and extracellular signals can regulate the cytoskeleton , yet while knockout of the sole Gβ protein in Dictyostelium completely blocks chemotaxis , basal cytoskeletal dynamics and other directional responses such as shear-flow-induced motility and electrotaxis are still present , although somewhat reduced [2 , 3 , 44 , 47 , 48] . Gβ requires plasma membrane localization in order to signal; thus , removal from the plasma membrane should prevent it from activating downstream effectors . As Gβ is continually exchanged between membrane and cytoplasm with a half-life of 5 s [49] , it should be possible to trap it by association with an internal anchor . We built a Gβ sequestration system using a chemical dimerization approach whereby the association of two protein domains ( FKBP and FRB ) is induced by the small molecule rapamycin [50–54] . Starting with Gβ-null cells [44] , we expressed an FRB—Gβ fusion protein and an endoplasmic reticulum ( ER ) -localized FKBP ( FKBP-calnexinA [55] ) . Thus , addition of rapamycin should drive Gβ relocalization to the ER and suppress its signaling function , effectively rendering cells Gβ-null in an acute fashion ( Fig 1A ) . To test for rapamycin-induced sequestration , we measured the extent of ER-localized Gβ in single cells over time following rapamycin addition . We computed the correlation between each cell’s fluorescence intensity in the ER anchor and Gβ channels to assess co-localization . FRB-RFP-Gβ was rapidly sequestered from the plasma membrane and increasingly co-localized in large clusters with FKBP-YFP-calnexinA ( S1 Movie ) . Sequestration is fast: half-maximal correlation occurred 5 . 6 min after addition of the highest dose ( 5 μM ) of rapamycin that was tolerated by cells ( Fig 1B and 1C , S1 Data ) . Sequestration kinetics were similar for both 5 μM rapamycin and 1 μM rapamycin . Therefore , unless indicated otherwise , we used the lower concentration for subsequent experiments . Gβ-null cells fail to transmit many signals triggered by G-protein-coupled receptors ( GPCRs ) [44 , 56–59] , and we should be able to recapitulate these defects with our sequestration approach . We thus assayed whether relocalization of Gβ to the ER inhibits transmission of signals from GPCRs to downstream effectors . Stimulating wild-type cells with chemoattractant ( cAMP ) triggers transient responses , including phosphorylation of PKBR1 , and this response is abolished in Gβ-null cells [33 , 56] . We found that introducing our FRB-Gβ construct in Gβ-null cells rescued the PKBR1 response . Acute sequestration of FRB-Gβ to the ER anchor blocked PKBR1 phosphorylation , but only when all three components of our system—the ER anchor , FRB-Gβ , and rapamycin—are present ( Fig 1D ) . Unfortunately experiments using the inducible sequestration system in developed cells were often problematic: Tagged Gβ and anchor components were frequently degraded during starvation and , likely as a consequence , cells failed to complete their developmental cycle . However , this problem was not observed in vegetative cells , in which the sequestration components remained intact . Gβ-dependent , chemoattractant-stimulated responses in vegetative cells , such as Ras activity and PIP3 production [57 , 59 , 60] , could also be blocked by Gβ-sequestration ( S1 Fig ) . Taken together , these results demonstrate that in the absence of rapamycin , our inducible sequestration system sustains key Gβ-dependent signaling events . In the presence of rapamycin , Gβ is sequestered from its site of action , thereby blocking receptor-based signaling . In this respect , sequestration of Gβ recapitulates Gβ-null cells . To probe for phenotypes that may only be apparent after rapid loss of Gβ , we turned to directional motility assays . We measured the behavior of Gβ-sequestered cells presented with two different directional cues—an attractive chemical ( folate ) or electric fields—and compared their responses with wild-type and Gβ-null cells . While chemotaxis is strictly dependent on Gβ , electrotaxis , the directed migration of Dictyostelium cells in response to electric fields , is not . While Gβ-null cells cannot move up a chemical gradient , they can move down electrical potential [44 , 47] . We took advantage of the heterogeneity in expression of components in Gβ-sequestered cells to internally control experiments . We can distinguish behavior of cells that , in the presence of rapamycin , are functionally wild-type ( expressing RFP-FRB- Gβ , but no CFP-FKBP-anchor ) , Gβ-null ( with no detectable RFP-FRB- Gβ expressed ) , or Gβ-sequestered ( expressing both RFP-FRB- Gβ and CFP-FKBP-anchor ) . For chemotaxis , we further compared these populations to true wild-type and true Gβ-null cells . We find that just as unsequestered cells resemble wild-type cells , Gβ-sequestered cells behave similarly to Gβ-null cells in chemical gradients . In the presence of Gβ , cells move directionally , while in the absence of functional Gβ ( through sequestration or knockout ) , directionality is lost ( Fig 2A ) . Furthermore , Gβ-sequestered cells ( 0 . 4 +/- 0 . 1 μm/min; n = 31; +/- SEM ) as well as true Gβ-null cells ( 0 . 4 +/- 0 . 05 μm/min; n = 98; +/- SEM ) move at a reduced speed compared to unsequestered ( 1 . 1 +/- 0 . 2 μm/min; n = 30; +/- SEM ) or true wild-type ( 2 . 5 +/- 0 . 15 μm/min; n = 97; +/- SEM ) cells . In contrast , in electrical fields , the behavior of Gβ-sequestered cells differs from the Gβ knockout . Compared to wild-type and Gβ-null cells , Gβ-sequestered cells show a significant decrease in their directionality during electrotaxis ( Fig 2B and S2 Movie ) . Furthermore , the speed of translocation in Gβ-sequestered cells ( 2 . 1 +/- 0 . 22 μm/min , mean +/- SEM; n = 34 ) was reduced compared to wild-type ( 3 . 8 +/- 0 . 23 μm/min , mean +/- SEM; n = 34; Student’s two tailed t test: p < 10-6 ) and Gβ-null cells ( 2 . 9 +/- 0 . 23 μm/min , mean +/- SEM; n = 33; Student’s two tailed t test: p < 0 . 006 ) . Closer examination of Gβ-sequestered cells by confocal microscopy revealed a striking change in the organization of the actin cytoskeleton . While wild-type cells have fairly stable levels of cortical and cytoplasmic actin , sequestration of Gβ induces striking oscillations of LimE-GFP , a reporter for dynamic F-actin ( Fig 3A and 3B ) [61] . Periodic loss of cytoplasmic LimE-GFP intensity is accompanied by a corresponding accumulation of F-actin around the entire periphery of the cell ( S2 and S3 Figs ) . The cytoskeletal oscillations induced by Gβ sequestration are present in the majority of cells and have well-defined characteristics . By automatically tracking cells over time and measuring their cytoplasmic LimE-GFP intensity , we identified oscillating cells from the characteristic peak induced in their Fourier spectrum ( S4 Fig ) . After rapamycin addition , the fraction of oscillating cells rises from 6% to 52% , but only when the ER anchor is co-expressed ( Fig 3C and S1 Data ) . The period of oscillation ( measured as the peak frequency of the Fourier-transformed signal ) is tightly controlled across all oscillating Gβ-sequestered cells ( 12 . 9 +/- 3 . 2 s , n = 83 ) ( S4 Fig ) . We also observed a second F-actin phenotype upon acute loss of Gβ . In ~10% of cells , waves of F-actin polymerization travel around the cell perimeter with a similar period as the whole field oscillations , taking 10–20 s for a full cycle ( S5 Fig and S3 Movie ) . Two lines of evidence confirm that acute Gβ loss of function through sequestration is required to initiate this actin oscillation phenotype . First , oscillations are not observed when the ER is forced into proximity of the plasma membrane , arguing against an ER-specific recruitment phenotype ( S3 Fig ) . Most importantly , when Gβ is overexpressed and sequestered in wild-type cells ( which harbor endogenous Gβ that cannot be recruited ) , no actin oscillations are induced ( Fig 3D and S1 Data ) . Individual cells transition abruptly into the oscillatory mode . Oscillations become apparent as soon as rapamycin-induced sequestration of Gβ can be observed ( S6 Fig and S3 Movie ) and can continue for days ( see later; Fig 4C and S1 Data ) . By treating cells with both rapamycin ( the FKBP-FRB heterodimerizer ) and a competitive inhibitor of heterodimerization ( the small molecule FK506 , an FKBP-FKBP homodimerizer ) , we titrated Gβ levels over the full dynamic range of the sequestration system ( S7 Fig ) . As the amount of sequestered Gβ is increased , the properties of the oscillating state such as its period and amplitude did not change ( S8 Fig ) . The oscillations have characteristics of an all-or-none behavior: only the percentage of oscillating cells increased ( Fig 4A and 4B , S1 Data ) . These phenotypes—whole-cell oscillations and traveling waves of actin polymerization—are reminiscent of previously observed actin-based activator—inhibitor systems [1 , 2 , 5 , 6 , 16–20] . However , the oscillations we observe here are triggered , persistent , and have an unusually large spatial range and high amplitude . This suggests that acute loss of Gβ pushes the cytoskeleton into an unusual state . Our observation that cortical F-actin oscillations follow acute sequestration of Gβ raises a key question: why did previous Gβ-null analyses fail to uncover this striking cytoskeletal phenotype ? Consistent with published work [2 , 3] , we find that very few Gβ-null cells display LimE-GFP oscillations cells ( Fig 4C and S1 Data ) . We reasoned that if cells compensate for the loss of Gβ function over time , the phenotype induced by acute sequestration of Gβ should approach the Gβ-null phenotype after sufficient time has passed . Consistent with this hypothesis , the fraction of oscillating cells decreases over days of continuous Gβ sequestration and eventually approaches the small fraction seen in Gβ nulls ( Fig 4C and S1 Data ) . Similar compensatory phenomena have been previously observed in other Dictyostelium signaling contexts . For example , the effect of LY294002 , a PI3K inhibitor , on Dictyostelium cell migration fades during prolonged treatment [2] , likely due to compensation by redundant signaling pathways [35] . In another case , the actin nucleator WASP relocalizes to the leading edge and compensates for SCAR/WAVE function when SCAR/WAVE is deleted [6] . Our findings suggest that a compensatory mechanism is also at work here: the globally oscillating state is suppressed in Gβ-null cells . Our results highlight the value of using acute inhibition to uncover protein function . We have used rapamycin-induced Gβ sequestration to interrogate loss-of-function phenotypes along two “axes” ( Fig 4D ) . By titrating the amount of sequestration while retaining its fast timescale ( axis 1 ) , it is possible to interrogate how a phenotype emerges , distinguishing between an all-or-none or gradual transition . Conversely , varying the timescale of perturbation ( axis 2 ) reveals whether phenomena such as cellular compensation can mask an acutely induced phenotype . Applied to Gβ sequestration , we find that a new phenotype—a globally oscillating F-actin cytoskeleton—can be uncovered at points in this “phenotypic space” that are not accessible to standard genetic perturbations . Multiple oscillating actin foci localize around the cell periphery on the basal surface of chemotactic cells . These foci often originate from previously aborted pseudopods that remain attached to the substrate . Internal and external signaling inputs are thought to entrain these foci , but how their dynamics are controlled for this to happen remains unknown ( e . g . , oscillation dynamics are unchanged in Ras , PI3K , and Gβ nulls ) [2] . The large-scale cortical actin oscillations we observe here are similar in period to the previously described oscillating foci ( 13 +/- 3 s versus 9 +/- 2 s , respectively ) , suggesting that these two forms of cytoskeletal dynamics may be closely related . Thus , we tested whether our acute sequestration of Gβ would reveal signaling control over these oscillatory actin foci . To analyze individual actin foci , we collected confocal movies imaged in the plane where cells contact the coverslip . We developed a computational approach to comprehensively track and quantify the dynamics of actin foci by automatically identifying each cell’s periphery , subdividing it into ten degree sectors ( thereby generating 36 tracked regions per cell ) , and measuring the mean intensity in each sector over time ( Fig 5A ) . Consistent with previous results [2] , we found large-amplitude oscillations in LimE-GFP intensity in some sectors ( Fig 5A , right panel ) but not others , with a mean period of approximately 10 s ( S9 Fig ) . Regulators of F-actin formation localize to the same structures and oscillate as well: the peak of actin-nucleating SCAR/WAVE complex member HSPC-300 precedes that of LimE by about 2 s; Arp2 and the F-actin binding domain ( ABD ) of ABP120 peak at about the same time as LimE; and the peak of Coronin , a regulator of actin disassembly [2 , 62] , lags behind LimE by more than 2 s ( Fig 5B , S10 Fig , and S1 Data ) . These data suggest that focal LimE oscillations report cycles of polymerization and disassembly of F-actin . We next addressed how the dynamics of actin foci compare between wild-type ( Gβ-unsequestered ) cells and Gβ-sequestered cells that exhibit whole-field oscillation . In both cases , individual sectors oscillate . However , the mean LimE intensity across all sectors in Gβ unsequestered cells does not show a marked oscillatory behavior ( Fig 5C ) , whereas the mean intensity of sectors in Gβ-sequestered cells clearly oscillates ( Fig 5D ) . Thus , the whole-field oscillations we observe upon Gβ-sequestration in the middle plane of cells ( Fig 3A ) are also reflected in the behavior of membrane-plane actin foci . What properties of these individual oscillators change as cells transition to whole-field oscillation ? We reasoned that changes in the amplitude , period , or the synchronization in phase between individual oscillating sectors could be responsible . We developed an automated approach using the Hilbert transform [63 , 64] , which has been used extensively to analyze neuronal activity [65 , 66] , to quantify the amplitude , period , and phase of individual oscillators over time ( S11 Fig ) . Using this algorithm , we extracted the oscillation phase ( i . e . , whether currently at a peak or trough ) as well as the instantaneous period ( i . e . , how fast the phase is changing ) at each timepoint . Strikingly , only the phase synchrony differs in Gβ-sequestered cells ( Fig 5E , S12 Fig and S1 Data ) . Yet although synchrony increases , it is not perfect: individual sectors can fall in and out of phase with the group over time ( S13 Fig ) . Taken together , our data suggest that global oscillations in Gβ-sequestered cells are caused by increasing synchronization among preexisting membrane oscillators . Downstream of Gβ , three signaling pathways , defined by PI3K , TORC2 , and PLA2 , are known to instruct actin-based motility in Dictyostelium ( Fig 6D ) . Ras activity can feed into both PI3K and TORC2 , and downstream , Rac activation is thought to connect these signaling modules to the actin cytoskeleton [31 , 33 , 38 , 56] . Enhanced activity of these pathways leads to wider , more stable zones of actin polymerization compared to the isolated oscillating foci . We investigated whether Gβ uses any of these signaling pathways to regulate spatial coupling of actin foci . First , we analyzed the dynamics of Ras activity , PIP3 levels , and Rac activity in single cells . Gβ sequestration neither induced oscillations nor caused any other apparent changes to these signaling currencies on a timescale of minutes ( Fig 6A ) . Second , we perturbed the activities of members of these pathways in wild-type cells to determine whether global LimE oscillations would emerge . Neither inducing Rac activity ( Tet-On: GFP-Rac1A[V12] ) , blocking all three pathways ( using a pharmacological cocktail: BEL|LY294002|pp242 ) , nor raising the levels of intracellular Ca2+ ( a messenger commonly oscillating in other systems [22 , 67] ) led to global oscillations of F-actin ( Fig 6B and S1 Data ) . Third , we interfered with these pathways in Gβ-sequestered oscillatory cells to determine whether their activity was required for synchrony . Acute inhibition of all three pathways caused only a very small decrease in the number of oscillating cells , while unbalancing Ca2+ levels did not inhibit global oscillations at all ( Fig 6C and S1 Data ) . We conclude that Gβ’s control over the coupling range of actin oscillators likely involves a different , currently unidentified mediator . How can hypercoupling between cytoskeletal oscillators lead to a defect in directed cell migration ? The coupling state among the oscillators might be an important parameter for upstream cues to polarize the cytoskeleton—a prerequisite for cell motility . To investigate this question , we tracked individual Gβ-sequestration cells over time , simultaneously monitoring cytosolic actin dynamics and cell migration in both the presence and absence of rapamycin . For this analysis , we returned to confocal imaging in the midplane of the cell . Here , polarization events are distinguished by a relatively stable actin patch that coincides with a substantial drop in cytoplasmic LimE-GFP reporter levels ( Fig7A and 7B and S14 Fig ) . In both control and Gβ-sequestered cells , polarized patches are of similar intensity ( S15 Fig ) , and phases of polarity alternate with apolar phases , which can easily be visualized in t-stack kymographs ( Fig 7A and 7B; left panels ) . In this representation , the y-axis represents time , and the lateral surface of the cell is shown for each timepoint along the x-axis . We found that Gβ-sequestered as well as Gβ unsequestered cells were capable of cycling between polarized and apolar states ( S4 and S5 Movies ) . Consistent with our prior results , acute sequestration of Gβ induced large-amplitude oscillations of F-actin . However , long-term imaging revealed that these oscillations are largely restricted to apolar phases—times when the cell is not undergoing protrusion or migration ( Fig 7B and S5 Movie ) . Thus , phases of polarization appear to be incompatible with whole cell oscillations . While increased coupling in Gβ-sequestered cells did not affect the lifetime of poles once they successfully formed ( Fig 7C and S1 Data ) , Gβ sequestration significantly ( p < 10-4 , Student’s two-tailed t test ) impaired the establishment of new poles ( Fig 7D and S1 Data ) . Consistent with a reduced number of cell polarization events , sequestered cells translocate at a significantly reduced speed ( p < 0 . 003 , Student’s two-tailed t test , Fig 7E and S1 Data ) . Taken together , our data show that appropriate control of coupling between localized cytoskeletal oscillators is essential for efficient polarization and motility as well as directional sensing . Increasing the strength of coupling—through acute loss of Gβ—synchronizes actin dynamics , which hampers the entrainment of the actin cytoskeleton by both internal polarity cues as well as entrainment by the external cues that are necessary to direct motility ( Fig 8 ) . One of the most remarkable features of chemotaxis is the ability of migrating cells to accurately sense extraordinarily shallow chemical gradients [68] . Previous work has suggested that the signaling network downstream of Gβ plays a crucial role in this input sensing [5 , 29 , 69 , 70] . Here , we have uncovered a separate link between Gβ and the cytoskeleton in tuning coupling between actin oscillators . Might oscillator coupling also play a role in input sensitivity ? We reasoned that oscillator-to-oscillator coupling might represent a means of sharing information between nearby regions of the cell periphery . By comparing noisy receptor—ligand interactions at multiple locations , cells might improve their ability to discriminate signal from noise when choosing a migration direction . To test this hypothesis in a simple context , we built a mathematical model representing input sensing at the cell’s periphery ( Fig 9 ) . It should be emphasized that this model is not meant to capture the full complexity of the cell’s gradient sensing and chemotaxis pathways , but rather represents a minimal model to quantitatively interrogate the essential elements of oscillator-to-oscillator coupling and entrainment to an input . Our model incorporates a circular lattice of actin oscillators representing the cell’s cortex . Oscillators are coupled to one another by a term that increases sinusoidally with their difference in phase [71] and can also be coupled to an oscillating input signal using the same mechanism . Although the chemoattractant signals presented to a real cell are unlikely to oscillate in this fashion , the exact mechanism for input coupling is unknown , and our simplifying assumption allowed us to model oscillator-to-input and oscillator-to-oscillator coupling in a single unified framework . Our model includes three parameters that define the coupling between an external input and the nearby membrane ( kIN ) and the coupling between membrane oscillators ( parameters k1 and k2 for input-coupled and non-input-coupled membrane oscillators ) . We also include a term ( σ ) to represent noise in input-to-oscillator coupling . Our model reproduced well-known features of coupled oscillator systems . Increasing oscillator-to-oscillator coupling showed an abrupt transition to global synchrony , consistent with prior work modeling the synchronization of weakly coupled oscillators as a phase transition ( S16 Fig ) [68 , 69] . This is analogous to the effect observed after Gβ sequestration , in which the transition to global oscillation appears to be all-or-none in individual cells ( Fig 4 and S1 Data ) . To test how coupled oscillators are affected by features of the input signal , we set out to determine how oscillator-to-oscillator coupling affected sensing of weak inputs ( low values of kIN ) or noisy inputs ( high values of σ ) . We found that increasing coupling could not improve sensing of weak noise-free inputs but rather led to spontaneous synchronization as coupling strength is increased ( S21 Fig ) . In contrast , oscillator-to-oscillator coupling markedly improved sensing of noisy inputs ( Fig 9 ) . For simulations with little or no coupling , the effect of noise was dominant , and membrane oscillators were unable to accurately couple to inputs ( Fig 9; k1 = 0 . 1 ) . Conversely , for very strong coupling , oscillators became synchronized to one another so strongly that they were completely input-insensitive ( Fig 9; k1 = 3 . 5 ) [25 , 71] . Between these two extremes , our model revealed an optimum of input sensitivity at an intermediate coupling strength ( Fig 9; k1 = 2 . 5 ) . If weak oscillator-to-oscillator coupling was indeed beneficial for input sensing , one would expect wild-type cells to exhibit some coupling between oscillating foci . Indeed , we find experimentally that in wild-type cells the relative phases of oscillators are not random but loosely correlated ( Fig 5E , asynchrony; phase distribution width Θ50 < 90 [deg . ]; S12 Fig and S1 Data ) . Thus , we propose that upstream signaling cues optimally entrain the cytoskeleton when the coupling strength between its dynamic units is of intermediate strength .
A dynamic actin cytoskeleton drives eukaryotic cell migration . Waves , flashes , patches , and oscillatory actin foci have been observed in Dictyostelium [2 , 5 , 6 , 15–19] , neutrophils [1 , 4] , and other mammalian cells [17 , 20 , 72 , 73] . Underlying these phenomena are nonlinear reaction processes that exhibit a range of behaviors including excitability and oscillations [1 , 2 , 26 , 29 , 70 , 74] . These cytoskeletal dynamics are shaped further by upstream cues such as internal polarity signals [1–3] and external chemoattractant [4 , 5] . Here , we show that signaling also directly regulates the strength of coupling between local cytoskeletal processes . Acute loss of Gβ leads to strong synchronization of actin oscillators , which has detrimental consequences for cell polarity , motility , and directionality . Electrotaxis revealed this new role for Gβ in directed cell migration . However , we expect the link between Gβ and cytoskeletal dynamics to be essential for interpreting other cues as well . During chemotaxis , when the activity of the Gαβγ heterotrimer is proportional to the amount of the chemical signal the cell experiences [75] , fine control of the magnitude and intracellular distribution [76] of oscillator coupling may be possible . In future work , it will be important to learn more about how Gβ exerts this control . Common signaling pathways involving PI3K , TORC2 , and PLA2 appear to be not essential . Similarly , perturbing levels of Ca2+ , a messenger known to oscillate in many systems [22 , 67] , including chemotaxing Physarum polycephalum cells [77] , shows no obvious effect on coupling between actin oscillators . P . polycephalum , however , is a beautiful , conceptual precedent for the idea that cell movement may be governed by the coupling between independent oscillators: in this organism , periodic streams of small pieces of cytoplasm can become entrained to each other , which , through further modulation by attractants or repellants , supports directional movement [78] . Recent evidence in Dictyostelium shows that Gβ interacts with Elmo , which suggests a possible direct link to the cytoskeleton bypassing the other signaling pathways [79] . We observed oscillation of HSPC-300 , a member of the actin-nucleating SCAR/WAVE complex . This may be the most upstream oscillator , with F-actin reporters and disassembly factors ( e . g . , Coronin ) following its dynamics . In this case , SCAR/WAVE’s relevant regulators will need to be identified [80] . Mechanistically , how could loss of Gβ increase the strength of coupling ? Based on the mechanisms through which oscillators are coupled in other systems , possible explanations include ( 1 ) increasing the density of oscillators at the periphery while keeping the coupling range of each constant [81] , and/or ( 2 ) directly increasing the range of a diffusible or mechanical signal that is generated by the oscillators [26] . Our experimental data support the first hypothesis . In strongly coupled Gβ-sequestered cells , a larger fraction of sectors contain actin oscillators compared to Gβ-unsequestered cells or Gβ-null cells ( S17A Fig ) . Moreover , upon Gβ sequestration , the number of membrane sectors that contain an actin oscillator increases , while the amplitude of the oscillators remains constant ( S17B Fig ) . Additionally , we find that during cell polarization , oscillators largely disappear from the sides and back of the cell ( S18 Fig ) . Taken together , our data suggest that the number or density of oscillators is regulated , and this may be used as a mechanism to control coupling strength . Additional mechanisms could affect the firing threshold or the refractory period of the oscillators . Our simple mathematical model helps to guide intuition on why coupling between oscillators could be advantageous for polarity and directional movement . For both cases , the signals that need to be interpreted can be noisy , and in these scenarios moderate coupling between oscillators can provide an advantage—input-coupled oscillators can “share” information to filter noise and better entrain to an input signal . Our results are consistent with recent predictions in bacterial chemotaxis , in which an optimal membrane distribution of receptors balances sensitivity to spatially correlated external noise and spatially uncorrelated intrinsic noise ( which can be filtered out by a similar mechanism of local information sharing ) [82] . Our work highlights limitations in classical genetic approaches . Genetic nulls are the most common means of assaying gene function in Dictyostelium . However , many genetic mutants give no or mild phenotypes , and they often require combined hits in multiple signaling pathways to significantly inhibit chemotaxis [6 , 32–36] . In theory , two mechanisms can account for this: a selection on the population levels can favor a subset of cells ( potentially carrying suppressor mutations ) that best cope with the genetic change . Alternatively , intrinsic redundancy with parallel pathways or slow compensation via negative feedback can obscure the true role of a gene in cell behavior [42] . Such compensation enables robust function and is a widely employed characteristic of adaptive/homeostatic systems . For example , pharmacological inhibition of synapses transiently inhibits signal transmission , but homeostatic mechanisms restore function within minutes [39–41] . The motor of bacteria is another example . It compensates for persistent changes in the level of internal signaling components to maintain the robustness of chemotaxis [41] . Which mechanism is at play in our case ? Both Gβ-null knockout and wild-type cells lack excessive coupling of actin oscillators , albeit likely for different reasons . In wild-type cells , Gβ suppresses the coupling , while Gβ-null knockout cells have , over time , arrived at a Gβ independent steady state that does not support oscillations . Transformation of Gβ-null knockout cells with the sequesterable Gβ construct restores wild-type physiology , which can become transiently unbalanced upon acute Gβ sequestration . This imbalance remains for days—long enough for us to observe its effect on oscillator coupling—but eventually the steady state of Gβ-null knockout cells is assumed again , potentially due to compensation from parallel pathways . We favor this possibility over genetic suppression based on the speed with which oscillations disappear again after induction . As a consequence of compensation , different modes of gene inactivation can result in strikingly different phenotypes . In zebrafish , gene knockdowns can produce strong phenotypes that are masked by compensation in genetic knockouts [83] . Our data suggest that additional phenotypes appear when proteins become inactivated even more rapidly . Gβ-knockout and knockdown cells have been extensively studied in Dictyostelium and other systems [43 , 44 , 84] . Although defects have been reported for a wide range of chemoattractant-stimulated responses , including directed migration [44] , these cells display normal basal polarity and actin dynamics [2 , 3] . Acute sequestration was essential to uncover the role of Gβ in tuning cytoskeletal dynamics and initiating cell polarity . In this light , our work suggests that much can be learned by revisiting classical mutants with acute perturbation approaches , and not only in instances in which a loss-of-function mutation is lethal .
Dictyostelium strains were grown at 22°C in HL5 medium ( ForMedium ) in Nunclon tissue culture dishes or in suspension in flasks shaken at 180 rpm . Cells were routinely used from nonaxenic cultures . In this case , cells were grown in association with Klebsiella aerogenes ( K . a . ) on SM agar plates and used for assays when bacteria began to get cleared [85] . Growth under these conditions gave the strongest responses to stimulation with folate , so this condition was used for most subsequent rapamycin-mediated sequestration experiments . However , sequestration of Gβ also induced oscillations in F-actin when cells were grown in HL-5 instead . For imaging experiments , a scrap of cells was seeded in 200 μl HL5 in a Lab-Tek II 8 well chamber ( Nunc ) , allowed to settle , and washed one to two times in KK2 ( 16 . 5 mM KH2PO4 , 3 . 9 mM K2HPO4 , 2 mM MgSO4 ) immediately before the assay . Rapamycin ( SIGMA ) was freshly prepared at 2 μM in KK2 and added 1:1 in sequestration experiments . To render cells responsive to cAMP ( Fig 1C ) , aggregation-competent amoebae were prepared by resuspending washed cells at 2 x 107 cells/ml in KK2 , starving them for 1 h while shaking at 180 rpm , followed by pulsing the cells with 70–90 nM cAMP ( final conc . ) for another 4 h . Before stimulation with 1 μM cAMP , cells were basalated ( shaking at 180 rpm in the presence of 5 mM caffeine for 20 min ) with or without 5 μM rapamycin , washed in ice-cold KK2 , and kept on ice until stimulation . For analysis by western blotting , samples were resolved on 4%–15% SDS-PAGE gels . After electrophoresis , proteins were transferred to PVDF membrane and probed with anti—phospho-PKC ( pan ) antibody from Cell Signaling Technology ( 190D10 ) , which was used to detect the activation loop ( T309 ) phosphorylation of PKBR1 , and anti—pan-Ras antibody from EMD ( Ab-3 ) . Sequestration of Gβ was confirmed by fluorescence microscopy . PhdA-GFP , LimEΔcoil-GFP , LimEΔcoil-RFP , GFP-Arp2 , Hspc300-GFP , ABD-GFP , GBD ( PAK ) -YFP , Coronin-GFP , and YFP-RBD ( PI3K1 ) have been described previously [2 , 5 , 35 , 86] . Standard methods of molecular biology , including reagents from Quiagen and Zyppy Plasmid Miniprep Kits from Zymo Research , were used to generate the following constructs: SRC-YFP-FRB ( pHO34 ) was assembled in pDXA-YFP by subcloning FRB ( XhoI/XbaI ) from pOW578 with a synthetic sequence ( HindIII/Nsi1 ) encoding the myristoylation tag from SRC . cAR1-RFP-FRB ( pHO39 ) was assembled in pDXA-YFP by replacing YFP with a fragment containing cAR1-RFP ( HindIII/XhoI ) and adding amplified FRB ( XhoI/XbaI ) . CalnexinA-CFP-FKBP ( pHO232 ) was assembled in multiple steps . CalnexinA was amplified from a published plasmid [55] and inserted into a variant of pDXA-YFP encoding FKBP ( pHO167 ) or CFP and FKBP ( pHO232 ) . A Gateway-compatible vector derived from pDM448 [87] encoding FRB-RFP was generated ( pHO436 ) , into which Gβ was inserted with an LR reaction to build FRB-RFP-Gβ ( pHO536 ) . A tetracycline-inducible variant of GFP-Rac1AV12 ( pHO578 ) was built by enzymatic assembly ( Gibson ) in pDM369 [87] . To generate stable cell lines , cells were transformed by electroporation ( Genepulser Xcell , Bio-Rad ) using 10–20 μg DNA per 4x106 cells ( 100 μl ) in 1 mm cuvettes ( Bio-Rad ) . Two consecutive pulses with a 5-s recovery period between were delivered at 750 V , 25 μF , and 50 Ohm . For overexpression , cells were plated in bulk and selected with G418 ( 10 μg/ml ) and/or hygromycin ( 50 μg/ml ) the next day . The time course of inducible sequestration ( Fig 1B ) was benchmarked in strain HO543: A Gβ-null strain ( LW6 ) derived from DH1 [44] was used as the base strain into which the sequestration system was engineered . First , pHO536 ( FRB-RFP-Gβ ) was introduced , and transformants were selected with hygromycin ( 50 μg/ml ) to give HO535 . This strain was then transformed with pHO167 ( calnexinA-YFP-FKBP ) to give HO543 or simultaneously with pHO232 ( calnexinA-CFP-FKBP ) and LimEΔcoil-GFP to give HO547 , with pHO232 and phdA-GFP to give HO548 , with pHO232 and pOH250 to give HO549 , and with pHO232 and PAK ( GBD ) -YFP to give HO630 . Additional anchors , such as a NLS or the transmembrane domain of Miro , were tested , but yielded poor depletion of Gβ . Transformants were selected with G418 ( 10 μg/ml ) . When appropriate for comparison , parent strains DH1 expressing LimEΔcoil-GFP ( HO618 ) or LW6 expressing LimEΔcoil-RFP ( HO595 ) were analyzed . The following strains were used to control for the effect of rapamycin mediated recruitment: DH1 expressing LimEΔcoil-GFP , pHO232 and pHO39 ( HO620; G418 resistant ) , Ax2 ( Kay lab ) expressing LimEΔcoil-RFP , pHO232 and pHO34 ( HO621; G418 resistant ) and Ax2 ( Kay lab ) expressing LimEΔcoil-GFP , pHO232 and pHO536 ( HO626; G418 and hygromycin resistant ) . For dual color oscillation experiments ( Fig 5B ) , Ax2 ( Kay lab ) cells expressing LimEΔcoil-RFP together with GFP-Arp2 ( HO632 ) , Hspc300-GFP ( HO634 ) , ABD-GFP ( HO638 ) , or Coronin-GFP were analyzed . A spinning disc Nikon Eclipse Ti fitted with a spinning disc head , 405 nm , 488 nm , and a 561 nm laser line and appropriate emission filters were used to record CFP , RFP , and GFP ( or YFP ) double- or triple-labeled cells at room temperature . Images were routinely recorded using a 60x ( 1 . 45 NA ) objective , a Clara Interline CCD camera ( Andor Technologies ) , and NIS Elements software . After analysis , when necessary for presentation , contrast was adjusted uniformly using ImageJ or Photoshop , and to image sets of some experiments a uniform Gaussian Blur was applied . To quantify oscillations , a single two- or three-channel image was taken to assess Gβ sequestration , followed by a 2-min movie ( 1 frame/second ) to record behavior in the reporter channel at the lowest laser intensity necessary for reasonable signal-to-noise . Longer imaging periods ( 10 min ) and/or adjustment of the focal plane close to the coverslip were used when necessary ( e . g . , to record individual oscillating foci or alternating polar and apolar states ) . For Fig 5 , Ax2 cells expressing LimE-RFP were analyzed for 2 min ( 1 frame/second ) immediately before and for 2 min ( within 5 min ) after applying perturbations . For Gβ sequestration , only oscillating cells ( strain HO547 ) were considered . Ca2+ and ionomycin were used at 10 mM and 10 μM , respectively . For triple drug inhibition , Bromoenol lactone ( BEL 5 μM ) was washed out after 5 min of treatment , after which acute application of LY294002 ( 50 μM ) together with pp242 ( 40 μM ) followed . BEL and LY294002 have been demonstrated as effective inhibitors of PLA2 and PI3K in Dictyostelium before [38]; pp242 is an inhibitor of TOR kinase and inhibits TORC2-mediated phosphorylation events in Dictyostelium ( S19 Fig ) . Expression of tet-on GFP-Rac1A ( V12 ) , was induced overnight with 100 μg/ml doxycycline . The effect on oscillating , Gβ-sequestered cells was additionally tested by treatment with U73122 ( 5 μM ) , EGTA ( 10 mM ) , and Ca2+ ( 10 mM ) . The electric fields were applied as previously described for vegetative Dictyostelium cells [88] by using μ-Slides ( Ibidi ) . These tissue-culture-treated slides with small cross-sectional area provide high resistance to current flow and minimized Joule heating during experiments . To eliminate toxic products from the electrodes that might be harmful to cells , agar salt bridges made with 1% agar gel in Steinberg’s salt solution were used to connect silver/silver chloride electrodes in beakers of Steinberg’s salt solution to pools of excess developing buffer ( 5 mM Na2HPO4 , 5 mM KH2PO4 , 1 mM CaCl2 , and 2 mM MgCl2 , pH 6 . 5 ) [89] at either side of the chamber slide . EF strength is empirically chosen ( ~10V/cm ) based on our previous experience [90] and measured by a voltmeter before and after each experiment . Fields of HO547 cells were chosen based on the presence of Gβ and anchor expressing cells , which were distinguished by fluorescence imaging ( see Microscopy section for details ) . High-definition DIC movies ( 1 frame/30 s ) were recorded at room temperature for at least 30 min after the electric field was switched on . To quantify directionality and speed , time-lapse images were imported into ImageJ ( http://rsbweb . nih . gov/ij ) . Tracks were marked by using the MtrackJ tool and plotted by using the Chemotaxis tool described [91] . All experiments were repeated and produced similar results . Data are combined and presented as means +/- SEM ( standard error ) . To compare group differences , unpaired , two-tailed Student’s t test was used . A p-value of less than 0 . 05 is considered significant . HO543 , DH1 , or LW6 ( Gβ null ) cells were grown in HL5 medium containing 20 μg/ml G418 and 50 μg/ml hygromycin . Two days before the experiment , 2x105 cells were mixed with an overnight culture of K . a . in 250 μl streptomycin-free HL-5 medium and plated on an SM agar plate . On the day of the experiment , cells were washed off the SM plate with DB buffer , washed once , and resuspended in DB at 2x107 cells/ml . Suitable amount of cells were transferred to LabTek II chambered coverglass ( Nalge Nunc ) containing DB with 5 μM rapamycin and 0 . 05% DMSO . For folic acid chemotaxis , Femtotips microcapillary pipettes ( Eppendorf ) filled with 1 mM folic acid were used . Microscopy for this set of experiments was carried out with a Nikon Eclipse TiE microscope illuminated by an Ar laser ( YFP ) and a diode laser ( RFP ) . Time-lapse images in bright field , YFP , and RFP channels were acquired by a Photometrics Evolve EMCCD camera controlled by Nikon NIS-Elements . Tracks of cell migration were analyzed in ImageJ to obtain directedness and speed of cells . For all other analyses , cells were identified , tracked , and processed to extract various properties ( e . g . , cytoplasmic fluorescence , membrane fluorescence , extent of polarization , angle of polarization ) using custom code written in Matlab . First , initial locations for each cell were provided by hand-drawn masks such that each mask contains a single cell at the first timepoint . At each subsequent timepoint , each cell was tracked by extracting a 100x100 pixel box centered at that cell’s prior location in the LimE-GFP fluorescent channel . To identify the cell within this box region , interior pixels were separated from background intensity using a fixed intensity threshold , followed by binary erosion with a single-pixel structuring element ( to remove isolated noncell pixels ) and a hole-filling operation ( to fill all pixels within the cell ) . The largest connected component within this image was assumed to be the cell . For each cell and at each timepoint , we extracted the following features: To identify which cells in a population were oscillating and characterize the timescale of oscillation , we turned to a Fourier approach ( for the analyses of Figs 3 and 4 ) . We found that the cytoplasmic LimE-GFP levels undergo strong , regular periodic fluctuations . From each cytoplasmic intensity timecourse , we subtracted a 30 s moving average to center cytoplasmic fluctuations on a mean value of zero ( eliminating intensity fluctuations during cell movement or photobleaching ) and computed the discrete Fourier transform of this mean-centered signal . Cells were then marked as “oscillating” if any sampling frequency between 0 . 05 and 0 . 2 Hz contained at least 10% of the cytoplasmic signal’s total power ( see S4 Fig for oscillating and nonoscillating representative cells ) . These frequencies correspond to periods ranging from 5 to 20 s , which covered the range of frequencies we observed in a preliminary analysis across more than 50 oscillating cells . Each cell’s oscillation frequency was then taken to be the sampling frequency at which the power was maximal . To understand how cortical LimE dynamics relate to those of other cytoskeletal factors , we sought to correlate LimE-RFP with other reporters ( GFP fusions to HSPC300 , Coronin , the ABD actin binding domain of ABP120 , and Arp2 ) . To identify cells expressing both LimE and a second reporter , we thresholded cells using both GFP and RFP fluorescence . The cell’s cortex was identified as a 5-pixel-wide shell of this thresholded image for each cell . To compute the intensity of cytoskeletal foci around the cell’s cortex , we then subdivided the cortex into 36 equal-angle segments ( sweeping out 10 degrees each ) and measured the fluorescence intensity in both the GFP and RFP channels . We then sought to compare the temporal dynamics of GFP and RFP in each spatial region from each cell . To do so , we calculated the cross-correlation between these two channels . For uncorrelated cytoskeletal factors ( e . g . , myosin , paxillin ) , we found that dynamics in GFP and RFP were uncorrelated , leading to a low-magnitude , flat cross-correlation . For correlated cytoskeletal factors ( e . g . , HSPC300 , Coronin , Arp2 , and the actin binding domain ABD ) , the cross-correlation peaked at the characteristic delay time between LimE and that particular cytoskeletal factor . We estimated this delay time by fitting a Gaussian distribution to the cross-correlation to identify the location of this peak—the resulting delay times are shown in Fig 5B . From the centroid and center of mass measurements described above , the direction and extent of polarization was determined by computing the vector between the center of mass ( c→ ) and centroid ( n→ ) . The magnitude of p→ describes the extent of polarization , while its direction reflects the pole’s orientation . We were also interested in identifying periods of time in which cells exhibit long-term , stable polarization ( for the analyses of Fig 7 ) . By inspecting many cell trajectories , we found that stable polarization was associated with a consistent direction of polarity—cells would retain a pole with a similar directional orientation , and changes in direction were associated with the formation of a new pole . Conversely , during unpolarized phases , fluctuations of actin around the membrane would lead to frequent changes in the direction of p→ ( S14 Fig; lower panels ) . Thus , we implemented a greedy search algorithm to find continuous periods of time when the angle of polarization was contained in a 1-radian window and lasted at least 25 s , and measured the number and duration of these polar regions for each cell ( S14 Fig shows two representative cells ) . To assess the synchrony of oscillation between different membrane regions of a cell , we set out to measure each region’s oscillation phase at each timepoint . The phase of oscillation describes the current position of an oscillating signal on a sinusoidal curve ( i . e . , the rising or falling edge ) , and periodically rises from 0 to 2π . Thus , by comparing the phases between different regions of the membrane , we could assess whether they were oscillating in synchrony , with the phase rising and falling together , or whether at a single timepoint different membrane regions were at different points in their oscillating trajectories . The analytic representation of a signal provided by the Hilbert transform is an ideal way to measure instantaneous properties of a signal containing periodic fluctuations such as the oscillation phase . For the time-varying LimE-GFP intensity in the nth membrane sector xn ( t ) , the analytic signal x˜n ( t ) =xn ( t ) +i xn ( t ) *1πt is a complex-valued function from which instantaneous properties of the signal’s oscillation can b e calculated , such as its instantaneous oscillation phase φn ( t ) =∠ x˜n ( t ) and frequency ωn ( t ) =φ˙n ( t ) . Phase measurement can be improved by first applying a low-pass filter to avoid noisy fluctuations from being interpreted as oscillation . Thus , we first applied a low-pass filter ( an 8th order Butterworth filter with a cutoff of 0 . 2 Hz ) to each membrane trajectory before calculating its Hilbert transform , using custom Matlab code . We found this procedure to yield highly robust measurements of oscillation phase ( S11 Fig ) in both Gβ-sequestered and Gβ-functional cells . The instantaneous frequencies we measure from this approach are closely centered at ~10 s ( S9 Fig ) and are strikingly similar to those measured by Fourier analysis of cytoplasmic oscillation ( Fig 3 ) . To assess synchrony between different membrane regions , we measured the breadth of spread in oscillation phase between them , at all timepoints during oscillation . We first computed the “group phase”—the vector sum of all regions’ individual phases , weighted identically . We assessed synchrony by computing the phase difference between each membrane region and the group phase at each timepoint , and measured how broad this distribution is in oscillating Gβ-sequestered and nonoscillatory Gβ-functional cells ( S12 Fig shows histograms of two representative cells ) . To characterize the migration of Gβ-sequestered and Gβ-unsequestered cells , we tracked individual cells during 10 min movies , where fluorescent images were acquired once per second . Cells were automatically segmented by thresholding the fluorescent channel , and the centroid of each cell was automatically determined at each timepoint . At least 28 cells were tracked in each condition . From each cell’s centroid data , we calculated the root-mean-squared displacement xrms over time for each cell , choosing 300 distinct 5-min intervals for each cell during the 10 min movie . We fit the data to the simple diffusion model xrms2=2dDt , where d = 2 is the dimensionality , D is the diffusion constant , and t is the current time . From this model , we estimated the diffusion coefficient for each cell , and computed the p-value for a difference in diffusion coefficients between Gβ-sequestered and Gβ-unsequestered cells ( Fig 7E ) . | The actin cytoskeleton of motile cells is comprised of highly dynamic structures . Recently , small oscillating actin foci have been discovered around the periphery of Dictyostelium cells . These oscillators are thought to enable pseudopod formation , but how their dynamics are regulated for this is unknown . Here , we demonstrate that the strength of coupling between individual actin oscillators controls cell polarization and directional movement . Actin oscillators are weakly coupled to one another in wild-type cells , but they become strongly synchronized after acute inactivation of the signaling protein Gβ . This global coupling impairs sensing of internal cues during spontaneous polarization and sensing of external cues during directional motility . Supported by a mathematical model , our data suggest that wild-type cells are tuned to an optimal coupling strength for patterning by upstream cues . These observations are only possible following acute inhibition of Gβ , which highlights the value of revisiting classical mutants with acute loss-of-function perturbations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"physiology",
"cell",
"motility",
"dictyosteliomycota",
"cell",
"polarity",
"model",
"organisms",
"cellular",
"structures",
"and",
"organelles",
"cytoskeleton",
"research",
"and",
"analysis",
"methods",
"contractile",
"proteins",
"actins",
"slime",
"molds",
"proteins",
"protozoan",
"models",
"ras",
"signaling",
"cell",
"membranes",
"biochemistry",
"cytoskeletal",
"proteins",
"genetic",
"oscillators",
"signal",
"transduction",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"dictyostelium",
"cell",
"signaling",
"protists",
"organisms"
] | 2016 | Gβ Regulates Coupling between Actin Oscillators for Cell Polarity and Directional Migration |
The growth of scleractinian corals is strongly influenced by the effect of water motion . Corals are known to have a high level of phenotypic variation and exhibit a diverse range of growth forms , which often contain a high level of geometric complexity . Due to their complex shape , simulation models represent an important option to complement experimental studies of growth and flow . In this work , we analyzed the impact of flow on coral's morphology by an accretive growth model coupled with advection-diffusion equations . We performed simulations under no-flow and uni-directional flow setup with the Reynolds number constant . The relevant importance of diffusion to advection was investigated by varying the diffusion coefficient , rather than the flow speed in Péclet number . The flow and transport equations were coupled and solved using COMSOL Multiphysics . We then compared the simulated morphologies with a series of Computed Tomography ( CT ) scans of scleractinian corals Pocillopora verrucosa exposed to various flow conditions in the in situ controlled flume setup . As a result , we found a similar trend associated with the increasing Péclet for both simulated forms and in situ corals; that is uni-directional current tends to facilitate asymmetrical growth response resulting in colonies with branches predominantly developed in the upstream direction . A closer look at the morphological traits yielded an interesting property about colony symmetry and plasticity induced by uni-directional flow . Both simulated and in situ corals exhibit a tendency where the degree of symmetry decreases and compactification increases in conjunction with the augmented Péclet thus indicates the significant importance of hydrodynamics .
Some corals are known to have a high degree of morphological plasticity along different environmental conditions [1]–[3] . Light and water motion are the main environmental factors related to morphological variations of a coral colony [2] . The water movement over a reef is responsible to transport the necessarily nutrients [4]–[6] , to flush the waste products [7] , increase particle capture [8] , and enhance photosynthesis and respiration [9]– . There is evidence to suggest that different morphologies emerge in response to changes in hydrodynamic energy [12] , [13] . In particular for branching corals Pocillopora damicornis , Seriatopora hysterix and the hydro coral Millepora alcicornis , measurements showed that their morphology had changed in accordance with the intensity of the ambient flow – a compact shape under high flow environment , and a more open growth form with thin branches under lower flow [13] . Earlier experiments with branching corals in flume studies [14] suggested that densely packed branching colonies act as a solid body . Water flow ( even for a relatively high flow velocity about 20 cm s−1 ) starts to circumvent the colony creating a stagnant region inside . In a low energy environment , sparely spacing branching structures allow flow to penetrate deeper inside the colony as opposed to densely compacted branching . Chang and collaborators [15] reported a variation in intra-colony flow pattern in scale models of Stylophora pistillata . Colonies from the low flow environment were reported to distribute the flow velocity more evenly throughout the interior compared to the colony collected from higher flow regime . Reidenbach et al . [16] estimated mass transfer in unidirectional and oscillatory flows of three coral species - Stylophora pistillata , Pocillopora verrucosa and Pocillopora compressa . Mass transfer rate was shown to be a function of the physical flow characteristic and the morphology of branching structure . For unidirectional current , mass transfer rate of sparsely branched corals was observed to be larger for increasing flow velocity . At a higher frequency oscillatory flow , however , the compactness in branch spacing greatly enhance mass transfer rate through the inner structure of the branches . Water motion also influences the coral's physiological mechanisms during photosynthesis and respiration [10] , [17] . At the tissue level epithelial transport of ions , Dissolved Inorganic Carbon ( DIC ) and oxygen occurs by passive diffusion due to a standing gradient of these compounds around the coral branches . These gradients , created in the Diffusive Boundary Layer ( DBL ) , are directly influenced by the effect of flow [4] , [18] . In two studies by Nakamura et al . [19] , [20] the susceptibility of the coral Acropora digitifera to bleaching during hot periods and under low flow conditions was demonstrated . Another recent study [11] focused on the effect of flow-dependent efflux of oxygen during photosynthesis , showing the importance of water movement to remove the excessive oxygen from the organism , as oxygen accumulation inhibits photosynthesis . Under natural conditions branching corals ( e . g . Pocillopora sp . , Stylophora pistillata , Madracis mirabilis ) tend to develop symmetrical branching colonies - as shown in Figure 1A . Two hypotheses can be found in the literature to explain this phenomenon . According to some authors gene regulation is controlling the symmetrical shape [21]–[23]; whereas an alternative explanation is that the symmetry of the colony can be explained by environmental influences and the supply of nutrients [24] . In order to investigate these hypotheses , Mass and Genin [25] conducted an in-situ control flow study . The authors observed that Pocillopora verrucosa growing in unidirectional flow condition developed an asymmetrical branching form with strong growth in the upstream direction of flow leading to the conclusion that extrinsic factors control the colony symmetry [11] , [25] . These results show that environmental factors contribute greatly to the asymmetry in coral colonies . Former simulation models were used to address the influence of flow on the morphology of scleractinian corals . The first approach used a three-dimensional aggregation model ( represented on a cubic lattice ) coupled with a hydrodynamic model [26] . However a major problem in this model was that the morphology of three-dimensional aggregates on a cubic lattice was very artificial and almost impossible to compare with the morphology of a branching coral . Subsequent work [27] used an accretive growth model in which the branching mechanism of the coral was controlled by local curvatures . In both approaches the advection-diffusion equations were solved using the lattice Boltzmann method combined with the momentum propagation method [28] . Still , a major issue with the latter method was to become unstable for relatively very low flow velocities . In three more recent papers [24] , [29] , [30] , an improved version of the accretive growth model has been used - the polyp-based accretive growth model , in which branching is no longer controlled by a branching rule but it is an emergent phenomenon of the model and is caused by local differences in simulated nutrient concentrations at the surface . With this enhanced accretive growth model combined with the lattice Boltzmann method and the momentum propagation method it was possible to compare simulated morphologies to real coral morphologies . Because of the numerical instabilities , only morphologies in the diffusion-limited range have been simulated so far . In the present work we use the polyp-based accretive growth model to simulate coral morphologies and we are , for the first time , able to increase the flow speed up to 5 cm/s by using the advection-diffusion solver from COMSOL Multiphysiscs [31] . In order to study the emergence of symmetrical and asymmetrical growth forms under different hydrodynamic conditions , we used a previously published computational growth model [29] coupled with a model of nutrients transport through flow [31] . We computed local concentrations of “nutrients” around a simulated object , where the growth rate is driven by the absorbed quantities of simulated nutrients . The improved accretive growth model generated a number of flow-induced growth forms , depending on the model parameters . The simulated growth forms were compared to Computer Tomography ( CT ) scans of real corals exposed to different flow conditions in a controlled flume setup [25] , [32] while changes in symmetry and compactification were evaluated with regard to corresponding changes in Péclet numbers
The experimental data used in this work originated from samples of Pocillopora verrucosa colonies collected after the end of two flume tank experiments conducted at H . Steinitz Marine Biology Laboratory , Eilat , Israel ( Red Sea , 29°30′N , 34°56′E ) [25] , [32] . Both experiments took place in an in situ controlled flow environment . The first two samples ( Figure 1C and 1E ) were scanned using CT scanning techniques [27] with a voxel spacing of 0 . 35 mm × 0 . 35 mm × 0 . 30 mm and 262 to 378 numbers of slices per scan , depending on the colony size . The remaining samples ( Figure 1 A , B , D ) were scanned with a voxel spacing of 0 . 59 mm×0 . 59 mm×0 . 40 mm . Visualization of the CT scans was carried out with Paraview [33] . The simulations in this study are based on the accretive growth model [29] . The growth process is initialized with a triangulated spherical object ( Figure 2A ) . Each vertex of the triangle represents a polyp which is interconnected with other simulated polyps . The growth of the simulated polyps is constructed along the direction of normal vector ni of the vertices vi ( Figure 3 ) . The development of the simulated coral can be viewed as an accretive process of the evolving interface consisted of layers constructed on top of each other ( see Figures 2C and 3 ) . The growth of simulated corals is assumed to be locally limited by the availability of nutrients ( e . g . Dissolved Inorganic Carbon ) from the surrounding water . Simulated nutrients are absorbed at the surface nodes and can be translocated to their neighbor by a surface diffusion process . The simulation domain consists of two compartments: a rectangular channel ( 60 cm length × 60 cm width × 40 cm height ) , where fluid is supposed to enter and exit the domain from left to right . The simulation is initialized with a triangulated spherical object with a diameter of 6 cm representing the ( initial ) simulated coral . The discretization of the simulation domain is done using the Galerkin finite element method in COMSOL Multiphysics [31] Figure 4 demonstrates an example of the finite element mesh from the discretization of the simulation domain . The flow field is given by the incompressible Navier-Stokes ( NVS ) equations and the continuity equation ( 1 ) where F is the external volume force , which was set to 0 , u is the fluid velocity , p is the pressure , is the dynamic viscosity and is the stress tensor that results from the fluid viscosity . Fluid entered the simulation domain with the initial velocity where is the uniform inlet velocity profile along the normal direction of the entry point and was set to 0 . 05 m s−1 . The boundary conditions of the substratum and surface of the simulated corals were set to the no-slip condition where velocity vector was zero . At the outlet the pressure and viscous stress is set to zero . This boundary specifies vanishing viscous stress along with a Dirichlet condition on the pressure . In general , this perscription is numerically stable and admits total control of the pressure level along the entire boundary if the viscous stresses downstream are homogenous [34] . The rest of the boundaries were open boundaries with no viscous stress . This condition describes boundaries that are open to large volumes of fluid , which is free to enter or leave the simulation domain . The impact of the flow is described using the non-dimensional Reynolds number , defined as ( 2 ) in which L is the characteristic length ( m ) , is the mean velocity of the fluid ( m s−1 ) , is the kinematic viscosity ( ) , is the dynamic viscosity of the fluid ( Pa s ) and is fluid density ( kg m−2 ) . Assume that coral branches can be roughly approximated by a circular cylinder . Depending on the Re number , the wake behind a circular cylinder behaves differently from steady larminar flow ( Re<40 ) to unsteady with periodic behavior ( 40<Re<200 ) . The wake becomes unstable for Re>200 and completely turbulent for Re>5000 . For the dynamic viscosity = 1 . 0 e−3 Pa s and u∼0 . 05 m s−1 , for a 2 cm branch ( L∼0 . 02 m ) , we will have Re_branch∼1000 . If the whole colony is assumed to be a circular cylinder , Re_colony will be ∼5000 . From this estimation , the flow through coral branches will become turbulent thus different eddies of different sizes and strengths can be expected along the wake [16] . This gives rise to instabilities in the Navier-Stokes equations . In our simulations turbulence is suppressed by increasing the viscous forces over inertial forces ( = 5 . 0 e−2 Pa s ) . Augmenting fluid viscosity 50 times more than water , the Re_branch and Re_colony are greatly reduced ( Re_branch∼20 , Re_colony∼100 ) resulting in a laminar streamline flow thus allows reasonable time to find the steady-state solutions . For the analysis of stationary time independent solutions , we use a nonliner solver with an iterative method to solve the linearsized equations . The stationary solutions were calculated with the biconjugate gradient stabilized ( BiCGStab ) linear system solver and Vanka preconditioner from COMSOL Multiphysics [31] . Vanka update was done two times in the preconditioner iteration and solved using generalized minimal residual ( GMRES ) method . The relative tolerance of the iterative solver is set to 1 . 0 e−3 . The initial velocity was used only in the first simulation step whereas in the subsequent simulations , the solutions from the previous time step were reused as the new initial conditions . This makes the solution converge faster due to small changes in geometry for each simulation of the growth step . In the supplementary Text S1 , we provide a MATLAB script describing COMSOL routines ( v3 . 5a ) and parameter settings that were used for our advection-diffusion simulations . After the solutions of NVS equations had been found , simulated nutrient entered the simulation domain from all sides and were absorbed by the simulated coral . Subsequently , the amount of absorbed nutrients were determined by solving the advection-diffusion equation , ( 3 ) where c is the concentration gradient , is the velocity vector obtained from the solution of the NVS equations and D is the diffusion coefficient . The concentration of nutrients was initialized with the idealized value of 1 . 0 mol m−3 on all boundaries except the substratum and the simulated object , which were set to 0 . The stationary time independent solutions were computed using the BiCGStab linear System Solver and Vanka preconditioner [31] . After the relative tolerance of the iterative solver had reached a stopping criterion of 1 . 0e−6 , the absorbed flux at each of the vertex was outputted by interpolating the concentration of nutrient at a small distance ( l ) along the normal direction of the vertices . Since the concentration on the coral's surface is zero , we normalize the absorbed flux by the maximum concentration cmax , which results in a flux that is independent of D and l and has a maximum of a unit concentration 1 mol m−3 . These absorbed quantities were translocated to the neighboring vertices by means of surface diffusion , ( 4 ) where is the surface diffusion coefficient . The surface diffusion process can be interpreted as a simplified mean to translocate energy supply , in terms of absorbed concentration in this case , between adjacent regions on the simulated object's surface . Diffusivity on the surface is controlled by , slower leads to minimun translocation and induces a small branching pattern , whereas faster facilitates bulky branches due to scattered resources on the surface ( Filatov et al . [30] ) . Since the branching pattern is sensitive to the variation of surface diffusivity , to address the impact of flow , we kept constant to a value of 3 . 0e−4 m2 s−1 in all of the simulations – this value induces intermediate branching pattern . After calculating surface diffusion , the translocated concentrations of nutrients at each vertex were used to determine the thickness of a new growth layer . For a vertex vi , the length of the newly extended growth vector li along the normal direction of vi is regulated by the growth function: ( 5 ) where g ( Ci ) is a nutrient response function that obey saturation kinetics , which converts the translocated nutrients Ci at vi into the local thickness li of the new growth layer . In order to relate the growth rate to the translocated nutrient , it is assumed that the growth rate approaches a saturation state at high nutrient uptake thus limiting the extension of the local thickness . This type of growth function is a sigmoid function: ( 6 ) where Lmax is an asymptotic maximum growth rate , is the translocated nutrient at a vertex vi . The exponent n defines the kinetic order of the growth rate with respect to Ci . The model parameters and K denote the characteristic growth curve of the growth function . For every growth simulation , Lmax was set to 1 . 0 e−3 m and K was set to 1 mol m−3 . This setup corresponds to an extension rate approximately about half a millimeter because is normalized to 1 mol m−3 . Since the growth rate of a coral colony is approximately about 1–2 cm per year , the size of the simulated object after a consecutive run of 150 growth simulations ( ∼10 cm ) is comparable to the size of the speciements collected fron the in situ experiments ( ∼10 cm ) . The last parameter exponent n was increased to n = 1 . 2 to yield a slightly steeper growth curve . In the appendix we demonstrate the influence of the parameter n on the simulated morphologies ( Figure S2 ) . To compare the influence of hydrodynamics to the morphology obtained from the simulations and CT-scanned corals , we used the Péclet number ( Pe ) . In the absence of fluid motion diffusion dominates transport and results in a lower Pe , while large values of Pe denote a regime where advection dominates . ( 7 ) where is the mean flow velocity and L is the characteristic length , and D is the diffusion coefficient . For each simulation , the impact of flow over diffusion was simulated by gradually decreasing the diffusion coefficient D in Equation 3 by a factor of 10 from 1 . 0 m2 s−1 to 1 . 0e−5 m2 s−1 . The variation allowed us to study the resulting morphology by keeping constant . The characteristic length L used here was the average diameter of the terminal branches of the simulated coral ( dc ) . This parameter can be considered as a morphological invariant property of a branching object [35] . The list of the parameters we used in the simulations can be found in Table 1 . While for simulated objects exact Pe numbers were calculated , the value of Pe for the real corals could only be approximated . Here we used the characteristic velocity acquired from the flume experiment [32] . The characteristic velocity in which the “reduced-flow coral” ( Figure 1B ) and the “enhanced-flow coral” ( Figure 1C–E ) were grown was assumed to be 1 . 0e−2 cm s−1 and 15 cm s−1 respectively . The diffusion coefficient was assumed to be 1 . 0e−3 m2 s −1 and invariant for the whole estimation of Pe number . The calculation of characteristic length L was the same as the length scale calculated from the simulation , which was the average diameter of the terminal branches ( dc in Table 1 ) . Due to the complexity of coral's geometry , the morphological analysis required an alternative approach from traditional landmark-based morphometrics [36] , which is more suitable for unitary organisms . In our work , we used a morphometric method that extracted a number of localized variables such as terminal branch thickness ( dc ) and local direction of growth of the branching tips [35] . These morphometric traits are further used to quantify the morphological resemblance between CT scans of corals from field experiments ( Fig . 1 ) [25] , [32] and simulated corals . An important pre-processing step of the morphometric analysis is the construction of a morphological skeleton of a 3D object , shown in figure 5A . Skeletonization reduces the object to a network of thin lines , one pixel or voxel thick , running through the centers of the object . A morphological skeleton or medial axis has the same topology ( a similar branching structure ) as the original object , and occupies the same spatial extent in the image . We used the skeletonization algorithm previously described [35] . The diameter dc of the sphere c located at the endpoints of the morphological skeleton was determined and represents the local thickness of the terminal branches ( see Figure 5B–C ) . In addition , global geometric properties from simulated and CT-scans corals were extracted: surface , volume and surface/volume ratio estimations ( Table 1 ) . To quantify the symmetry of branch formation in corals and simulated objects , we introduce two extra morphometric variables - the symmetry angles hangle and vangle ( Figure 5E ) , and the symmetry magnitude smmag ( Figure 5F ) . Let us consider a morphological skeleton graph involves N distinct vertices , where are vertices connected to root node in three-dimensional Euclidean space ( ) . We define a horizontal plane Pxy passing through and the orthogonal vertical plane Pyz . The projection of N-1 vertices on Pxy and Pyz consists of 2 ( N−1 ) projected points and respectively . For each vertex , the angles between and are defined as hangle and vangle ( and in Figure 5E ) . The two angles give us the orientation of vertex to Pxy and Pyz respectively . Here we define the symmetry angle of the skeleton graph by looking at the distribution of hangle and vangle . The morphological skeleton graph is considered to be symmetric ( in the horizontal and vertical direction ) if the difference between hangle and vangle is close to zero i . e . the two angles are the same ( ) . Furthermore , consider a reference point s chosen in the upstream direction of a coral that was exposed to uni-directional flow or s as an arbitrary reference point for corals that were exposed to other conditions , is a projection point of s on Pxy . Let be an arbitrary vertex in a morphological skeleton graph , the scalar projection of vector to as a magnitude of , where is the angle between and . Let us imagine another vertex in the skeleton graph that lies exactly in the opposite direction of , the sum of the two projected vectors ( to and to ) negates each other and thus the vertices and are considered to be symmetric in the direction of the vector i . e . along streamline direction . The symmetry magnitude smmag is defined as , where denotes the angle between and . The skeleton graph is considered to be symmetric ( along stream line vector ) if . The distribution of the symmetry angles hangle and vangle gives an overview of branch orientation with respect to flow direction . A skewed distribution of either hangle or vangle indicates asymmetrical branching distribution in some direction whereas the sum of smmag provides the symmetry of branches along streamline direction .
Under diffusion-limited conditions ( no-flow ) in the simulations , branches emerge in all directions leading to a relatively symmetrical shape ( Figure 6A ) . In the subsequent simulations we increased the impact of flow over diffusion by gradually lowering the diffusion coefficient ( D in Equation 3 ) . We observe that the branches asymmetry increase as the Pe number becomes larger . Branches tend to form in the stream upward direction , while they are gradually suppressed in the downstream sides is ( Figure 6D–F ) . Similarly to the the coral from the flume tank experiment ( Figure 7 ) , an asymmetrical simulated growth form emerges , particularly at a higher Pe value , with branches mainly formed on the upstream side ( Pe_branch = 1 . 13 , Figure 6D and 7 ) . Considering the flow patterns around the simulated objects , we observe an asymmetrical branching trend with a high degree of compactification with the increasing Pe ( Figure 8 ) . Flow inside the colony , i . e . between branches , was substantially reduced with most of the flow circumventing the coral's colony ( Figure 8D–E ) . Furthermore it can be observed that the along-stream flow gradient of the simulated object becomes steeper for higher Pe numbers leading to a higher degree of absorption of simulated nutrient in the upstream part of the simulated object . The reason for this is because the simulated morphology becomes more compact at a higher Pe and hence friction is greater , not because the flow patterns vary with the Pe . In other word , the simulated morphology has changed with the increasing Pe . To quantify the degree of compactification of the corals and the simulated forms under the influence of different flow conditions , we computed the surface/volume ratio of each form . In general , the surface/volume ratio of Pocillopora verrucosa is higher than other species such as Madracis mirabilis , Acropora sp . and Montipora sp . [35] , [37] . A higher surface/volume ratio indicates a higher area which is in contact with the environment and ( potentially ) a higher amount of mass transfer . The roughness of the coral surface also influences mass transfer rate [16] . Factors that influence the degree of surface roughness include the living tissue and polyps . In reality , it can be observed that surface area will increase when corals extend their polyps thus affecting the mass transfer rate . However , in our study we assume a smooth surface because including all the details about the roughness of a real coral would require 3D images with extremely high resolution , which is currently beyond the capabilities of the available medical scanners . The surface/volume ratios calculated from the in situ corals have a higher value when compared to the surface/volume ratios obtained from the simulations . The controlled coral ( CT_456 ) has the highest surface/volume ratio whereas for the rest of the corals , we found an increasing trend when Pe_branch and Re_branch increased . However increasing Pe number in the simulations tends to causes the surface/volume ratio to decline ( Table 1 ) . The degree of symmetry was analyzed by examining the distribution of symmetry angles ( hangle and vangle ) and symmetry magnitude ( smmag ) . The mean values ( hangle and vangle ) provide information about the direction of coral's branches while the sum of symmetry magnitude gives the notion of the colony-level symmetry . We first looked at the trend of these two morphometric traits from the simulations and then compared with the CT-scanned corals . For the flow simulations with lower Pe , the average values of symmetry angles are and the sum of the symmetry magnitudes is quite low ( Figure 9A and 9B ) thus indicating the morphology is symmetric . Under high flow , the symmetry angles diverge from and since they are symmetric to each other the sum of the two angles is always close to ( Figure 9C–9E ) . We verify the same trend in the CT-scanned corals , for which the sums of the symmetry magnitudes are small for a highly symmetric growth form , but there is a distinction in symmetry angles between the controlled coral and the in situ corals . Although the controlled coral ( CT_456 ) exhibits a low sum of symmetry magnitude smmag and the symmetry angles are close to ( both hangle and vangle ) , this does not hold true for the in situ corals ( TS_001 and TS_002 , see Figure 10A and 10B ) since their symmetry angles diverges from even though they still appear to be symmetric ( because of the low smmag ) . This suggests the relationship between the Pe number and the associated symmetry angles and magnitudes; that is increasing Pe tends to make the sum of the symmetry magnitude higher and induces the symmetry angles to diverge from ( Figure 10C–10E ) . The simulation approach provides an indication of how the simulated corals can change their degree of symmetry by the increasing Pe . Under the influence of uni-directional flow , branches form towards the direction of the incoming flow thus reducing the formation of branches on the downstream side . This phenomenon can also be observed in the in situ CT-scanned corals .
Using a simulation approach , we studied the impact of hydrodynamics on the growth of the scleractinian coral Pocillopora verrucosa . The 3D morphometric analysis provided a quantitative approach and enabled us to investigate the impact of uni-directional flow . We observed that hydrodynamics plays a major role in the simulated morphologies whereas an increase in Pe number induces the formation of asymmetrical branching growth forms ( Figure 6 ) . Branches tend to develop in the direction of the incoming current , resulting in an asymmetrical form . In the flow simulations , we found a decreasing trend of the horizontal symmetry angle ( hangle ) when increasing Pe , as opposed to the vertical symmetry angle ( vangle , see Figure 11A ) and the symmetry magnitude ( smmag , see Figure 12A ) . A similar trend was found in the CT-scanned corals from the in situ experiment ( Figure 11B and 12B ) . According to our analysis , corals seem to maintain their symmetry angles about 45° to the substratum at lower Pe . In the simulation these angles will change drastically if high Pe number is used . While the simulated corals exhibit a high degree of plasticity , in situ corals tend to maintain their symmetry to some extent . For low Pe number their symmetry angles were less than 45° but they were observed to be very symmetric , which implies plasticity due to the impact of uni-directional flow . Our simulation model predicts a decreasing surface/volume ratio when Pe increases ( Figure 12C ) which also occurs in many marine sessile organisms ( e . g . sponges , hydro corals , scleractinian corals [13] ) . In contrast , this trend does not agree with the in situ CT-scanned corals ( Figure 12D ) , since they were grown in the controlled experiments in situ , where flow ( uni-directional ) was limited . Hence it is impractical to compare them with the earlier measurements , which used samples from different hydrodynamic regimes growing at a different temporal scale . The surface/volume ratio provides a significant implication of how corals occupy a certain volume without taking into consideration the temporal scale of their growth . After a period of simulation time , objects from advection-dominated simulations occupy less volume and become more compact , reducing their surface/volume ratio . However , if spatial scale of the growth is used to evaluate the surface/volume ratio , at any rate , flow-induced object will exhibit a higher ratio . For example , considering the interim object ( growth step 98 ) of the diffusion-dominated simulation ( Figure 6A ) , the object occupies a volume of 7 . 71e−5 m3 and has a surface area of 2 . 23e−2 m2 corresponding to the surface/volume ratio of 289 m−1 . This value is definitely lower than the ratio measured from the advection-dominated simulation ( Figure 6D ) that has the surface/volume ratio of 416 m−1 ( volume = 7 . 74e−5 m3 and surface area = 3 . 22e−2 m2 ) . Although our simulations provide a reasonable approximation of a coral growth process and various growth forms emerge in response to the varying Pe number , we still face a challenge to include a full scale level of detail . Our simulation cannot approximate all the fine details of Pocillopora verrucosa especially the small-scale roughness of the branches with many bumps ( hence the Latin time verrucosa ) but we can simulate other coral species with a relatively simple corallite structure quite well ( e . g . Madracis see [30] ) . If the exact approximation is needed , we may have to include the role of corallite for species with complex corallite structure - such as Pocillopora or Acropora - in our model . Another interesting property of branching corals that we cannot simulate is known as anastomosis –the fusion of branches , a phenomenon which is observed in many branching organisms such as scleractinian corals , hydro-corals and sponges [13] that are exposed to high energy regimes of the reef . If we look at the simulated form in Figure 6 , the object with highest Pe ( Figure 6F , Pe_branch∼11 . 3 ) is packed with a lot of braches . In reality some of these braches can easily be fused . This property is probably important for the study of coral growth in a higher range of Re number . In our model , we also address the relevant importance of the growth function ( Equation 6 ) to the growth rate . The model parameters from Equation 6 can be interpreted as species-specific variables . The implication of this is the possibility to use real data and attribute them to the simulation model . Although the underlying physiological process is highly simplified , the diffusion-limited assumption still plays an important role to the study of coral's growth [38] . Furthermore , based on the surface diffusion and diffusion-limited assumption , our accretive growth simulation is a good candidate to study the so-called “two compartment proton flux model” [39] , where a simplified version of calcification can be incorporated into the accretive function to study the translocation of fixed-carbon energy supply from zone of photosynthesis and zone of rapid calcification . Hence , to assess the effectiveness of relevant parameters such as Pe number , Re number and surface diffusion , to name a few , an extensive sensitivity analysis is required together with a quantitative comparison between the simulated objects and the real corals ( see also [30] ) . The methods presented in this paper for modeling accretive growth and the impact of hydrodynamics , in combination with a method for the quantitative analysis of three-dimensional complex shape can be applied to a large class of marine sessile organisms ( e . g . scleractinian corals , hydro corals , sponges , rhodoliths ) . Morphological plasticity is a major issue in different fields of marine and coral biology ( e . g . ecology , taxonomy , paleontology ) with applications in environmental studies ( e . g . coral bleaching and ocean acidification [38] ) . To our knowledge , only few of the existing computational models aim to understand the emergence of growth and form under different hydrodynamic conditions . Most examples are from physics [40] , [41] , and a few studies about hydrodynamics effect on growth and form of bacteria colonies [42] or bio-films [43] . Thus far , we have overcome the instabilities issue of the earlier version of the accretive growth model [44] , [45] and we are able to simulate coral morphologies at a more realistic flow speed ( 5 cm s−1 ) . We varied Pe number by lowering diffusion coefficient ( D in Equation 3 ) and maintained a constant Re number throughout the simulations ( Re_colony∼100 ) . While this is crucial to avoid numerical instabilities of the Navier-Stokes equations , our results ( the simulated forms ) could be slightly different if the Pe number was changed by varying the velocity rather than the diffusion coefficient . This issue could be overcome by introducing turbulence flow simulation and study the steady state approximation of Reynolds's Average Navier-Stokes equations . With our provided framework , this can be done in the future using CFD package with turbulence flow solver in COMSOL Multiphysics [31] . To date , our coupled accretive growth model is the first example of a computational model of growth form that can be used to generate objects with a high resemblance to biological growth forms under different hydrodynamic conditions . We can compare and quantify our simulated objects and the real corals using three dimensional morphometrics . Our study also shows that the formation of symmetrical branching forms [25]–[27] can be explained with a biomechanical model . In reality most scleractinian corals will not be growing under uni-directional flow but will be exposed to bi-directional current where the flow direction is reversing twice a day because of the tidal movements . This leads to a next question: is a bi-directional current causing the formation of symmetrical branching colonies ? This will be a subject of our future research . | A long-standing question in marine biology and coral biology is the morphological plasticity of corals , sponges and other marine sessile organisms and the influence of water movement . Usually branching species tend to develop symmetrical colonies where branches are being formed in all directions . There is a long standing discussion if this process in which colonies develop symmetrical colonies is controlled by genes or by the environment . In this work , we address this question for the scleractinian coral Pocillopora verrucosa . We first have acquired coral colonies from a controlled in-situ flow experiment where the coral was growing under uni-directional flow conditions . The corals colonies were scanned using a Computed Tomography ( CT ) technique used for medical imaging and industrial imaging . We have developed a simulation for the growth and form of corals and the influence of water movement . We have compared the simulated morphologies to the three dimensional images obtained with the CT scanner . We have found that coral's branches predominantly develop in the upstream part of the colony and an asymmetrical colony is being formed under uni-directional flow conditions . Our results confirm that growth of the coral is strongly influenced by the flow conditions . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"biophysic",
"al",
"simulations",
"marine",
"and",
"aquatic",
"sciences",
"biology",
"computational",
"biology",
"marine",
"biology"
] | 2013 | Modelling Growth and Form of the Scleractinian Coral Pocillopora verrucosa and the Influence of Hydrodynamics |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.